Choosing Capacitors to Recap Old Electronics

[@borayurt66]

Tons of info here, very useful. But, why not talk about other critical properties of capacitors like ESR? Also, why not talk about properly testing a capacitor to see whether it really needs replacing? Not all caps puke electrolyte when they go bad, and not all of them lose their capacitance when they go bad. Actually, it is very common to see capacitors go up in capacitance when they go bad. This is of course not an actual increase in capacitance but a metering error (if a less than ideal tool like a DMM is used) due to electrical leakage, which is another common fault in aged capacitors. It is very easy for a novice person to measure a 100uF cap and see that it is reading 110uF and decide that it is still good.

[@JDW-]

Thank you for watching and for making time to type a thoughtful comment. Yes, you are correct about my not having talked in depth about ESR in this video, and it is a good topic to learn about. My reason for not having taken the deep dive into ESR in this video probably won't satisfy you, but here it is... No fluid-filled electrolytic capacitor has eternal life. No manufacturer datasheet suggest eternal life for caps either. And while I did present the excellent DE-5000 LCR meter in my video, which ranks among the top hand-held LCR meters you can buy in terms of features and accuracy, I have tested very old, bad caps that measured good ESR and even good capacitance across the full range of frequencies, 100Hz to 100kHz. But after swapping them out, the device began to work properly. And you are correct about some old caps showing higher capacitance than brand new caps too. So measuring ESR and capacitance alone really isn't good guidance for knowing when to swap out possible bad caps. That's why I prefer to replace capacitors in old in accordance with their age, rather than ESR or other aspects. The number of years it has been soldered into that PCB matters. If the fluid-filled cap has been on the board for over 30 years, it pretty much goes without saying it needs to be replaced. That's not true of SOLID electrolyte capacitors (ceramics, film caps, etc.), and I talked a little about that in my video. I disagree with folks who tell me about caps in their experience that are 40 years old and still going strong (mostly HAM radio operators, or those who use vintage radios), because even though those examples are real and exist, most capacitors that are 25 years old and older don't fair so well, and certainly not the smaller SMD electrolytic caps, such as are used in the Apple Macintosh SE/30, Classic, etc. I watched a KZbin video only an hour ago by a gentleman who purchased a SEALED-IN-BOX Macintosh Classic (from 1990) that he proceeded to open on-camera, and he took it apart to examine the circuit boards and didn't see anything visually wrong, but when reassembled and tried to power it on, there was no bong or screen display. He then found it had blown a fuse. One swipe around the SMD caps with a swab showed yellow goo that had come out. Again, this was a brand new (old stock) machine, where the seal on the box had never been opened, and the machine had never been used (outside factory testing). So again, AGE really is the right determiner of whether an old fluid-filled electrolytic cap should be changed, and it is also an easy to understand rule for novices who may not have a good LCR meter. That rule about age remains appropriate even for caps that have never been used at all. If it's 25 years old or older, it needs to be replaced. If you're talking 15-25 years, it might still work, and perhaps ESR testing could help. But to do appropriate ESR testing, you need to desolder and remove the cap in most cases to do a proper test. Most people don't go that far, which again, is why using AGE as the determiner of whether a fluid-filled cap should be replaced is rooted in sound reason. When I recap old computers, I not only recap the power supply and motherboard, I even recap the floppy drives too, even if they work fine. Just remember that no fluid-filled cap has eternal life. I hope this helps!

[@borayurt66]

@@JDW- Thank you for this amazing and detailed answer. I totally agree with you on the 25 years rule. I replace them all regardless of how they test. A 25 years old cap may be fine today, but certainly there won't be much healthy life left in it, so better to replace it now before it starts leaking current (which is unavoidable) and damaging hard to get much expensive components. I stopped arguing and explaining things to anti-recappers many years ago. I take them as a branch of the flat earthers, and let them be in their fantasy world. I am in fact a ham radio operator and I can easily say there are as much anti-recappers in the retro world as in the ham radio world 😂. Electrolytic caps age even unused, sitting in a drawer (much like the example you gave with that sealed in box Mac) This is exactly why one should know how to properly test caps, not for deciding whether to replace them or not, but to test the caps that are bought "new". This is more significant with axial caps. You can never know how long they have been sitting in a drawer in the vendor's shop before you bought them. Not many manufacturers still produce "fresh" axials, most of them have an extensive amount of NOS parts. Better to know how to test a cap properly rather than blindly believing them to be good just because you bought them as "new" This is why I think explaining concepts as ESR, Vloss etc is crucial in a video on recapping. Most people still think a DMM with a capacitance range is enough to test capacitors, the truth can't be further away. Another trick for how to choose replacement caps is to choose higher voltage and temperature rated ones than the ones you take out. A 35V cap will last longer than a 16V one, even though it only sees 5V. Same with temperature, a 105°C cap will last longer than a 85°C one even there is no such temperature in where they will be working. Modern caps being much smaller than the older ones, it is very easy to fit ones with higher ratings to the same spots. Thanks again for the time you took to answer me, take care!

[@JDW-]

Thank you for providing such useful feedback. Your comments about testing newly purchased axial lead capacitors are spot on! Absolutely! Everyone needs to do that. I also do exactly as you do and purchase replacement capacitors with a higher voltage specification than the stock capacitors. I usually need to do that in order to ensure the capacitor body is similar in size to the stock capacitor, which has advantages not only in terms of the lead spacing but also the durability of the capacitor overtime. I tend to trust the physically larger body aluminum electrolytic capacitors over the smaller ones. This is an important consideration because capacitors today are smaller than the old capacitors were even though both have the same capacitance and voltage specifications. So by going to a higher voltage, you get a physically larger capacitor, and often a better one too. And of course, it usually results in lower ESR, which is good in most situations. However, you have to be careful and recapping switching power supplies because some old supplies require a minimum ESR for stability. Thanks again for sharing your excellent thoughts! I think our dialogue is important for everyone to read, so I am making this a pinned comment that will perpetually stay at the top of all other comments. Thanks again!

[@ESEJESEJ]

Hey guys, you two seem to really know what you are talking about. I have a problem, or rather a whole bunch of problems now that I know more about caps and their lifespan. Anyway, long story short: Got hold of a B&O Beovision mx5000 crt tv recently, made in 1988. *Smelled really bad when being on for 10+ minutes. *Turned out to have lots of dust inside. Cleaned it and smell went away. *Another smell emerges and upon deeper inspection, both of the internal batteries had gone bad, leaked and even damaged the board to some degree. *Replaced both batteries and cleaned the leakage away = smell finally gone. * Using it for three days and the smell is now back! It smells like a toxic electronic smell that fills the entire room. Can't stand it so have to turn it off. So to my question: is there a condition within electronics, where due to some electrical failure, even brand new batteries will get "eaten"? A leaking battery smells really bad. Would you say the same applies to leaking caps? For pictures of my tv, if at all helpful, search for the following on reddit, "Images of my B&O MX5000" Cheers.

[@santospoland]

The significant notes are a signature of the amazing attention to detail and considerations you make to your craft. Thank you for sharing all of this with us. Your work is a treasure!

[@JDW-]

Likewise, I treasure your extremely kind words, Alex. Thank you for having been the inspiration for this video!

[@johnmitchell2741]

Words cannot describe my appreciation for you taking the time to make this video and explaining it in layman's terms for a novice like myself to understand 👍Tons of valuable information that I can come back to as needed

[@JDW-]

Thank you for both watching and making the time to let me know the video was helpful to you, John. Best wishes!

[@threethousandyen]

Great overview! Digi-Key does let you sort by price at quantity. Really great to find discounts if you plan to recap multiple machines.

[@retrocomputinggrotto]

Very well explained! Anyone just starting to re-cap old(er) electronics should watch this :)

[@sketchpalosotherchannel]

Just recapped my IIsi today, it's working! all I need is a monitor and peripherals and I will finally have a really nice vintage mac. I unfortunately had not watched this video at that time, but here I am now. I did buy the logic board caps from amazon, but the power supply board caps are from Mouser.

[@JDW-]

Glad your IIsi is now working! I can't speak about caps from Amazon, but I can say that when you buy from vendors like DigiKey or Mouser or RS Components, you will get quality parts.Sometimes when you buy from Amazon, you don't know what you're getting until it arrives, and that remains true even if the Amazon page advertises a big name cap brand like Nichicon. Also, if you need cheap caps quickly, yet from a very reputable vendor, you should check out Console5. I've not purchased from them only because I always get parts from Mouser, but their kits are top notch and sell for great prices.

[@sketchpalosotherchannel]

@@JDW- Yeah, I kinda, just put in the uF and volts and went from there, I got lucky on amazon because the size was good enough and I did have to replace some axials with radials. I used a pic guide from @BranchusCreations IIsi video. Wish I had watched your video sooner lol. Oh well, lessons learned. kzbin.info/www/bejne/j3_WgaVnoq2IqNk

[@cberge8]

If you aren't able to determine the original voltage raring of a cap. Look for other caps on the board that are in parallel to the one being replaced. The voltage rating on those would be the same as the one in question.

[@JDW-]

That sometimes works, but not always. It's still a guessing game. For example, the SONY PSU inside an Apple HD20SC external hard disk drive enclosure uses two 2200uF caps side by side, but one is rated for 16V while the other 10V. In that case I can see the voltage ratings, but let's assume for the moment I could only see the voltage rating on one and not the other. Well, I thought it okay to buy two 16V replacements because they would physically fit, but I later found the output became unstable due to the fact the higher voltage rated cap had different ESR specs which impacted the require minimum ESR of that particular power supply. You can see that in my video below and in the spreadsheet I made which shows replacement cap specs side-by-side. So if you look at a given PCB and pick a replacement cap based on the voltage rating of nearby caps, it may or may not work as expected, at least, not in a switch-mode power supply. • VIDEO: kzbin.info/www/bejne/h6DJXoqgn8Sljas • SPREADSHEET: bit.ly/3yFsoOT

[@fritzj6803]

Great tutorial. You can also measure the resistance or microhenries of the inductors with the MESR-100 ESR meter. The reactance formula is 2𝛑FL. That would be 2 x 3.14 x frequency of esr meter which is 100 kHz x 100 (uH) Just plug it in to get the reading. For resistance R=0.628 X L(uH) For Inductance L(uH)=R(ohms) ➗ 0.628

[@gyrgrls]

Every great tutorial deserves a recap.

[@buriedbits6027]

This is a video I always come back to. Thank you so much for putting this video together.

[@_P_M_]

Good information! I enjoy vintage synthesizers with aging power supplies. I've learned enough to source and replace old LCD's and batteries but I haven't had the nerve to do any recapping. I hope to learn more so I can keep them running.

[@micinput]

There's one useful source of information you didn't specifically mention. Often you can track down "service manuals" for your equipment. This is usually the source for the schematics, but "service manual" is usually a better Google search term. Even if the service manuals only have block diagrams, they sometimes list the "bill of materials". For example, I have a music studio effects unit from Lexicon, and its service manual lists all the capacitors used, their specifications, how many of them there are, and what they are labeled on the silkscreen on the motherboard, i.e.: "CAP,ELEC,22UF,25V,RAD 4 C27,28,31, 104" Having watched your video I can now interpret that as "Capacitor, electrolytic, 22 microfarads, 25 volts, radial. There are 4 of them, specifically the ones labeled 'C27', 'C28', 'C31' and 'C104' in white on the motherboard." I'll still need to measure them with calipers, but I'm a lot farther into identifying replacements than I was before.

[@JDW-]

Thank you for making time to let me know my video was helpful to you, and for sharing your personal experience. The primary reason I did not mention “service manuals,“ is because I myself recap vintage Macintosh computers, especially from the 1980s. For those machines, there are no service manuals with schematics or BOM‘s. Larry Pina books offer a complete capacitor list in the appendix for certain Mac models, but for other Mac schematics you need to search online, especially vintage Macintosh discussion forums or the Macintosh Garden. For applications like yours though, you are correct, there are a good number of service manuals out there which provide extremely detailed and helpful information. Thank you for bringing that to our attention. I wish you the best in your recapping adventure!

[@AladimBR]

Very helpful. Thanks for sharing your knowledge. I'm a mechanical engineer and electronics are not my expertise. Despite that, I've always been interested in computers and have build my own machines since 386/486 days. At high school, I had an Apple II+ plugged into a 26 inch Philco TV (valve tech), which I had to "dispute" with family. A lot of nice memories. Anyway, I have some retro stuff at home: kept P2, P3 and some Athlon XP mainboards, plus 3dfx video cards, and an Apple II plus, IIe and IIGS I've collected over the years. Most of these boards (and some PSUs) need a recap job. I've trained over the last year on junk electronics (meant to scrap) in order to learn how to weld and now I feel confident I can do it. I decided to learn this skill because many things break at our homes due to failing/ageing capacitors and most business just want to replace parts, not fix. A total waste and very hard to keep our loved retro stuff working. That's how I've arrived at you video. Watched 2 times already. Made a list of the capacitors I need and have browsed over mouser.com. One thing I didn't get, it is how to choose capacitors for these mainboards concerning ripple current and ESR: what should I pick? If I understood correctly, ESR is ok if not too high, but it doesn't have to be the lowest also. What about ripple current: low or high? That was not clear to me. Hope you or someone reading the comments can clarify that.

[@JDW-]

Typically, a low ESR capacitor will also allow for a much higher ripple current. I hesitate to give you a general rule for choosing ESR and Ripple only because it really depends on what you are recapping. Power Supplies are most sensitive to ESR and Ripple. If you are not recapping a power supply, but instead doing a motherboard, in many cases the capacitors are just what they call "bypass capacitors" which prevent voltage drops from occurring. In that case, you can most often use the lowest ESR caps possible and will be fine. In fact, I recently released a video on Overclocking a Macintosh LC575 motherboard (aka Color Classic MYSTIC), and I used solid polymer OS-CON capacitors, which have only about 0.025 Ohm ESR, and discovered that kind of low ESR capacitor is actually required to overclocking to the highest speeds. Two associates tried overclocking with solid tantalum capacitors (0.7 ohm) and found they could not overclock as high! So low ESR capacitors definitely have their place. Often times, very low ESR caps can be used in power supplies too, but some power supplies require a minimum ESR, and it's very hard to know that without a schematic, yet most of the time we don't have a schematic! I came across that problem when I recapped the Apple HD20SC external hard disk enclosure. I did a video on that as well. The power supply had a brief whining sound with a very low ESR replacement capacitor, but when I swapped out with a slightly higher ESR cap, the noise went away. (Basically the switching frequency became unstable at power on due to the low ESR cap.) All said, the best general advice I can give is to go with low ESR caps where possible, as those will work most of the time, but you need to be prepared to try different caps on some power supplies. If you choose a polymer aluminum cap, I recommend the "solid" kind rather than "hybrid" because the hybrid types have a liquid inside that theoretically could leak out many year from now. Solid type caps cannot ever leak.

[@AladimBR]

@@JDW- Great! That's what I needed... some pointers... I got your point, don't worry I understood your rational... and have now a starting reference. Thanks a lot / kind regards

[@JDW-]

Wishing a very Merry Christmas to all my fellow recappers! ✝️🌟❄️☃️🎄 🎁 🎅🏼

[@allanknox8216]

Any cap in a power supply should also probably be checked for ripple current as well. Mouser and DigiKey will help you find almost anything if you use their customer service sales guys.

[@santospoland]

Hello James, thank you for putting this together! The video was both entertaining and most educational. Unlike yourself, I don't have an electrical engineering background but my collection of old Macs does require that I know what I am doing to the best of my ability and your expert advise on all things capacitors has given me much need confidence and a heck of a lot of information that I was able to easily grasp. I would also like to thank you for the practical examples based on the SE/30. As you know, collecting vintage equipment does require a number of skills, key sets of knowledge and repair/diagnostic equipment to effectively maintain the health and wellbeing of those devices. Again, your expertise has elevated my confidence in effectively choosing the correct capacitor for the application. I sincerely appreciate the time and care you put into making this video and I am definitely bookmarking this video for future reference. When I am better able to, I will drop something in the PayPal tip jar as a token of my appreciation!

[@JDW-]

Thank you, Alex!

[@davidgreen4406]

Thank you for making this video! I found the information within it to be very useful.

[@xaviergonzalez5828]

Thank you Sir! My native language isn't English but I get you at 100% great video that no one talk about it!

[@JDW-]

Thank you for your kind words, Xavier!

[@TradeWorks_Construction]

Great use of Mousers site to illustrate the thought process used toward picking out a replacement option. I prefer Mousers site myself and thought i would throw my 2 cents in about one of the features I love. This may have already been mentioned by someone else but Mousers site actually will show how many fit your criteria BEFORE you hit apply. At the Far left of the screen under the Products Section in Blue it keeps track of the shown RESULTS # -> To the RIGHT in RED it keeps track of the current results remaining according to your filters. To actually view those results you still need to click apply filters however it’s a very helpful tool to know ahead of time which filters you’re choosing might be excessively limiting your options. ESPECIALLY if any of the filters were preference and not strictly necessary for the operating criteria.

[@JDW-]

Yes, it is convenient. Thank you for pointing out that feature. I show it at about 25:38 in the video, but I do not discuss it. You can see "Results" at the far left (a large number) and then I click a Voltage Rating filter (but don't Apply it), and "Results Remaining" appears in red, showing you what to expect when you do actually click Apply.

[@sierrawhiskey5155]

Many thanks, I'm a middle aged guy curious about recommissioning the 70's Hi-fi gear I grew up with and recently inherited from my fathers estate. It's a pity more people don't realise much of this vintage stuff can be salvaged.

[@JDW-]

Thank you for watching and for your reply. With high fidelity gear, it’s important to keep in mind that you may have problems other than just the capacitors. But replacing leaky capacitors is important. I hope my video is helpful to you in that endeavor. Best wishes!

[@ashleyjwilliamshand]

Thank you for taking the time and effort to make this in depth video. This gives the Layman (like me!) a complete masterclass in this niche topic. As the number of boards from this era dwindles, I can see this video being discovered for a very long time. You're helping save things of little value now, that will be of great value (and increasing rarity), long down the line.

[@JDW-]

Well, I certainly appreciate your kind words, Ashley. I actually do try to make videos that will remain significant in years to come. A lot of the tech channels here on KZbin will talk about recent announcements like a new MacBook Pro, and while that content is very useful at the time those machines come out, those videos become irrelevant after the next edition computer comes out. So that’s why I don’t make many videos like that. Instructional videos on how to choose capacitors though is a different matter altogether. The information will remain relevant as long as current capacitor technology remains relevant. And I don’t see that changing over the next 10 years. Thank you for watching and have a great day!

[@Yreq]

That is video i was looking for. A lot of useful information. I'm using Wurth tools allmost everyday. Prices are astronomical, but quality is (mostly) great, but I've never heard about Wurth caps:D

[@brycelynch2138]

Console5 is my go-to for getting recap kits. Not getting paid to say it, they're just awesome.

[@JDW-]

Even if you were getting paid to say that, I wouldn’t mind. I’m rather libertarian when it comes to what people post here in the comments. With that said, I myself have never purchased from them because whenever I need to place a capacitor order, I always need some capacitors which console five doesn’t carry, and that leads me back to Mauser again. If they had all of the capacitors I needed at one particular time, I suspect it would be cheaper to buy from Console five. They really do have some great pricing.

[@happyday2678]

U finally did this explaining video for this confusing topic. Well done!

[@ioutoubeu]

Thanks a lot for sharing this! I thought interesting to add that in hifi environment another source for identification of the needed capacitor can be read directly on the board itself once the old cap has been removed. With a caveat.. there can be errors about the "direction" of the capacitors (+/-) Thanks again!

[@JDW-]

Thanks. When you talk about polarity "errors" though, are you saying the silkscsreen markings on some PCBs are wrong?

[@richardsinger01]

@@JDW- yes, it is not uncommon. Best to photograph the board before removing capacitors rather than relying in the silk screen printing.

[@JDW-]

@@richardsinger01 In my case there is no need for photos since I have everything shot on 4K video. 🙂

[@Dawstering]

Thank you for another great video! Very informative :)

[@itswilliamevans]

Thanks for the video! Learnt a lot, especially about the derating. I'm new to this and looking at replacing the capacitors on a Macintosh Classic. I have an iMac G3 and G4 that I feel I should do at the same time, seeing as they're about 20 years old now. And thanks for suggesting Mouser, apparently they will ship to me internationally for free if I spend over a certain amount.

[@burtharris6343]

I generally use Digikey first. There, I can easily sort by price at any quantity you want. I'd be surprised other distributors don't have something like that; finding it may be the trick.

[@cali4tune]

This is a great channel! Excellent videos and I appreciate the parts links. Hope the YT algo will pick it up soon.

[@JDW-]

Thank you for your kind words. KZbin doesn’t like me because I keep Ads disabled. No doubt my channel would grow faster with Ads, but I can’t help hating Ads and wanting a more clean and friendly experience for my viewers.

[@cali4tune]

@@JDW- We won't fault you for wanting to grow the channel. Better a few ads than no content in the long run IMO. Thanks for doing great content!

[@JDW-]

@@cali4tune Thanks, but I really do hate ads personally, which is why I have done my best to keep them switched off. Sadly, I received an email from KZbin just yesterday saying that they will be unleashing unwanted Ads on all videos outside the US (which affects my channel) from June 1st. And in cases where I choose NOT to monetize, they will display Ads and take all the revenue. The "good old days" of KZbin are gone now, I'm afraid.

[@Wes12940]

this video is gold! criminally undersubscribed channel! thank you sir!

[@FrankConforti]

Thank you for a very in depth content regarding capacitors. For most hobbyist types like me, caps have always been a bit mysterious. Almost all other familiar components with exception of chokes maybe, it’s pretty straight forward. I really appreciated how you took the Mouser filter to the nth degree. And you description of tantalums was dead on. A question… is there a modern replacement for RIFA caps? I like the idea of having filtering on the mains.

[@JDW-]

Hi, Frank. In the Mouser cart I provide in the text description under my Mac Plus Analog Board recap video, I use a Ceramic Safety capacitor with X & Y ratings as a replacement for C33 & C37: www.mouser.com/ProductDetail/80-C971U472MUWDBWL40 So when replacing a RIFA, you just want to make sure your ceramic has the correct high voltage rating, then make sure it has the X1 & Y1 safety ratings, ensure the capacitance is the same, and make sure the lead spacing is such that it will fit nicely. Some of my Analog Boards have RIFAs while others have the ceramics -- all original parts from back in the day. So there's nothing inherently wrong with substituting a plastic and paper RIFA with a good ceramic safety capacitor.

[@Touchstone64]

Honestly, thank you so much for taking the time to pragmatically walk through your process, I feel like I've absorbed a lifetime of learning in about an hour 🤣 I'm a lot more informed about choosing replacement caps - and when I should and maybe should not - but there are still practical problems I can't solve no matter how much logic I apply or how many explanations I watch. For example - how to replace a surface mount cap when the code is obscure or doesn't match a pattern I know? May I ask, where do you look @JDW (or anyone with experience here!) Again, for example: an SMD cap defines its properties in three lines of text : first line "6 5", second line "100", last line "25A". Does the second line (100) mean 10pF or 100uF, and last line 25V and "series A"? And then what's "6 5"??? I'm stumped and hoping for some inspiration. There HAS to be a definitive key, doesn't there? Yours in hope!

[@JDW-]

The world of SMD chip capacitors is a mess of inconsistency. You can get started examining that world by reading through this web page: www.iequalscdvdt.com/Markings_and_Codes.html But there are still times when you will need more help than that web page. Sometimes you are lucky and have a spare board of something you're recapping, such that you can remove a good cap from the board and then use an LCR meter like the DE-5000 to measure it. That will tell you the capacitance. But you then need to figure out other aspects of it like voltage. But if you know the maximum voltage it will see, then you can choose the voltage spec of your replacement cap, in accordance with the type of cap it is. For example, solid tantalum must be twice the working voltage. Polymer Tantalum can be less than that. Ceramic needs to be a lot more than double to avoid DC Bias voltage impacts on Capacitance. Also, if you know the TYPE of cap it is, the BODY SIZE will often tell you the voltage rating. You can check similar SMD caps of the same type (tantalum, ceramic, etc.) and see what I mean on Mouser regarding body size as it pertains to the voltage spec. Another help is a SCHEMATIC. You can then know voltage and capacitance, and sometimes you can figure out the TYPE of capacitor as well. Unfortunately, the SMD world isn't as friendly when it comes to markings as through-hole capacitors that are very easy to figure out much of the time. SMD takes more time and effort. And in the worst case, you may need to post about your replacement in an online forum like the EEVBlog Forum and ask people with a photo of your cap. Some of those folks can be surprisingly helpful!

[@nuckenfutz9983]

Thanks a lot for this video. Watched all the way through and filled in some knowledge gaps. Especially about ripple current handlings effect on ESR/Impedance. Never made that connection. I was hoping you would go into more detail of.. The importance of ESR, The possibility of swapping electrolytics with polymers, Identifying a given capacitors manufacturer/data sheet for as close to 1:1 replacements as possible or when upgrades are in order And when to select a capacitor based on its low frequency ripple handling, or high frequency. Oscilloscope time? I figure a multimeter isn't advanced enough to properly show what's going on. These concerns stem from dealing with switching mode power supplies (more complicated than I expected), electronics with older electrolytic ultra low ESR VRM capacitors which aren't manufactured anymore, and low profile disk drive capacitors which are limited on space for identifying marks.

[@JDW-]

Thank you for watching and for providing feedback. One good example demonstrating "the importance of ESR" is shown my other video here (42min, 8sec): kzbin.info/www/bejne/h6DJXoqgn8Sljas If you are not recapping a switch mode power supply, you can mostly swap electrolytic for low ESR polymer caps. But as you could see in the video I just linked, with switch mode supplies, even swapping an electrolytic for another electrolytic can be problematic if the switch mode design expects minimum ESR. But recapping motherboards and the like actually benefit from the super low ESR of polymer caps, so you basically just need to make sure you get the same capacitance and voltage ratings, then swap out those electrolytic. I would recommend solid electrolyte polymers over the "hybrid polymers" because the "hybrid" ones are fluid filled and could possibly leak one day. Because you can swap out electrolytic caps in most cases (other than switching PSUs), you need not give such a close eye to finding perfect 1:1 replacements much of the time. For example, if I swap out an SMD electrolytic 47µF 25V cap with a 47µF 25V solid electrolyte polymer cap on a motherboard, my replacement is technically not 1:1 but the replacement will still be better than the original. If you want to get deeply involved in ripple, you need to go beyond the data sheet and have an oscilloscope to actually test the ripple once your cap replacement is done. If you are new to scopes and just need something basic, there's not need for anything fancy. This GW Instek scope costs less than US$300, is from a reputable vendor and made by a very reputable manufacturer in Taiwan, with a fast UI (compared to other low end scopes in its class), and has a strong following on the EEVBlog forum: www.tequipment.net/Instek/GDS-1102B/Digital-Oscilloscopes/ (The photo shows the more expensive 4-channel model, but the link I gave you is to the 2-channel version, which honestly, is all you need for a starter scope. That same model is also 100MHz, which is really all you need.) Low profile (short, not tall) capacitors can be challenging to find in the right specs needed for a swap. When recapping floppy drives, for example, I often want to swap with solid tantalums but can't because they are taller than the stock electrolytic due to the fact I must properly voltage derate the tantalums. So yes, I agree that recapping switch mode power supplies in general can be troublesome due to the ESR requirements of some designs, which is further compounded by the fact you sometimes have very little space for the replacement caps and find the best choices for replacement simply will not physically fit. Sometimes you can lay a cap on its side to get around height limitations, but not always.

[@nuckenfutz9983]

@@JDW- Almost lost this notification. Appreciate you taking the time going into further detail regarding this. Between the knowledge you, Tech Tangents and a few others have given me, I feel pretty confident sourcing my own replacement caps without doubts. Thanks for sharing.

[@justincase3880]

Nice job .. thx for taking the time on this obscure, but relevant subject …

[@JDW-]

Thank you for your kind words, Justin.

[@ecco222]

I was hoping you would discuss ripple current. I always ignore this specification when capacitor shopping and there are times when I shouldn't.

[@JDW-]

Basically "ripple" is any residual AC waveform on a DC power line, whether that be from AC to DC conversion or from a switch mode power supply. The higher the capacitor ESR, the hotter it will get when the cap is used on a DC voltage rail with a lot of "ripple." So without going into extreme technical detail, the approach to mitigate the heat up inside the cap is to choose a cap with low ESR. If you look at capacitor spec sheets, you will see that a lower ESR corresponds to the capacitor being able to handle a higher amount of ripple. It gets more complex than this, but you won't always know the amount of ripple you have in a given circuit you didn't design, especially if you don't have advanced test equipment to check it. So when replacing stock caps, such as in a vintage computer, it's usually best to go with a capacitor that is rated as having "low ESR" where possible. You can always feel your caps too, and if they get really hot when in typical use (not heat flowing on them from other hot components, but from their own internal heat), then you know you probably need a lower ESR capacitor when and if available.

[@ecco222]

@@JDW- The problem with finding "low ESR" is that manufacturers usually don't provide an ESR value for typical through hole aluminum electrolytics, the kind found in power supplies. It's not a criteria that can be filtered for when shopping. Data sheets have ripple current, Q, dissipation factor, and tan values and I think these are all related to ESR but I don't know what to look for. This makes me wonder how websites that sell "cap kits", like Console5, determine that the capacitors included in their kits are "low ESR". Did they find data sheets for all of the original capacitors and then source modern equivalents? And what value is considered "low" for ESR? I think Console5 would need to have more information about the circuit, such as data sheets for the ICs within the circuit to find out what ESR value is required. ESR requirements seem to be more dependent on the application and saying "low ESR" is arbitrary. I have also read about situations where having too low of ESR is not desirable.

[@JDW-]

@@ecco222 "Low ESR" (which involves Q, DF, TAN, etc.) is a complex subject beyond the scope of this comments section, but "low ESR" is not "arbitrary" terminology insofar as if you search for 1uF radial electrolytic caps in the 50V range, for example, a good number are marked precisely as that - "low ESR" - which denotes "lower ESR than typical." What is typical? Well, basically, it's the standard ESR of most of the the caps in that category. So a smaller subset have ESR that is lower than the majority average. Answering the question of "how low is best or good enough" is not something I can easily specify because it really does depend on the circuit board design. A lot of it is trial and error unless you are the electrical engineer designing a given PCB. For example, I didn't know that ultra low ESR OS-CON SMD caps installed on a Macintosh LC575 motherboard would allow the CPU to overclock to 50MHz -- well beyond it's stock speed of 33MHz (when paired with the right CPU and fast VRAM). Previously, overclockers of that board had not been able to go so high in large part because they recapped with solid tantalum, which has much higher ESR than OS-CON. Again, the point here is that I being an EE suspected that very low ESR caps might do some good on that LC575 board because higher clocked CPUs require more sudden bursts of power and therefore will tax the caps, so I bought the ultra low ESR OS-CONs and tested, and then overclocked, and found I was able to achieve something others were not. I had no schematics and had not done circuit analysis. I just decided to go with trial and error with regard to very low ESR, and it had a positive net result. So long as you avoid ceramic caps (unless you know they are needed when recapping vintage electronics), you can usually get away with very low ESR from solid polymer caps like OS-CON. The only time ultra low ESR can become a problem is in some older switching power supplies which require a minimum ESR. Most of the time, you won't have a schematic to know that, so again, it's trial and error. I came across that when recapping an Apple HD20SC power supply (and I have a video on that too).

[@DrBroncanuus]

hi, thank you for this vid I learnt a lot but could you do a vid about selecting audio caps for the crossover in speakers ?

[@JDW-]

Thank you for your kind words, but I need to defer to the audio experts for that kind of video, I'm afraid. Your best resource for that kind of info is on one of the many audio related forums. The experts in those forums should be able to help you with all the specifics you need for your particular application.

[@ronniepirtlejr2606]

Super Walmart now sell a hardplastic digital calipers for under $10. They are extremely accurate and repeatable on measurements. I was very surprised! They are nowhere near mitutoyo quality! For a hobbyist they are perfect!

[@JDW-]

Ronnie, you’re very kind recommendation can potentially save people a lot of money so I will make it a pinned comment. Also, could you please confirm if the product shown at the URL below is the one you were talking about? www.walmart.com/ip/Digital-Caliper-0-6-Calipers-Measuring-Tool-Electronic-Micrometer-Caliper-Large-LCD-Screen-Vernier-Auto-Off-Feature-Inch-Metric-Conversion/960691125?athbdg=L1700

[@ronniepirtlejr2606]

@@JDW- yes, that is close enough. Mine has black buttons. Thank you!👍

[@surgingcircuits6955]

If money is king, Harbor Freight also sells them.

[@CloudSpecter]

Love your channel, you have a really relaxing voice, you are like the Bob Ross of Macintosh!

[@JDW-]

I'll have to work on my hair a bit before I can accept that honor, but thank you for your very kind words! :-)

[@Breakfast_of_Champions]

Agree on AGE being the main criterium. The electrolyte degrades even more from disuse. Powering it on with full voltage will lead to "self healing" degradation. The voltage rating could have been explained by the volts RMS of the circuit in reality including higher peak voltage, versus the absolute value printed on the cap. Tantalums are convenient but touchy, they mainly die from ripple AC from aging caps in power supplies.

[@commonsense5105]

JDW, very good informational video. I plan on doing a recap on a mid 70s vintage stereo receiver and based on your rework experience, do you have a soldering tip shape you prefer in doing that type of work. Thanks and this video will be a great reference for decades to come.

[@JDW-]

For equipment that old, even a fairly large sized rounded/conical tip would probably work just fine. The all-around best tips for general soldering though are beveled/angled tips.

[@isovideo7497]

I find that capacitor size is useful. If capacitors have become smaller, then I may end up getting a capacitor with a larger voltage rating than needed - that's ok. Lead spacing is measured by calipers across the outer sides of the two leads after cutting off the old capacitor, and then subtracting a lead width, which can also be measured easily by the calipers.

[@JDW-]

Correct. And by going with a higher voltage to get a physically larger capacitor, you will often find the ESR is slightly lower too, which is most often a good thing. Just make sure the HEIGHT of the capacitor will fit within your application's enclosure.

[@brianzwuup6875]

TANK you for this really awsome video! at first i wasnt sure what to do when replacing a capasitor. but now with all of the information from your video in got my dead monitor alive again!!!! its a: Samsung Syncmaster 225BW

[@JDW-]

That's wonderful to hear, Brian. Thank you very much for having kindly made the time to let me know. I'm always overjoyed to hear repair success stories like yours. Best wishes!

[@rebootedelectronics]

25:00 Possibly no datasheet because it's a Marketplace product. Sold through Digikey, but, in this case, shipped by TubeDepot

[@michalgajdos7575]

Loathe LEAD-FREE SOLDER,JOINTS. Especially dry joints manufactured in early days of Lead free implementation. And when I returned to soldering after two decades of total passivity, I didn't know what is going wrong.. soldering was easy when i was barely TEEN,and now I couldn't joint two cables together, leave alone audio connector...

[@JDW-]

Indeed, and here's an excellent overview of the TIN WHISKER PROBLEM pertaining to Lead-free solder: www.analog.com/en/technical-articles/tin-whiskers-are-real-and-complex.html

[@mikeconnor3602]

This is so true, last I soldered a substantial amount was Hafler Amp and preamp kits in the mid 80s I just completed a blue ESR meter and within minutes noticed that lead free sucked and required higher temps. Thank heavens I still had legit fine rosin core solder from the mid 80s that flowed beautifully. You gotta love so called "progress". I'm going to be on a hunt for the real stuff, is it available anywhere?

[@deleteduser3749]

Thanks for the great radio. Can't wait to recap my cb radio. I wonder if the lower noise capacitors would be an upgrade??

[@JDW-]

That would depend on what you mean by "lower noise capacitors". But I lack experience with CB radios of any kind, so I am not really equipped to answer your question satisfactorily.

[@deleteduser3749]

@jdw- Sorry, I was thinking about the aluminum organic polymer ones with the lower ESR. Thinking the lower resonance would change the signal quality? Perhaps I'm just guessing though...?

[@JDW-]

Without having a schematic or otherwise a firm understanding of the electronic design, it’s really difficult to say. But sometimes you don’t really know until you actually change out the capacitors for a different type. There’s no harm in trying so long as your desoldering and soldering skills are reasonably good.

[@deleteduser3749]

@@JDW- yes, thanks for the inspirational video. I can't wait to experiment.

[@HobbyOrganist]

I have a Radio Shack STA 85 AM/FM amp I bought new in 1979, used it on all my computers basically 10-12 hours a day for powering my wall speakers, it started putting out periodic noise about 4 years ago so I bought a new unit, but I miss the STA 85 because my gear in all on a rack in the basement and I turn the power off to the amp, printer and monitor when I'm at work, with the STA 85's mechanical controls it always was set at the same volume level etc when the power was turned on, the new stuff uses electronic controls and I have to go downstairs and turn it on and set the volume every time, thinking maybe I could replace the caps in the STA 85 and that might be usable once more!

[@JDW-]

Assuming no other components have gone bad, a complete recap of all the aluminum electrolytic capacitors very well may bring the old STA 85 back to perfect working condition again.

[@Dennis-uc2gm]

I've seen aged tantalum's (teardrop style) short and burn more than once in older test equipment. Usually these will be 10uf or less but multiple distributions across the entire voltage buss throughout the entire circuit board. I usually replace them all with electrolytic. Yes the ESR value is not as low but have yet to see it really effect the operation .

[@JDW-]

The age of the equipment doesn't have much to do with it. Basically, it comes down to bad design. Many EE's out there simply design the board wrong. You absolutely must voltage spec your solid tantalum capacitors at no less than twice the board's working voltage. Replacing tantalum caps with electrolytic will replace one problem for another. Electrolytics will never burn like tents, but when it gets very cold, ESR will exponentially increase. The other problem is you will need to swap the electrolytic about every 20 years too. I would suggest using polymer tantalum or OS-CON (solid, not hybrid) where possible if you don't want to use solid tantalum. ESR is low, stable, and the caps won't need to be replaced in 20 years.

[@Dennis-uc2gm]

@@JDW- In all the times I've seen them short and burn I have to say they maybe were rated only 5 volts or so better than what the rail voltage they were on. I'm OK with the electrolytic's, in 20 years it will be the next care takers job to maintain :-)

[@JDW-]

@@Dennis-uc2gm On a 5V rail, a 10V rated solid tantalum would suffice, but on a 12V rail, a 16V solid tantalum would be inadequate for long-term reliability. You definitely need 25V solid tantalum parts on 12V lines. But of course, you solution to use electrolytic is fine, especially if your devices are not use in sub 0°C environments where higher ESR may become an issue.

[@pauldow1648]

Maybe you mentioned this.... When leaving cap legs to resolder to - consider a heat sink, if needed. To prevent heat damage to new cap.

[@JDW-]

You are referring to 43:45 in my video where I show how I did a couple Axial lead capacitor replacements on an SE/30 motherboard. That was done many years ago before I had a hot air station. With hot air, you can more easily and safely remove the old leads completely from multi-layered boards. Even so, not everyone has a hot air station or knows how to properly use it, in which case my advice in that section remains sound. Heatsinking is fine, but it is not needed if you solder correctly, which means keep your soldering time on each leg to about 3 seconds or less. This is what most data sheets will tell you. Aluminum heatsink clips can protect the cap from excessive soldering iron heat in theory, but the heatsink also draws heat from your soldering iron tip, which could possibly result in a cold solder joint for someone inexperienced. And since one of the existing leg nubs is soldered to a ground plane which is a large heatsink, that leg will take a fair amount of heat for the solder to properly stick to it. When soldering to existing leg nubs, it's important to ensure your new cap is definitely soldered on well. Adding a little flux might help.

[@instantwow]

This is an awesome video. This helped answer many of my questions. I have a circuit that has a 0.33uF polarized electrolytic cap. I couldn’t find a polarized electrolyte version, but did find a non-polarized / bipolar electrolytic capacitor. This is for a video time base corrector. The voltage rating is the same. Would this be an appropriate substitution?

[@JDW-]

Are you absolutely sure it is 0.33 µF instead of 33 µF? If so, it’s not surprising that most replacements you are finding are not polarized. Many capacitors less than 1 µF are often not polarized. The good news is that for those small capacitance values, you may even benefit from using a film capacitor instead, as that contains no fluid and would last the life of your device. In other words, it would never need to be replaced in the future. Of course, the biggest issue about using a capacitor of a different shape is the lead spacing and the spacing around the capacitor itself, so you need to measure and determine what will fit into that space before you buy a replacement.

[@instantwow]

@@JDW- thanks for the quick reply. Good question about the 0.33 uF value. That is correct and it is also listed at such in the service manual. May consider a different capacitor type. Aside from bipolars being more expensive, can they be subbed for a polarized electrolytic in a pinch?

[@JDW-]

You can use non-polarized capacitors in a pinch or really anytime. The only real disadvantage to non-polarized capacitors are cost and capacitance choices. For a deeper dive into the subject, I highly recommend the following article… www.elprocus.com/non-polarized-capacitor/

[@EJEuth]

Amazing video lecture! Thank you for posting it.

[@tanveernaeem1157]

Most often paper and wax capacitors in the older gear are also the cuprits.

[@darrellsharrock3859]

An ESR meter is useful too . Dc leakage in cheap caps is not always obvious

[@darrellsharrock3859]

Thanks for listening , have fun . Ps you can build your own E.S.R meter as a kit

[@JDW-]

Please expand the text description I wrote under the video and keyword search for "DE-5000". That is the single best handheld ESR meter money can buy. It's the one I use, and it's fabulous. You can choose 100Hz, 120Hz, 1kHz, 10kHz or 100kHz. Most data sheets use either 120Hz or 100kHz, so the DE-5000 has you covered. Here's a little tip... If you set the DE-5000 to capacitance mode and then try to measure ESR but find it is too low to register, switch the meter to Resistance mode. It will still allow you to check using 120Hz, 100kHz, etc, but now you can see the lower value that wouldn't display in Capacitance mode. It's really an awesome meter that should be in every electronic hobbyist's toolbox!

[@darrellsharrock3859]

@@JDW- thanks .

[@thehumbleone1983]

Hi your video is very good and helpful for someone like who knows nothing about this so I have a question do you think I can use Tantulum capacitors like the one you used to recap the motherboard in this video 👍 your video 👍

[@JDW-]

Yes, you can use solid tantalum to recap an SE/30 motherboard.

[@jeremygieske165]

This was a great and timely video as I was just looking to put in my first order with Digi-key to replace capacitors on a Vtech Laser 128 (Apple II clone). A couple questions that I hoping you can help a newbie with: 1. From what I read, higher voltage is fine - so thinking that most modern caps are smaller, I was going to buy more in bulk of the different capacitance values I needed but basically get all 50v caps, even though the original board had a mix of 16v, 25 and 50v ratings according to the parts list in the service manual. My thought was that many of these capacitance values are common in other vintage computers as well and I could essentially have a stock of on-hand of viable capacitors for several computers without specing each one out. Based on your video I'm not sure if that is a great idea now . . . 2. What is your opinion on pre-emptively recapping? I was going to just recap the entire Laser, knowing all the caps in there have to be 35+ years old; however, I've read others say it's only good to recap if you see damage and/or a high ESR for a specific cap. Thanks again - appreciate the advice!

[@JDW-]

A lot of people make mistakes with vintage electronics because they listen too much to "consensus." They check social media groups and online forums, then see what most people say, then decide based on that. It seems logical to conclude that a group of minds should be largely correct, but that just isn't the case. Sometimes listening to the voices works out well, but the problem is most of those people are not electrical engineers. Most of those people on those online groups say things like, "I've never had a problem with certain capacitors, so that means those caps are ok." But that is incorrect. If you look at data from all the major capacitor manufacturers, they will tell you how to calculator capacitor life and they put a disclaimer that regardless of that life calculation, the manufacturer will not guarantee a service life of beyond 15 years (for fluid filled aluminum electrolytic capacitors). That doesn't mean your old cap will die on its 16th birthday, but it does mean that if you have fluid-filled caps older than 20 years, you need to replace them regardless of what others may recommend. And although it is true I am giving you my recommendations right now, (1) I am basing my recommendations on what the manufacturers say in their official documents, and (2) I am an electrical engineer who is also experienced in recapping old electronics. "Pre-emptive" or "before it breaks" recapping is the best choice for any device that's 20 years old or older, when it comes to fluid-filled aluminum electrolytic capacitors. Buying a lot of aluminum electrolytic capacitors and slowly pulling from that stock to recap various devices isn't a bad idea but there are a couple caveats worth mentioning. First, when recapping power supplies, you will find that some are sensitive to capacitor ESR (resistivity). If you replace an old cap with a new low-ESR cap, there could be trouble. Sometimes you won't know that until after your recap. I know this from personal experience and from my own knowledge of switching power supply design. That's why it's important to check output voltages on an oscilloscope before and after your power supply recap to ensure the output noise is no worse that it was before your recap. And if there is a problem, you may need to buy a different capacitor which has higher ESR to solve the problem. The second caveat is that if you use up your capacitor inventory very slowly, over the course of say 10 years, then you have to consider that there is a shelf life. All manufacturers publish information about shelf life for aluminum electrolytic capacitors. I personally think it's fine if you use capacitors that are several years old, especially if they've been stored in a climate controlled environment. But it's important to keep in mind that nothing in this life lasts forever. Everything degrades over time. So I wouldn't want to recap something with an old stock of fluid-filled capacitors that is 20 or 30 years old, for example. Lastly, it's important to keep in mind I am talking to you about Aluminum Electrolytic Capacitors. Tantalum capacitors, and Organic Polymer Aluminum Electrolytic capacitors with a solid electrolyte could be 20 years old and still work great in a recapping project. They have no fluid that dries up over time. There are also other caps like film capacitors which also have no fluid inside and have a very long shelf life (if stored properly). Mostly, you need to recap fluid filled aluminum electrolytic capacitors. The exceptions to that rule are X & Y rated capacitors that are sometimes called "RIFA" caps, which are used at the AC input side of a power supply. You will find these in the Macintosh 128K, 512K and Plus, for example. Those sometimes die due to age and voltage spikes. It's often good to replace those if they are 20+ years old, although it isn't so critically important as the need to replace fluid-filled capacitors. Tantalum capacitors may also need to be replaced if you see that any are burned in a circuit, but keep in mind that burned tantalum capacitors means there could be other problems in a circuit that needs fixing. I hope this information helps. If you have further questions, please do not hesitate to ask.

[@jeremygieske165]

@@JDW- This helps a great deal - thank you very much for the detailed response and information. More for me to think about. I like the idea of using tantalum from the perspective of not having to worry about potential leaking caps again 15-20 years from now; however, I'm not sure with the Laser if there are good options as it's through-hole, and for some reason some of the existing caps on the board have high voltage values (ie a 47uF with 50v). When looking on Mauser I didn't see anything practical with that high of values. Regardless, I am leaning back towards recapping the entire board, vs. piece-meal. Thanks again for the advice/information!

[@JDW-]

@@jeremygieske165 I'm sorry I cannot give specifics. I've never owned even a single Apple I or II or clone. My guess is the highest voltage is only 12V on the board, but you would need a schematic to know for sure. But don't let those 50V stock caps fool you. A lot of small capacitance value caps high very high voltage ratings like that. The higher the voltage, the bigger the cap, but the lower the ESR, you see. So try to find a schematic, or post in Apple II or Laser forums online (or in Apple II FB Groups) to see if anyone knows if the board has a voltage higher than 12V. If not, you know that any SOLID tantalum cap you choose must be rated for 25V (or higher) to be safe. (If you need thru-hole, it means you cannot choose Polymer type Tantalum, only SOLID.) Keep in mind that solid electrolyte OS-CON caps sometimes come in radial (thru-hole) types too, and those are just as good as tantalums, so long as you get the solid electrolyte versions. Don't buy any Organic OS-CON cap that says "hybrid" because those have a fluid electrolyte that could eventually leak. You want solid electrolyte replacement caps for longevity. Lastly, just be sure to measure your board to see how much diameter and height space you have, then choose replacements that will fit within that space. Try to measure the distance between the leads too because "Lead Spacing" is also part of the capacitor specifications. I hope this helps.

[@jeremygieske165]

@@JDW- All great advice. Thank you again for the thoughtful and detailed response!

[@NullStaticVoid]

I only go for Nichicon or Wima for stuff that is in the audio path.

[@dwheat9003]

Most certainly I have had to be sure solder mates to leads more than I have with lead solder sometimes. One good reason I like lead free is my grandkids won't be subjected to harmful lead. Bought a VOM new old stock that was in-op, looking at soldering on board it simply had poorly flowed lead free solder in about 5 instances which was the cause of some of its functions being non operative. My point it seems as though the lead tin types pool in faster and more assuredly seems like.

[@JDW-]

Whether using leaded or lead-free solder, using supplementary flux can be critically important at times so you can get the solder not only to flow nicely but also attach itself to PCB pads and component pins. In my own work, I've found extra flux to be more important for SMD work than thru-hole, but it still benefits both. I prefer leaded solder to avoid the tin whiskers problem inherent to most lead free solders, and because I like flowing solder with lower temperatures. There's actually more harm that can come from breathing in flux fumes than from touching leaded solder. So long as you don't ingest leaded solder, you will be fine, especially if you wash your hands after a long soldering session.

[@smop8388]

Great video, as always. Thank you ! I'm just surprised how little has been said about MLCC (ceramic) capacitors. I've seen them used fairly often for vintage Mac recaps. What's wrong with them ?

[@JDW-]

Thank you for your kind words. If you’re looking for a capacitor with very low ESR, you can't do better than a ceramic capacitor. But let's ponder why NOT to use a ceramic (MLCC). First, the physical size of ceramic capacitors increase and decrease (changes so small it's imperceptible to the human eye) as applied voltage increases and decreases. It sounds pretty crazy, but it’s true. So while ceramic capacitors are very often chosen for the output of switching power supplies due to their extremely low ESR, they can also make a lot of noise because they vibrate the entire PCB, which amplifies the sound like a speaker. If you've ever heard a constant buzz when using a product that has a switching power supply, ceramic capacitors are the reason! There are some expensive specialized ceramic capacitors that are made to eliminate that vibration on the PCB, and sometimes you can design a PCB with holes cut around the capacitor itself to attenuate the vibrations of the PCB itself -- but for recapping that is not practical. Yet another big problem with ceramic capacitors is that they have an even more strict voltage Derating than solid tantalum capacitors. Why? Because the Capacitance of ceramic capacitors decreases according to the voltage (DC bias) you apply to them. That's right... The higher the voltage across a ceramic, the lower its capacitance! Sometimes capacitance can decrease as much as 80%. Imagine a situation where you want to install a 1µF capacitor in a circuit that sees 10Vdc. You might need a 100v rated ceramic capacitor so as to maintain a stable 1µF of capacitance. The downsides to higher voltage rating for ceramic capacitors are Cost and Physical Size of the capacitor. Even if you have enough circuit board space to accommodate a physically larger ceramic capacitor, and even if money is no object so the higher cost doesn't matter, you still may NOT wish to install a physically large ceramic capacitor because ceramic capacitors are ceramic and therefore brittle. Depending on the size of the circuit board or product, a fall to a hard surface from the height of even a couple feet could crack large ceramic capacitors.

[@smop8388]

@@JDW- Thank you again for this clarification. I will look into it. Your videos, your work, and the attention you pay to your viewers are absolutely remarkable. Your channel is undoubtedly one of the best of its kind on KZbin, if not the best.

[@JDW-]

@@smop8388 I'm truly humbled by your extremely gracious words, Smop. Thank you very much!

[@jd5787]

Hi! Super video, thanks for sharing! I am completely new to this. I would like to recap l a couple of cheap chi-fi preamps. How would you choose between caps types? Say the amp is using electrolytic caps. Would you choose better electrolytic? Or Film, Tantalum etc?... Woukd changing resistors make a huge difference in sound quality? Thank you

[@JDW-]

Thanks for watching and for your comment. I'm not an audio guy and I don't have a schematic of your circuit, so I would recommend you visit an audio related forum to ask specific questions about how the sound might be different with changes to resistors and or components other than the caps. Regarding the caps, I would suggest replacing the stock electrolytic caps with the same kind of electrolytic caps. While that should be easy to do in terms of matching the Voltage & Capacitance rating, as well as the physical dimensions and lead spacing, I cannot say what impact the replacement capacitor "ESR" will have. If your caps are bad, then measuring their ESR might not be helpful, even if they are not bad but very old, it could be their ESR is higher than normal. You might be able to find out what brand and model the stock caps are by asking at an Audio forum, and in that case, you know exactly what replacement caps to buy. Meaning, even if you can't get the stock caps because that model isn't made anymore, you would be able to look up data on the stock caps to know the ESR, and that would be your guide when picking out a replacement electrolytic. If you have no way to find out what the ESR of the stock caps were, you could opt for a general purpose capacitor to play it safe with ESR, or you could opt to buy a "low ESR" replacement and see if that works well in your circuit. If a low ESR cap causes trouble in the circuit, then you would just buy a general purpose cap instead, which should have sufficiently high ESR to remedy the issue. Lower ESR is usually better, but in some circuits, ESR can be too low and cause issues. That is why being able to find out the ESR of the stock caps can make it easier to select a suitable replacement. Having a schematic and knowing what the circuit means can also be helpful too. Best wishes!

[@drywinddotnet]

Hi James! Merry Christmas to you and your family!

[@JDW-]

Merry Christmas!

[@googleuser323]

Great video and thank you for sharing your knowledge. I am into repairing ECU's (electronic control unit) would this also apply to have electrolitic caps replaced by Polymer Tantalum? taking into account the voltage values. I haven't tried doing this but would appreciate your inputs. Again thank you for sharing.

[@JDW-]

Yes, everything you saw and heard in the video applies to caps on vehicle ECUs as well. The ECU should have a TVS diode or similar spike snubber on it's incoming 12V line, or be connected to a power source that does, which means your 12V line will see 14.4V with the engine running, and up to however much voltage the TVS diode is rated for (Vc). Keep in mind that Vc could be 33V. Meaning, if a 60V surge occurs, the TVS (with a Vc=33V) will cut that spike down to about 33V or less, and if there are other electrolytic caps in that area, they will shrink the voltage spike as well. (Electrolytic caps handle over-voltage well.) The reason I mention all this is because unlike vintage computers connected to a wall socket power supply, automotive spikes are more frequent, and whatever replacement caps are chosen need to be selected while pondering with the maximum voltage of spikes a given circuit is likely to see, which really depends on the spike filtering, which most often includes TV diodes. So if you have a TVS somewhere that limits the 12V line spikes to 33V, that's likely the maximum voltage your replacement caps would see (and then only rarely and briefly). Choosing replacement electrolytic caps at 33V in that case is a non-issue, but even polymer tantalum has a voltage derating of 20% lower than your polymer tantalum cap's specification voltage. Say you have a replacement tantalum cap rated for 40V. Well, 80% of 40V is 32V, which is probably safe enough, assuming the Vc of whatever TVS diode used is in fact 33V. You can't safely guess. You really need to know. So please give this some thought. Voltage spikes affect polymer and solid tantalum caps the most, which is why you really must ponder the maximum voltage level of spikes in a car application before replacing electrolytic caps with any kind of tantalum caps.

[@googleuser323]

@@JDW- Thanks JDW it gives me an idea on how to do about with this replacement in the near future. Stay safe and Thanks.

[@Epictronics1]

Very useful video, thanks!

[@AlsanPine]

some notes on audio gear which i have more than 40yrs experience in. i replace all ceramics i can. they are a big noise source along with composite resistors. not sure why you do not like lead free solder. i have used led free for decades and have never found a reason to go back. there are many outstanding led free solders available today. its not like it was in the old days. you do need decent soldering iron though 🙂 choosing between capacitor brands and models become more important in audio gear. it is largely irrelevant for other electronics. as i do mostly audio, the panasonic fc & fs and higher grade nichicon are the low-end caps for me. elna makes some great audio caps. for non-electrolytic caps, stick with film caps if you can. tantalum is great for computers but absolutely worthless for audio. and again, replace all your composite resistors with film ones i always use 1% which makes for more balanced output and easier fine tuning and better imaging. you will have a much more efficient system, greater headroom, better detail, and immersive soundstage. if you do the work yourself, the parts are dirt cheap. always shield your signal wires and pay attention to their routing. it takes a lot of time to do the work so do it right or not at all. cheers 🙂

[@JDW-]

I don’t doubt your good word or your 40 years experience with audio. But could you please explain why “tantalum is worthless for audio“? You have me curious. The reason I ask is because pretty much every vintage computer motherboard has an audio section on it.

[@AlsanPine]

@@JDW- true. i should have said good audio! computer audio does not have the stresses that a decent stereo has. tantalums are excellent in high frequency applications. they are more suited to digital signals. they sound positively crappy in analog. so yes they are heavily used in computers. i have extensive experience in them as well. i built my first one in the early 70's. the other issue with tantalums is their early problems with, well, blowing up! i am sure they are much better today but the reality is that they are not ideal for the variability and stresses of analog audio work. yes, they will function. in fact there are some vintage audio gear that came with tantalums. no, it did not go well 🙂 so, ok, worthless was a bit of hyperbole. but if you are going to put in all the effort and replace all the noisy analog components as i do, using tantalums will really ruin your day because it will limit the quality of the audio you can get. i have rebuilt 35w common sony integrated amps from 70's and blown away $10k discrete new audio gear in sound quality over and over again. i have a customer who spent insane amount of money on this fancy new setup and he barely uses it because the setup i made him for less than 1/5th the price sounds better. the secret is to eliminate all sources of noise even if you cannot hear it. the problem with noise is not just you hearing it. the problem is that it saps energy from the productive playing of music. this is why i shield signal wires. you may not hear the noise the composite resistors make but your amp see's it and has to deal with it which reduces your sound stage and head room as well as precision of detail. listening to such audio gear is like seeing a beautiful view through a perfectly clear window instead of a crud covered dirty one. it sounds natural and clear without all the crud in the way. but i digress 🙂

[@JDW-]

Thank you for providing further information. But what I meant was that even in a computer, it is not purely digital all the way to the speaker cone. The early audio stages are digital, but you still need an analog amplifier, however bad it may be in comparison to pro audio equipment, in order to drive that speaker cone. And so I was basically asking about capacitor types used in that analog output stage of a computer’s motherboard. When we talk about tantalum capacitors it’s important to consider that the ones which can catch fire are SOLID tantalum, but modern versions known as POLYMER tantalum do not burn and have a lower voltage derating than their SOLID counterparts.

[@AlsanPine]

@@JDW- yes, that was why i acknowledged utilizing a bit of hyperbole. technology advances and old mistakes are remedied. i do not, however, distinguish between pro and non pro audio. this is mostly a marketing scheme and my work shows it. yes there are purely crappy audio gear out there but many of the gear produced in the 70's and 80's was designed well. the problem was in attention to detail. companies have great design engineers create the perfect unit, then the bean counters get involved and push the manufacturing engineers to save a penny on each component. as an engineer yourself, you would have seen this first hand as it is an unfortunate but nearly universal problem although with my work for the government, i did not suffer from that at least. well, there are thousands of components and 10's of thousands of units produced so the company saves a lot of money. also, they can have many tiers of product moving up in "quality" so it is all great for the company. we who love audio lose. because most of the audiophile equipment is not so revolutionary. most of the cost is not the components. so when we rebuild a vintage gear and use the very best components, we incur, at most, $100 more cost (not including transformers) but the resulting gear blows away most of those "audiophile" gear out there. i have done this over and over again. with sony, with pioneer, with yamaha, with marantz, etc. if the fundamental design of the gear is good, we can bypass the corporate bean counting and create an audiophile gear at a fraction of the cost. in this world, tantalum caps do not exist, like i said, because they are not the optimum capacitor for audio. i was mentioning the historical explosions for one reason audio companies mostly stayed away from tantalums; they just do not bring anything to the table that is needed in vintage audio. their strength is their size efficiency and frequency response neither of which is useful. on computers, both are paramount hence their presence everywhere there. and since they are popular, yes even the sound circuits use them. they are not that horrible to ruin computer audio. i have however, recapped some computer audio circuits and you will hear the difference there too, again, if you have a higher tier audio card with better design. i do recall, years ago, creative labs put out a nice sound card that minimized, or avoided (can't remember) tantalums and sold it for a pretty penny. it had gold plated i/o and was a nice card. i think it was in the 90's? getting old sucks 🙂

[@JDW-]

@@AlsanPine In light of everything you said and considering your extensive experience with audio and knowledge of computer recapping, perhaps you could give me your opinion about a audio noise problem I am trying to diagnose on a vintage Macintosh Portable from 1989. Sadly, I am unable to get other Portable owners to give me feedback regarding what the typical noise floor is like, so I am somewhat working in the dark. Even so, the noise is quite noticeable when using headphones, and it's nothing like I've heard in my other vintage Macs, so I think there is a problem on the motherboard that needs to be solved. My latest video showing the troubleshooting and giving you an example of the noise is here (wearing headphones is recommended): kzbin.info/www/bejne/g6aogJWafaiVn8k My remarks about the motherboard audio circuit, motherboard schematics and other details are listed by me in the Tinker Different forum here: tinkerdifferent.com/threads/macintosh-portable-pop-noise-every-10s-from-speaker-headphones.3425/#post-29120 If you simply do not know or do not have time to give this any consideration, I fully understand. Not a problem. No pressure. But I couldn't help but consider asking you because you are an engineer with far more experience in audio than I am. I was thinking that perhaps by hearing the noise you might have some ideas I've not yet considered.

[@supergub]

I have a piece of '70s tech with mylar capacitors. Can those be left alone or should they be replaced?

[@JDW-]

Mylar caps are polyester (PET) and therefore part of the "Film capacitor" family which do not have a liquid electrolyte and do not need to be replaced (unless you know them to be bad). They have low ESR and handle high voltages well. Despite having been made in the 70's, they most likely are working just fine in the circuit.

[@supergub]

@@JDW- Thanks!

[@wmrieker]

reverse voltage on tantalums is always fun :)

[@JDW-]

You're referring to the explosion at 4:33. Yes, it is true. But sadly, even a voltage spike can trigger solid tantalums to go up in flames. This is why they need to be derated by 50% (i.e., rated for 25V if used on a 12V voltage rail).

[@yanfishtwig2356]

very detailed thank you i even read the paper you linked to another comment'er

[@JDW-]

Thank you for making time to let me know my video (and one of the comments) was of help to you. Best wishes!

[@CoffeeRunHobby]

Hi, I'm not sure I'm the only one but I did swap all caps with MLCCs with great success. I wanted to know your opinion about those. I am on the same page as you, as I don't want any liquid electrolytics. Here is where I don't agree with you: I wouldn't want any tantalum either, as those break in a dead short :) In the end I used X7R MLCCs with great attention to voltage and capacitance. The board is working fine, but I wanted to ask you why you don't recommend MLCCs? Nice video BTW

[@JDW-]

You have to be very careful when selecting an MLCC (ceramic) in a given recapping application. Here's why. The physical size of ceramic capacitors increase & decrease (a size change so small it's imperceptible to the human eye) as applied voltage increases & decreases. It sounds pretty crazy, but it’s true. So while ceramic capacitors are very often chosen for the output of switching power supplies due to their extremely low ESR, they can also make a lot of noise because they vibrate the entire PCB, which amplifies the sound like a speaker. If you've ever heard a constant buzz when using a product that has a switching power supply, ceramic capacitors are the reason! There are some expensive specialized ceramic capacitors that are made to eliminate that vibration on the PCB, and sometimes you can design a PCB with holes cut around the capacitor itself to attenuate the vibrations of the PCB itself -- but for recapping that is not practical. Not every recapping job with an MLCC will turn them into a buzzer, but this is neverthless a concern. Yet another big problem with ceramic capacitors is that they have an even more strict voltage Derating than solid tantalum capacitors. Why? Because the Capacitance of ceramic capacitors decreases according to the voltage (DC bias) you apply to them. That's right... The higher the voltage across a ceramic, the lower its capacitance! Sometimes capacitance can decrease as much as 80%. Imagine a situation where you want to install a 1µF capacitor in a circuit that sees 10Vdc. You might need a 100v rated ceramic capacitor so as to maintain a stable 1µF of capacitance. The downsides to higher voltage rating for ceramic capacitors are Cost and Physical Size of the capacitor. Even if you have enough circuit board space to accommodate a physically larger ceramic capacitor, and even if money is no object so the higher cost doesn't matter, you still may NOT wish to install a physically large ceramic capacitor because ceramic capacitors are ceramic and therefore brittle. Depending on the size of the circuit board or product, a fall to a hard surface from the height of even a couple feet could crack large ceramic capacitors. If you want even more reasons, you can read what I wrote here: tinkerdifferent.com/threads/spicy-oclock-project-has-started.128/page-3#post-7050

[@JDW-]

Regarding SOLID Tantaum, which isn't the same at all as POLYMER Tantalum (which have a benign failure mode, not shorted), it's all a matter of how to properly derate them. They are used in pacemakers and in space and in the US military. That's right. They don't always go up in a fireball when properly derated in accordance with manufacturer specifications of 50% or more. So if you will be using a tantalum cap on a 12V rail. Do NOT use a 16V rated SOLID tantalum! 12V x 2 = 24V, which is the proper voltage derating for SOLID tantalum in that application. Polymer Tantalum can be rated at 16V in that application though because it has a different voltage derating requirement. The issue with the Polymer types though is cost, which is why I tend to recap old Mac motherboards with OS-CON caps instead of Polymer Tantalum. Besides, OS-CON has lower ESR than Polymer Tantalum (which has lower ESR than SOLID Tantalum). All have lower ESR and more stable ESR than fluid filled Aluminum electrolytic capacitors though.

[@CoffeeRunHobby]

@@JDW- Thank you. I learned something new. I really appreciate your input. :) P.S. Now I remember seeing capacitors with bendy legs on some boards (in the shape of S) Maybe the reason was to stop the vibrations to reach the board.

[@KarlUppianoKarlU]

I would also mention interstage paper capacitors for people restoring vintage vacuum tube equipment. These become leaky over time, and begin to conduct DC that they're supposed to be blocking. This can throw off the bias of the downstream stage, leading to improper bias, over current and premature tube failure, as well as increased distortion, frequency response and gain errors. A simple ohmmeter test may not show the problem, because leakage can increase dramatically at higher voltages. Most restorers simply yard out all the paper caps and replace them with good quality film capacitors of similar voltage and capacitance. The newer film caps are usually much smaller than the originals, making positioning easier. Although, if the original appearance is of a high priority, it may be possible to hollow out the old paper cap and put a nice new plastic film cap inside. Electrolytic caps in vacuum tube equipment are usually always out of spec, and should be replaced as a general rule well. Again, newer devices are usually much smaller, and modern packaging is often much different. Some clever physical rework for esthetic purposes is optional. Mica and ceramic caps are usually not a problem in tube equipment, but it's not unheard of.

[@JDW-]

Hollowing out the stock cap in order to fit a modern film cap inside is a wonderful suggestion! Thank you for watching and for taking the time to write such a helpful and detailed comment.

[@peterfong2241]

Blessed love to all, great video bro Can you recommend a gud capacitor tester for me please an thanks, watching from Jamaica west Indies !!

[@JDW-]

Yes. In the text description under the video, I mention the DE-5000 LCR meter and put a link to Amazon. You might be able to buy it from your country too.

[@jakass]

Thanks for the trove of information

[@JDW-]

Thank you for watching and for your kind comment, JJ!

[@playstation2bigs]

Does replacing all electrolyte capacitors with SOLID capacitors makes my CRT tv last another 30 years ?

[@JDW-]

If the replacements are properly chosen, in terms of the capacitors, the answer to your question is yes. But it could be that another component within your CRT TV might fail in the next 30 years. Specifically, the flyback transformer is often a point of failure.

[@edwardbass9255]

What is the best way to determine the D-Rating of Capacitors on Mouser? Thanks

[@JDW-]

By "D-Rating" did you mean to say "Derating"? Or do you mean Dielectric constant or Dielectric strength? For now I will assume you meant to say "derating." Such is not given on Mouser, and it is not always given in capacitor data sheets. But guidance regarding capacitor Derating is typically given in general info publications by capacitor manufacturers. What I mentioned about "derating" in my video pertains mainly to TANTALUM capacitors, as they are most sensitive and susceptible to dying due to a voltage spike across them. So when I talked about "derating" for those caps, it pertains specifically to "VOLTAGE derating," which means you select tantalum capacitors with a voltage that is a certain amount higher than your circuit's working voltage. For SOLID tantalum, it's a 50% voltage derating, which means if your working voltage is 5V, you should select a SOLID tantalum cap that is rated for 10V or higher because 50% of 10V is 5V. POLYMER tantalum has a more forgiving Voltage derating of 20%, which means that in a 5V circuit, your Polymer tantalum cap needs to be rated at 6.25V or higher because 80% of 6.25V is 5.0V. There is no 6.25V rating for capacitors, so you choose the closest one, which is that case is 6.3v. Ceramic capacitors too (all types except NP0/C0G) have a DC Bias effect which dictates that you consider a voltage Derating in order to ensure you get the nominal capacitance you expect in your circuit. So for example, if your circuit is 5V and you choose a 5V ceramic cap, it won't fry due to a voltage spike like a 5V rated tantalum cap would, but you might get only 20% of your expected capacitance. Why? Because, quite sadly, ceramic caps like X7R lose capacitance as your applied voltage increases. It's the nature of the beast. NP0/C0G don't have that problem, but they are only available in very small capacitance sizes. So you normally must use a ceramic cap voltage rating that is much, much higher than your working voltage in order to ensure you get the full rated capacitance. If you use a 50V ceramic X7R cap in a 5V circuit, you will have no issues, for example. But the most accurate way to choose the exact voltage you need is to find the DC Bias curve (graph) in a ceramic capacitor publication, which shows you Voltage versus Capacitance, and then you can very accurately figure out what your ceramic cap's voltage rating needs to be to ensure you get the minimum capacitance it needs to supply in a given circuit. I hope this helps.

[@RickL_was_here]

Interesting vid. I'm still fairly new at board repairs, no Apple stuff though. What got me started however, was the amp board on my B&W sub. That thing had a couple smaller (ely) caps that, although didn't looked bad themselves, had some serious burning on the board around them. I decided to pull them off after watching a few videos and change them out. What I found doing the research though was that the higher voltage was ok, just so long as the uf remained the same (couldn't find the correct voltage with my search at the time). Anyway, it works flawlessly. Just wanted to point that out and if I somehow made a mistake with the higher voltage, what would that be and why would others claim that to be ok? Just for high powered stuff?

[@JDW-]

I consider "high powered stuff" to be either in the hundreds of volts or multiple Amps of current, but regardless of that, the advice you were given is sound, and that is what I present in my video too. If you have a 25V 10µF capacitor, for example, replace it with a 25V or higher 10µF capacitor. In other words, 50V, 100V, etc. will work fine. Can you go lower than the stock voltage spec? For example, use a 16V instead of a 25V stock cap? Actually, you can if you know the maximum voltage that capacitor will see. The voltage spec is the maximum voltage the cap should see on a regular basis, so never use a 5V rated cap on a 12V voltage line. But most people lack schematics or otherwise don't know the maximum voltages in their circuit, so using the stock caps as a guide is often the best you can do. But going with a higher voltage cap has benefits of marginally lower ESR. You can also swap with the same type of cap, such as swapping an electrolytic for electrolytic; but you can sometimes swap an electrolytic for a solid tantalum, polymer tantalum, polymer aluminum electrolytic, or even film capacitor. Most of that is discussed in my video.

[@RickL_was_here]

Appreciate the response, wasn't expecting one from an older video. I understand the "high power" response, I've worked as an electrician in the past and have a ticket for "low" voltage. Most people don't distinguish the differences (low being 600 or less). By "high", what I meant was in regards to amplifier circuits (or other devices with massive caps). I must have misunderstood what I heard you talking about, I thought you'd indicated going with a higher voltage could cause problems. I was listening to this while working so I will try not do that next time. I was aware of the ability to change them out for different types but have yet to experiment (aside from the higher voltage caps I swapped). Cheers and thanks again for the response!

[@JDW-]

@@RickL_was_here I read and reply to every comment under any video that needs a reply, regardless of the age of my videos. I try to help out wherever possible. But no, I did not say in my video that using a higher voltage spec'd capacitor (relative to the voltage rating on the stock cap) would cause problems. It won't. You might see a modest reduction in ESR when you go with a higher voltage rated cap, but most of the time, your circuit really won't care about that difference. And most of the time, lower ESR is better anyway. Just as one example, on vintage Macintosh analog/power-supply boards, there is a cap labeled C1 which originally was 3.9µF 35V (not polarized). That is often swapped out for a 100V non-polar 3.9µF cap, but I suggest a 3.9µF Metallized Polypropylene Film Capacitor, because they are easy to find on Mouser. Film caps are not polarized, and the replacement I recommend is rated for 450V. While the lead spacing differs from the original, it does fit and the higher voltage rating makes no difference at all.

[@Iruleyou1]

This was a great video but I have a question regarding purchasing caps for my Pentium 3 motherboard. I understand the basics of a cap like stated in this video but Digikey list many caps with things like, coupling, decoupling, low ESR (I get audio and automotive are just a better grade). A capacitor can meet the specs based on the typical things discussed in this video but can be one of those other designations in instead of general purpose. Do they matter in my case of recapping a P3 motherboard and if not when do they matter?

[@JDW-]

A motherboard doesn't have a switching power supply on it and therefore there is no worry about your new replacement capacitors having ESR that is "too low." So you can use "low ESR" replacement capacitors, if you like. However, "low ESR" tends to refer to electrolytic capacitors, not other capacitor types like Tantalum, Polymer Tantalum, Polymer Aluminum caps like OS-CON, or Ceramic. And if you replace all your electrolytics with more electrolytics, you will have to replace them again 20 to 25 years later. So when I recap vintage Macintosh motherboards like the SE/30 described in my video, I use capacitor types OTHER THAN electrolytics. And if you use SOLID Tantalum you must derate by 50%, and to know that you would need to know if some caps see 12V or not, and if you don't know, then your SOLID tantalum caps would all need to be rated 25V each to be safe. Or you could use Polymer Tantalum rated at 16V because they don't need as high a derating, but they are much more expensive. Or you could use OS-CON (my preferred cap these days for motherboards), because they can be rated at 16V too (or higher), and they are shaped like cans so they look stock (like electrolytics), and they have lower ESR than Solid or Polymer tantalum. You could use Ceramics, but those need to be rated at least 50V to avoid the DC bias effect where capacitance decreases according to applied voltage. I think OS-CON are better than Ceramics due to all the caveats of Ceramics. And so, when you understand all this, you can avoid any confusion about "coupling" or "decoupling" or "low ESR" or "general purpose" designations on Mouser, which again, tend to mainly describe fluid-filled electrolytics, rather than other capacitor types. Lastly, I don't use PCs or have experience recapping them, but it could be that your motherboard has large electrolytic capacitors on it which have such high capacitance that they absolutely must be replaced by capacitors of the same type, meaning, electrolytics. In that case, because it is a motherboard and not a switching power supply, what I said about ESR still holds true. You shouldn't have to worry about your replacement caps being too low ESR. So in that case, you can safely use "low ESR" electrolytic capacitors. It is only on some switching power supplies that you need to be careful with extremely low ESR caps because some power supplies want a Minimum ESR. But on a motherboard where there is no switching power supply, most all of the caps are "Decoupling" (aka "Bypass) capacitors that try to sustain voltage levels on motherboard chips so that nothing brings down the voltage on any chips, not even for a brief instant. And the lower the ESR, the faster those capacitors can respond to very brief sags in voltage. So what if there is a power supply on the motherboard? Well, I assume that would be something like a 5v to 3.3v down-converted or similar. And in the case, you would want to take care your replacement capacitors connected to that power supply aren't too low ESR. But if the capacitance is large, you would be replacing them with electrolytics, which don't have that low ESR even for caps which say "low ESR," at least not in comparison to OS-CON. So I think you probably would be fine using "low ESR" caps, if you like. When I replace caps in switching power supplies, there are many things I just don't know. I don't have a schematic. And I know in some cases ESR could be too low. So what I do is buy "general purpose" and "low ESR" and swap out the caps with my "low ESR" replacements and see if all is well. If all is well, I leave it like that. If it has issues (usually found by connecting the output to an Oscilloscope), then I can swap "general purpose" (higher ESR caps) in and out unil the problem is gone. If you don't have a scope, then recapping a switching power supply is more tricky. You would only know if your machine works or not, and you wouldn't know if power is clean or dirty because without a scope, you can't see any noise riding on your voltage output. A handheld meter won't show you that. But some scopes these days are pretty cheap, like the Rigol DHO804, which I purchased recently on AliExpress for only a few hundred Dollars, and since I use it a lot, that cash outlay was acceptable. Not worth it for only 1 recapping job though. I hope this helps.

[@Iruleyou1]

@@JDW- Thanks for the reply. This helped a lot. I wasn't asking so much if I could use low ESR caps, but more if I should be using them and if so when do I use one or the other just looking at a board. This is because you can't always tell just be reading the information on the cap. It sounds like in my case for the Pentium 3 board I really can't go wrong using them and in some cases they may be needed.

[@CrazyTechReviews]

Nice video!

[@sideburn]

Hmmm just fixed a 1977 Tektronix oscilloscope with shorted tantalums. It was quite a Smokey fire 🔥 before I found out. I’d say tantalum’s should be on the “hot list” of replacements. Also had bad tantalum’s on a Mac IIcx.

[@JDW-]

A lot of "solid" tantalums fail because the designers failed to properly voltage derate them. For example, if you put a 16V tantalum on a 12V voltage rail, you increase the likelihood of failure considerably. But if you use a 25V rated solid tantalum, it then becomes much less likely to fail. Want to reduce the risk even more while still using tantalum? Go with Polymer Tantalum instead. Not only will they last longer, they won't fail in a flame.

[@sideburn]

@@JDW- well this is a Tektronix 214. I think they are pretty high end designers. Pretty sure it was in spec. What fried was my batteries and part of the PCB because they should have fused the battery pack like the did the a/c input. I’ve since added 1a fuse to the battery and hopefully that will protect the tiny transformer behind those tantalum caps. I’ve heard these caps fail often and burn up the transformer. The tantalums i went with are “Solid Leaded”. these are 10v and 35v caps in the power supply on the transformer output side and i just noticed that the output voltages are 5v where the 10v caps are and 65v where the 35v caps are. But the caps are in series. Hmm 🤔

[@JDW-]

@@sideburn Some of the biggest names in the industry, including Apple, have used solid tantalum caps on boards and did NOT sufficiently voltage derate them. Trust me, engineers at big companies have made such errors. It also was not well understood in the distant past about the need to do a minimum 50% voltage derating for solid tantalum. It is now well understood, especially by NASA and the US Military. When a solid tantalum fails, it mostly fails shorted. So this may be while other parts in the same circuit fail too. All of the information I've been giving you about "solid" tantalums applies to both SMD and "leaded" types. Polymer Tantalums are only available in SMD types, unfortunately.

[@sideburn]

@@JDW- so should I not put these in the scope? I ordered them from mouser for preventive maintenance. I already replaced the shorted cap with an electrolytic. I was just going to replace these other 4 to be safe since I heard that they tend to short and fry the transformer. But I’ve since put a fuse on the battery so maybe I should just leave it alone.

[@JDW-]

@@sideburn If you ordered exact replacements for the stock caps, you will not trigger any immediate problems. All I am saying is that to get longer life out of solid tantalum replacements, you need to voltage derate them by 50%. That means if you have a 12V rail, you must choose a 25V rated solid tantalum. 50V rail? You need a 100V solid tantalum, and so on. What happens if you use a 16V solid tantalum on a 12V rail? Well, at some point a voltage spike will kill it, which means it goes up in flames and fails shorted. And while derating a solid tantalum by 50% may seem extreme, there are some high reliability designs where the engineers chose to derate by 1/3rd, which means the cap has a voltage spec three times the working voltage. The more you derate solid tantalum, the less likely it is to fail. The closer the voltage spec of the tantalum is to your working voltage, the greater the risk of failure by voltage spike.

[@MyPinkFloydian]

Thanks For the Videos they really helped me out!

[@JDW-]

That's great to hear. Thanks for making time to let me know! Best wishes!

[@jerryking7502]

Are film capacitors ok for audio electronics?

[@JDW-]

Yes, but the answer is not so simple for Audio, but I will defer to this audio engineer's excellent explanation which may help you: kzbin.info/www/bejne/o3TOZmOIidmlipI

[@iamhe999]

Thank you.. well done..

[@andershenriksen6997]

Tantalum caps should be checked as well. I've replaced sooo many of them. They might look pefectly fine, but really they are shorted - as in sub zero ohms. And that will kill adjacent circuitry. Also, they should be rated at double the voltage of that in the actual circuit. When it comes to electrolytics and all you have is the C-meter in you multimeter, then measure the capacitance. If it reads much much above the nominal rating, then it's turning in to a short. If it reads much below, then it's drying out, losing capacitance, increasing ESR. If it reads well within the limits given in the datasheet, then it's problably ok. What you really want is to measure the capacitance, the ESR, and the EPR. However, if you're repairing 30+ years old equipment, then replace the whole lot. New caps probably have better specs over all, and finding 105 °C - 20,000 hours caps from well-known manufacturers isn't difficult or expensive. Edit: Just recently I had a shorted electrolytic in the rectifying circuit i.n a power supply - a Bang & Olufsen. The C-meter was very unhappy and dissatisfied because of this. The ohm-meter said 2 ohms. At the other extreme I had two 8 μF caps read out to be 25 μF.

[@JDW-]

If the SOLID tantalums (not Polymer Tantalums) are shorted, the user will unfortunately find that out if they accidentally powered on before checking them, as those caps would be a dead short. The reality is that most people don't check them outside the circuit if they don't notice any burn marks on them. Most people start the recapping process when the flip on the power and find their device does not work, then they jump onto forums or FaceBook groups or Reddit to ask for advice, and then they learn about recapping. As to the 50% voltage derating, yes, you are correct. I actually discussed that in my SE/30 motherboard practical example in the video. But again, that applies to SOLID tantalums. You only need to derate the Polymer tantalums by 20%. That's why you need a 25V solid tantalum on a 12V rail, but only need a 16V polymer tantalum on the same rail. Even so, OS-CON caps are cheaper than polymer tantalum caps and offer lower ESR as well, which often make them a better choice than either tantalum type. You just have to be careful when recapping switch mode power supplies, as some require a minimum ESR to be stable. And yes, I agree that AGE is a better predictor of bad caps than usage and heat. If you have fluid-filled caps that are 30 years old, axial or radial, they do need to be replaced, even if the device seems to work fine.

[@andershenriksen6997]

We totally agree 👍

[@misterbool7349]

Hello Sir, The following questions are about old audio devices. What do you think about replacing the old electrolytic capacitors on the power supply board with low ESR capacitors to increase their life? These capacitors seem to suffer the most. Also, I have read that it is possible to replace the low capacity electrolytic capacitors with PP or PS (low dielectric constant) film capacitors to make your device more reliable. What do you think about it? Thanks :)

[@JDW-]

Whether you use film capacitors or electrolytic capacitors or something else to replace the stock caps in a switching power supply, the main issue comes down to ESR. There are some power supplies that were designed to require a minimum amount of ESR on the output capacitors, and if you go below that the output could become unstable. But the only way to know that is if you have a schematic, and most of us don’t have schematics on the old power supplies that we are recapping. So then it could be a situation of trial and error, where after recapping you find the output is not entirely stable, you may need to swap one or more of those output capacitors with a capacitor type that has slightly higher ESR.

[@JDW-]

Jump to the 39 minute timestamp in my video below to see an example of what I mean. kzbin.info/www/bejne/h6DJXoqgn8Sljas

[@misterbool7349]

@@JDW- Thanks, now i understand that it's not that simple. And i also know now that i have to buy a scope 😁

[@maxheadrom3088]

The DE5000 has been reviewed by a bunch of people and I heard only good things about it - the best cost/benefit and a great benefit by itself.

[@wekeeptruckin]

Great video thank you! I wonder if you can help me make a decision on caps for my 1997 vehicle ECU. I am keeping this vehicle long term and am working on recapping the ECU for preventative maintenance. Thankfully the 3 original electrolytic caps have not leaked as others with the same make/year have already leaked and ruined the PCB. Anyways, the 3 caps as I said are classic electrolytic, looks like a mini AA battery. The specs are Nichicon, H9633, 105C, 25V, 220uF. Since it's for a car ECU located in the engine bay they should also be AEC-Q200 rated. Being that there's only 3, the cost isn't that big of a deal to me. I don't mind spending a few extra bucks for something I will never have to worry about for the rest of mt life. Can I replace those old electrolytic ones with solid polymer caps of the same voltage, temp and capacitance and expect it to work the same? Some say you have to get caps with like 1/2 the capacitance if switching to polymer from electrolytic because of lower ESR. I really want something that will far outlast the old ones which have lasted nearly 27 years. I know you said you aren't into the hybrid ones but so far it sounds like they are my best option and have better functionality compared to the classic electrolytic. I saw one that matches my dimensions and capacitance and has a higher temp rating, 145C at 2000H. Some say the lifetime doubles every 10C drop in operating temp. Would that mean these are rated at 32,000H at 105C? Is there a capacitor I could put in there to replace the old electrolytics and basically know I will never have to replace them again? The dimensions are ~8mm diameter and ~16mm tall or long and they are in a horizontal resting position, radial mount type. Please let me know what you think, thanks!

[@wekeeptruckin]

Maybe its better to ask about solid vs liquid capacitors. Can I switch to solid polymer or ceramic of the same voltage, capacitance, temp? I've found a couple solid polymers and 1 dipped ceramic (very expensive) but none of them are AEC-Q200 rated. Would it be better to get those or go with a liquid cap rated for AEC-Q200?

[@JDW-]

To eliminate need to swap them out again, you would need a solid electrolyte capacitor. “Hybrid Polymer” only better in terms of certain specifications relative to normal aluminum electrolytic capacitors. But the fact remains that Hybrids have a FLUID (not solid) electrolyte, and anything fluid will find its way out over time. Even if the fluid by some miracle never leaks, it can dry out. This is why the idea capacitor has a solid electrolyte which can never leak or dry out. And such a capacitor can be found on Mouser, matching your physical size criterion, as well as capacitance, voltage and high hour rating at 105°C: www.mouser.com/ProductDetail/Chemi-Con/APSG250ELL221MHB5S?qs=nxZbHzLpdve2AQv2batcYQ%3D%3D Of course, my proposed replacement does not have AEC-Q200 certification. Many capacitors you look for in recapping applications don’t. Admittedly, you are the first person who has come along wanting to recap an automotive ECU. I have worked 30 years in the vehicle aftermarket, and I can say that for our aftermarket products, we almost never use capacitors with AEC-Q200 stress test certification to save cost, and we’ve never had a capacitor failure. However, we deal in vehicle security systems, not the ECU. The ECU is a critical vehicle component, so it is prudent to have the best capacitors possible. AEC-Q200 guarantees you a capacitor which has been tested in a way that comes close to matching the harsh environment of a vehicle. With that said, fluid-filled aluminum electrolytic capacitors are not great capacitors to begin with. Their ESR rises a surprising amount at 0°C and lower, and longevity centers on their fluid electrolyte not leaking or drying up. These are things that cannot be resolved by AEC-Q200 stress testing. In comparison, the solid electrolyte capacitor I linked for you above is superior in many ways to the stock capacitors in your ECU. They are simply not stress tested. Also, when considering hour-rating, you need to also consider that capacitor manufacturers don’t guarantee anything beyond 20 years or so, regardless of what the math calculates with respect to the hour-rating and the average temperature in your environment. For example, you can find some high-hour rating aluminum electrolytic capacitors rated at 125°C, and the math may suggest they have a 40-year life, but in reality, that fluid inside them limits their life, and as I said, the make of those parts doesn’t guarantee their operation beyond 20 years or so. My only concern about swapping out with a very low ESR type like the capacitor I linked for you is that I have no idea how they are used in your ECU. If there is a switching power supply on the ECU board and if those capacitors are critical components in that power supply, it could be the power supply requires a minimum ESR for the capacitors used. Probably not in the case of 200µF capacitors, but I cannot say for sure. What happens if the switching power supply controller requires capacitors with a minimum ESR but you use something lower than the minimum? Well, the output of that power supply would very likely become noisy at best, completely unstable at worst. But sans having a schematic, the only way to know would be to swap them out and test. But because an ECU is a critical car component, the best approach would be to put an oscilloscope on the voltage output to ensure it remains stable after recap. Meaning, you would need to see the noise on the output voltage before and after the recap to compare. And I’m guessing that you may not have a scope. This is the best advice I can give you, but I would strongly suggest you visit the EEVBlog forum and put your question to the engineers there because some will surely have automotive ECU design experience and can give you more experienced advice. Keep in mind though that recapping is not embraced by everyone. Some people think it’s a crazy idea. It is in some cases but not in others, so don’t be discouraged by people who claim you never need to recap anything. There are people like that. But it could be that posting in the EEVBlog forum will garner you more specific advice, especially in regard to AEC-Q200 certification for replacement capacitors you intend to use in an important vehicle component like an ECU.

[@wekeeptruckin]

@@JDW- Wow, thank you for taking so much time to help me! I am very interested in that polymer option you found, even though it doesn’t have the AEC-Q200 certification. I compared the stress test data from the datasheet with a liquid AEC-Q200 cap and they are fairly similar. The ECU I am working on is what I might call poorly designed for future maintenance. The circuit board isn’t even a circuit board, it’s like a giant flexible…ribbon cable? Which is adhered to a sheet of aluminum and bent over like a book. It was very hard to get open and has some minor damage just from that so I really don’t want to have to do it again. Most ECU’s are a typical metal box with a solid PCB screwed inside it. If it were a lot easier to open up I would be less concerned about putting liquid caps back in because I could just recap it every 15 years or so. Anyways. I am not very experienced with electronics, I’m your average diy. So I don’t know about the possibility of it having a switching power supply. (I also don’t have a scope or the knowledge of testing with one.) I wonder if I could get you a picture of the ECU where the 3 caps are located if you could tell me if it looks like it has what you were talking about or not? It may not be the best comparison, but there is a guy on youtube who replaced liquid caps with solid polymer in his ECU 4 years ago it seems he has had no problems. His was from a different make/year than mine though. If I do replace mine with the solid polymer I will be making some extra precautions for heat and humidity although I doubt there will be an issue. I couldn’t find a schematic of the ecu but I found a connector pinout which mentioned the voltage to the ecu is only 5v. I’m not sure if that tells us anything or not. I will likely check out the EEVblog as well. It will be a couple months before I do the recap but I’m getting my homework done ahead of time. Please let me know if a picture of the PCB would help decide if solid polymers will be a good idea or not Thanks again!

[@JDW-]

@wekeeptruckin it could be that only 5V enters your ECU but there may be a 5V to 3.3V step down converter on it. You could take a photo and post it to an online sharing service and try to post a link to that here, but it would need to be high enough resolution to see the part numbers on the chips to know if it’s a switching power supply. The entire point here is determine if there is indeed a switching power supply on the ECU and if so, what is the main controller chip part number, and then Google that part number so as to find a datasheet l to see what the input and output capacitor requirements are with regard to ESR (resistance of the capacitors). If there is no minimum ESR, then you have nothing to worry about when using very low ESR polymer capacitors.

[@wekeeptruckin]

@@JDW- Ok, I will look into the photo thing. I am a bit concerned about the humidity factor causing premature wear on the polymers. Do those need to breathe or could I coat them in silicone caulking or conformal coating to seal them off from humidity?

[@mlbabineaux]

Critical care medical devices use United Chemi-Con / Nippon Chemi-Con

[@donaldlemoine3032]

I have a question.. I am always tinkering with old motherboards and video card. Just about all of them have the aluminum electrolytic Capacitors. How do determine which solid to use in place of the aluminum electrolytic Capacitors?

[@JDW-]

And by "solid" you of course mean a capacitor with a solid (not fluid) electrolyte, such as a solid tantalum capacitor, polymer tantalum, or non-hybrid OS-CON. Film caps also are "solid" but those would not be picks for a motherboard or graphics card. In my video I discuss the voltage derating for solid tantalum caps. Polymer tantalums are a bit more flexible on that, but they are more expensive than Panasonic OS-CON caps. So in the example where I recapped a Macintosh SE/30 motherboard, you can choose solid tantalum at 25V, since the motherboard sees a 12V rail, or you can choose 16V polymer tantalum, albeit at a very high cost, or you could choose 16V OS-CON caps, which also have the benefit of looking stock and offering the lowest ESR. In a recap of a motherboard or video card, low ESR is good, so OS-CON caps may be the best pick of all among the 3 I just mentioned. Just keep in mind that OS-CON caps don't come in all the same sizes as aluminum electrolytic caps, and you need to keep physical size in mind too, measuring how much physical space on the PCB you have for the replacement cap, not to mention the lead spacing.

[@Wallygjs]

Thanks for explaining that. One thing I was hoping was that you might explain if there are any differences in behaviour of these capacitor types, in other words are they all exactly the same assuming you have the correct Capacitance and the voltage rating is sufficiently high enough. Being a newb to this sort of stuff I wonder if there are certain uses on a motherboard where you would prefer an electrolytic to a tantalum or a ceramic etc. I.e. why do we have such a range of different capacitors that on paper are doing the same thing?

[@JDW-]

ESR can affect behavior in the circuit. ESR is the resistivity of the capacitor. Use a very low ESR cap in a switching power supply, and you might get unwanted oscillations. I demonstrated that in my recap of the SONY PSU inside the Apple HDSC external hard drive enclosure. But at other times when the capacitor is used as a "bypass" or "decoupling" cap to prevent sudden drops in voltage, low ESR is desirable and a big benefit. The big disadvantage of fluid-filled aluminum electrolytic capacitors is not only their high ESR but the fact that ESR suddenly gets much higher at very low temperatures close to freezing or below. All capacitor types have their advantages and disadvantages though. You might find the comparison chart on page 5 of the following PDF useful: www.analog.com/media/en/training-seminars/tutorials/MT-101.pdf

[@Wallygjs]

@@JDW- Thanks for the information. I will have a look at the PDF. I suppose what I was getting at is, if you were doing one circuit would you prefer a tantalum over an electrolytic because of some feature of the way they work or is there absolutely no difference between them other than the things you mention above. I just wondered why there are so many different types of capacitor. I always thought there was some rule in electronics that says for this type of circuit you use a tantalum and for this other circuit you use an electrolytic. It would appear they all work the same, its just around how long they last and how they behave when over volted etc.

[@Wallygjs]

@@JDW- I had a look at the sheet and I can see from that the ceramics are good for High frequency circuits and the electrolytics are better for lower frequencies. I read a few pages and my head started to hurt but I get the gist!

[@JDW-]

@@Wallygjs The answer to your question is neither short nor simple because the different capacitor types are not all the same. For example, if you need a 3300µF 16V capacitor for a particular application, regardless of voltage, your only choice is a fluid filled electrolytic capacitor. Such high capacitance values, at high voltage ratings, are not available in other capacitor types. Here’s another example. Let’s say you’re looking for a capacitor with very low ESR. You can't do better than a ceramic capacitor. But let's ponder why NOT to use a ceramic (MLCC). First, the physical size of ceramic capacitors increase and decrease (changes so small it's imperceptible to the human eye) as applied voltage increases and decreases. It sounds pretty crazy, but it’s true. So while ceramic capacitors are very often chosen for the output of switching power supplies due to their extremely low ESR, they can also make a lot of noise because they vibrate the entire PCB, which amplifies the sound like a speaker. If you've ever heard a constant buzz when using a product that has a switching power supply, ceramic capacitors are the reason! There are some expensive specialized ceramic capacitors that are made to eliminate that vibration on the PCB, and sometimes you can design a PCB with holes cut around the capacitor itself to attenuate the vibrations of the PCB itself -- but for recapping that is not practical. Yet another big problem with ceramic capacitors is that they have an even more strict voltage Derating than solid tantalum capacitors. Why? Because the Capacitance of ceramic capacitors decreases according to the voltage (DC bias) you apply to them. That's right... The higher the voltage across a ceramic, the lower its capacitance! Sometimes capacitance can decrease as much as 80%. Imagine a situation where you want to install a 1µF capacitor in a circuit that sees 10Vdc. You might need a 100v rated ceramic capacitor so as to maintain a stable 1µF of capacitance. The downsides to higher voltage rating for ceramic capacitors are Cost and Physical Size of the capacitor. Even if you have enough circuit board space to accommodate a physically larger ceramic capacitor, and even if money is no object so the higher cost doesn't matter, you still may NOT wish to install a physically large ceramic capacitor because ceramic capacitors are ceramic and therefore brittle. Depending on the size of the circuit board or product, a fall to a hard surface from the height of even a couple feet could crack large ceramic capacitors. So as you can see, it’s important to weigh the pluses and minuses of each capacitor type before choosing it. And while a degree in electrical engineering or years of experience as an electronics technician can be extremely helpful, the product data sheet is a critically important resource that can help you make well educated decisions about whether or not to use a particular capacitor tied for your application. Datasheets pack an enormous amount of information that can overwhelm you at first, but time and experience can help you sift through it. Sometimes you cannot do that alone though, so you need to ask informed people like myself or become a member of a popular online electronics forum to ask people what a given datasheet spec means. I hope this helps!

[@Wallygjs]

@@JDW- Wow, thanks for your long and detailed reply. I am definitely getting a lot more understanding of Capacitors and the minefield of information around them. At the moment I am usually just replacing old capacitors on Vintage Macs and Amigas so I don't need to make decisions about which caps to use. In time though, I hope to do some night school or online training on electronics as I would like to build some circuits of my own such as accelerator cards or ram upgrades using modern chips. Ultimately I would like to come up with a drop in replacement TFT screen for Classic Macs and create a circuit board that allows you to just replace the analogue board, as increasingly the old tubes are dying as are the availability of the flyback transformers. Personally I would like to have a less dangerous classic Mac, as I have never been a fan of the dangerous voltages inside classic Macs. So thanks very much for taking the time to help me understand!

[@chironbramberger]

This is great! Thanks!

[@xav500011]

A lot of old retro computer mainboards are be very delicate. SMD capacitor pads will easily peel off which means a hot air gun will have to be used instead of a soldering iron. Which for novice hobbyists such as myself makes it more difficult. I have damaged a Commodore Amiga A1200 mainboard because of delicate SMD capacitor pads.

[@JDW-]

Hot tweezers are yet another option for those who either lack a hot air station or inexperienced with it. I personally have access to both tools and find them each to be beneficial at times.

[@paulklasmann1218]

Id never use hot air to remove SMD capacitors. Its a sledgehammer to crack a nut. Desoldering tweezers are what should be used. Alternatively a good soldering iron can be used. Theres no substitute for experience so plenty of practice on scrap PCBs first if you're a novice.

[@briansmith9698]

JDW I jhave old tube radio with a multi section capacitor color coded yellow 400 volts 30 mfd red 250 volts 30mfd blue 630 volt 20 mfd 20 mfd green What do you recommend replace it with? If I can I want put a can style capacitor in it would do you recommend as far as volgage and micofarad?

[@JDW-]

Brian, old tube radios are outside my area of experience. I've never owned one. Basically, it's a matter of properly identifying what is a capacitor, and what is a resistor, and not get them mixed up. If you are sure the values cited are Micro-Farad and not Milli-Farad, because you seem to know the voltage specification, you can rather easily do a keyword search for appropriate replacement "Electrolytic" capacitors using the capacitance and voltage values you told me, from places like Mouser or Digikey. Caps the size you mention should be polarized, although I am not 100% sure about that. You also need to measure the Lead Spacing so your replacement capacitor has the same spacing; otherwise the cap would stand up on its legs, and you don't want that to happen when installing physically large capacitors. The main issue would be to ensure that you know the Polarity. If the board isn't polarity marked and you can't see that on the stock caps either, you can often find schematics for old radios which would tell you that. A great resource about all that is the following URL: antiqueradio.org/recap.htm

[@ExPsy]

Tantalum capacitors need replacement when used in circuits prone to voltage overshoots (like cpu/gpu vrms) where the psu used might not be able to handle the load. I started seeing equipment dying cause the board engineers did not expect the psu or ac adapter used with their boards to have so significant overshoots.

[@JDW-]

With regard to solid tantalum capacitors and overvoltage, you are absolutely correct. However, the average electronics enthusiast would not necessarily know enough details about their old hardware to make an informed decision about whether or not to replace all of the existing solid tantalum capacitors. And there are a lot of cases where there is no such overvoltage and therefore you would not need to replace them at all. Also, if you measure and find there is overvoltage than you know you would need to replace those capacitors, but there are also situations where you may only get an occasional spike, and that is what I mean by saying it is harder for the average electronics enthusiast to figure out if those capacitors need to be replaced or not. The key point here is in determining if the solid tantalum capacitors in question are in fact voltage derated by 50%. In other words, a solid tantalum capacitor rated for 16 V should not be seeing more than 8 V normally and certainly not more than 16 ever.

[@stormriderkaos]

Leaded solder flows and sticks better. (?) True or Not?

[@JDW-]

Leaded solder flows better, yes. "Sticks better" depends on flux and technique. Why would Lead Solder flow better than Lead-free? Lead solder has a lower melting point, which really does make a difference. Lead solder is also a bit more shiny and mirror like than lead-free, and I prefer that look. Lead solder has no "tin whiskers" problem like lead-free solder does. I always use lead solder and never use lead-free.

[@multicyclist]

Just want to point out when replacing electrolytic capacitors, that too many people singularly fixate on 105C as better vs 85C. 105c does not mean longer life than 85C, just higher temperature when that spec is needed. In most cased where 85C is never even approached, it does not matter. What does matter that seemly everyone ignores is design life. Such as an old 85C 10,000-hour or even 18,000-hour capacitor they replace with a 105C rated capacitor with only a 2,000-hour rating thinking it is better. Many 105C 2,000 (or even 1,000) hour rated capacitor are often cheaper than a better quality 85C cap. Those low hour caps are meant for cheap disposable devices, not restorations that are expected to last for decades to come. Those vintage re-cap kits often sold on eBay boast of being a trusted manufacture 105C rated caps, but are actually the cheapest (low hour) series they can find, turning a restoration into waste of time and a future piece of crap.

[@JDW-]

Yes, you absolutely must consider the temperature rating with the hour rating. That "combination" is what matters. But there are times when you search for a replacement cap of a given capacitance value and physical size, only to find the best rated at a mere 2000 hours. And in that case, you need to go with what you can buy (for hobby use). If the voltage rating can be accommodated along with physical size and lead spacing, sometimes swapping out a small electrolytic for a dipped solid tantalum is a great choice. You simply need to make absolutely sure the tantalum's rated voltage is no less than twice the working voltage it will see. Do that, and you won't need to bother with a replacement in the future. Interestingly, I've also seen 85°C caps with lower ESR than their 105°C cap counterparts at times. I also see that when it comes to hour rating. Seems like many high hour rated caps have marginally higher ESR than lower hour rated caps. I'm talking 10,000 hour rated caps versus something like 2,000 hours.

[@multicyclist]

Yes, all valid points. I guess the point I was attempting to make is the spec I am least concerned about is the 85C vs 105C unless a 105C cap is required. Given the dimensions are correct, I look for hours lifetime, ESR, ripple current first and temp rating last as the least significant. Cost is also a factor in a selection but do not want to negatively impact longevity by using low hour caps. Tantalum capacitors often fail to a short when they do fail so I avoid using them. In my old Tektronics scopes the tantalum capacitors seem to be failing before the electrolytic. Case in point, I am rebuilding a Soundcraftsmen PCR800. There is no possible way the power supply filter caps would ever get near 85C (185F). My replacement is a Kemet brand with 16,000 hour rating and over 3 times the ripple current rating and lower ESR of the only Nichicon selection available, which is a 105c 5,000 hour cap with 1/3 the ripple current. And since these amps already have too small of power supply cap j(due to space constraints) s for their output, the 3X boost in ripple current should improve IM distortion during high current demands.

[@JDW-]

@@multicyclist There is much debate about SOLID tantalum capacitors, and I must emphasize "SOLID" because there is less debate about "POLYMER" tantalums, although most people tend to be somewhat ignorant about all the tantalum types (including WET). With SOLID tantalum, you absolutely positively must derate the voltage by 50%. This isn't my personal opinion. This is standard practice and is what the manufacturer's specify too. In some government use-cases they may even derate more. But solid tantalum capacitors are still widely used today specifically because they tend to avoid catastrophic failures when properly derated. It's voltage spikes that kill them. I've seen equipment from Apple and other big companies that have failed/burned tantalum caps only because their engineers decided to use a ridiculous 16V-rated solid SMD tantalum cap on a 12V rail. So even some experienced engineers make that huge Voltage Derating oversight. Sometimes those 12V rails can hit 12.6v. And all it takes is a brief bit of over voltage to exceed the 16V spec on that improperly chosen solid tantalum cap to light it afire. And yes, solid tantalums fail shorted most of the time. Had Apple engineers instead used a 25V rated SOLID tantalum, it would have been physically larger, but it likely would have survived and still be in use today on that nominal 12V rail. In contrast, POLYMER tantalums do NOT fail shorted and you only need to derate them by 20%. (Some say only 10% for low voltages under 10V, but I always derate them at 20%.) The problem with POLYMER tantalum caps is that they are very expensive. So in my SMD recapping projects (on vintage motherboards, for example), I tend to use lower cost Panasonic OS-CON instead, since those don't need a voltage derating and I get even lower ESR with OS-CON than Polymer Tantalum. (Polymer tantalum is lower ESR than SOLID tantalum, and solid tantalum is lower and more stable than the ESR of Al. Electrolytics.)

[@multicyclist]

@@JDW- Very informative! Thanks!

[@tjules7]

I took my NAD 7400 receiver for repair, but the guy is having a hard time finding transitors/capacitors repalcement. Can you suggest a place where to buy those components?

[@JDW-]

I'm not an audio guy, so I had to Google your part number to see it is a AM/FM Stereo Receiver. I have no schematics or capacitor lists for that device. The best I can do is suggest that you identify all the electrolytic capacitors that need replacing, then note all aspects of their physical size, including Lead Spacing, then order replacement capacitors from Mouser or DigiKey. Get the same capacitance, and get a cap rated for the same voltage or higher. That's pretty much what my video is about, so if you require more detail about using the Mouser website, then I would refer you to my video about that. Best wishes.

[@JDW-]

I see you also ask about TRANSISTORS. That is beyond the scope of my video, but you can buy those parts at Mouser and DigiKey as well.

[@tjules7]

@@JDW- Thank you so much for your quick response. I will definitely check them out.

[@andrewbarnett84]

Your selection process is in general OK, however there are three more variables that need to be taken into account. Temperature of operation, there are two types generally, 85 degrees C, and 105 Degrees C. You can fit 105 degrees to either but not the other way around. ESR, if ESR is mentioned it normally means a LOW ESR (Equivalent Series Resistance), Lower is better, used for particular power supplies. To speed selection on Mouser, DigiKey, RS, Element14, etc. Select 'in stock', first. RoHS compliant is not required for old electronics, and in fact its hard to find on that is not anymore. Old Electronics will mostly be using lead solder anyway, so using a RoHS parts is OK but will not make the old electronics any better.

[@JDW-]

Andrew, while I certainly appreciate your insights, lower ESR is not always better, especially when it comes the the output of switching power supplies. Some require damping when using ceramics, such that you have a super low ESR ceramic in series with a resistor. Yet other switchers don’t need damping and call for the lowest ESR caps possible to yield the lowest peak-to-peak ripple. Because of this, choosing ultra-low ESR caps when recapping a power supply can have unexpected consequences, and I discovered that when recapping the Apple HD20SC Sony power supply. I had to use a slightly higher ESR cap in one location to prevent an odd “chirp” sound “oscillation” at power-on. I have a video on that if you want to see the “chirp” waveform on my oscilloscope.

[@andrewbarnett84]

@@JDW- Hi, thanks for the reply. I would not replace ceramics unless they are damaged, and then try to be as exact to the manufactures original as possible. I was talking about Wet aluminium types. I've been an electronics technician for over 45 years, and have replaced millions of capacitors. In the RNZAF for example, I repaired AFCS on A4's AND P3's, the 99% failing item on them was pitch sealed tantalum capacitors, I personally replaced in excess of 100,000 in around 4 years. They were soldered on silver plated pegs embedded in Teflon buttons pressed into Aluminum trays for mechanical stability as the aircraft could pull many "G' in multiple directions. "cut, grab, rotate off, lift, rotate, bend new loop, solder in-place, form other end solder, cut excess" next. Next I looked after (field service), IBM 3270/3276 terminals and other brands like TELEX, Memorex, Raytheon, etc in government and Airline offices. They all had Nasty switch mode supplies, which were rated for 100-230 Volts AC input 50-60 HZ, our mains is 240 V at 50Hz, and most 'departments neglected regular vacuuming, so the PSU's over heated and died. During 25 years of the field service, traveling to all cities in New Zealand, all airports, and the pacific Islands of Tonga, Samoa, Amerika Samoa, Cook Islands, and Fiji, I carried bags of 1000 capacitors of each type, as well as replacement supplies. I would replace the supplies on arrival, do any cleaning, adjustments, repair printers as needed, and retire to a motel where I would replace the capacitors for the next stop. Later again I worked on robotic and CNC machines, the only difference really was the size of the failing capacitors they were much bigger. I worked on pick-and-place machines, mostly Japanese, the problems on those were NOT capacitors in the main but lack of cleaning of the mechanical and pneumatic systems. I've repaired Ham Radio transceivers, the capacitors in those have generally a long life, like 20-40 years, but people like me like the older radios so I'm replacing capacitors in 1970's and 80's machines. On my work bench is an HP1710B oscilloscope, it failed 10 years ago, no less than 12 companies have tried to repair it, and failed, which is why I obtained it for free. Its problem I have found is many differing mechanical faults and transistor failure, the capacitors chosen by the HP engineers are still in tolerance. And just for a completely weird one, I worked on a knitting machine that made jerseys, failures were caused by a combination of capacitors (30 yo), and broken soldered joints due to vibration. That was a fun one to repair as the instructions are in Italian, the owner a sheep farmer, and his helper a handyman. He would come in smelling of lanolin and sheep dip, to help me remove damaged power supplies in the bowels of the machines. Regarding de-soldering, especially on fragile PCB's, I use Goot brand, "gootwick" desolder braid, 2.0mm most commonly, and 'Chipquik SMD291 no clean flux, and Chipquik low temp desolder metal, which makes removal of small parts and stubborn parts easy. You must remember to remove the low temp metal before soldering new parts on though. Always clean with Isopropyl alcohol pads or the liquid and a q' tip.

[@pibbles-a-plenty1105]

Harbor Freight sells digital calipers for much less than a Mitutoyo costs and are of equivalent quality. Certainly good enough for gauging capacitor bodies.

[@JDW-]

I can't speak too much about the game of "good enough" because I only tried the one Chinese made digital caliper and found it was a fail, then I purchased the precision Mitutoyo and found it to be ideal, and I rely on it for accurate measurements. But if you buy a cheap pair of calipers and find it lets you measure accurately enough to buy capacitors to your size requirements, then of course that is all you need. Because I live in Japan, I have never used Harbor Freight, which makes it all the more interesting that I get daily SPAM from a source purporting to be Harbor Freight. Not sure why there's so much SPAM pretending to be them, but if it happens to me, who has never purchased from them before, customers of the real Harbor Freight should be careful about what emails arrive in their INBOX.

[@mansnilsson4382]

RoHS is a no-brainer. Also, in Europe, it is a requirement. Lead-Tin solder is forbidden for new production in the EU, except in avionics and space. Professionals may use it for repair. I use it, as amateur, for repairs on equipment with leaded solder since I have rolls of lead solder that will last my career. But all my new production is RoHS exclusively. Remember Flint, Mi.

[@wildman1978101]

What does the gross negligence and incompetence in water chemistry in Flint have to do with any of this?

[@TimPerfetto]

@@wildman1978101 He is clearly unaware how frustrating lead free solder is -- I suspect he also needs to resist the urge to eat leaded solder otherwise idk wtf the issue is either WTF STOP EATING HAIR

[@ramencurry6672]

Agreed. Planet 🌎 comes first.

[@texacomann]

Thank you for this very detailed Video! This helps me a lot! It is claimed that rohs does not affect the quality for the worse. But that is very questionable and certainly cannot be answered in general. So I'm rather conservative and also like to go without rohs, because most of the time it stupidly costs even more money. And we all have to work for our money. Inflation and the scarcity of components also increases the prices and restict your life. Considering environmental awareness and sustainability, mankind should be more concerned about maintaining peace on this beautiful earth than fooling around with nuclear weapons. Nuclear weapons must not fall into human hands, otherwise all environmental protection concerns are senseless pomposity, because nuclear weapons destroy the environment with 100 percent guarantee. On the other hand, climate change is more of a term that is misused for many political goals and should be viewed more critically. Nuclear weapons, on the other hand, never fail to destroy their environment 100 percent and can easily wipe out all life on earth. That's why I'm not interested in rohs to be honest. If I could do something against nuclear weapons with my actions, I would like to put a tick there if there was a checkbox for it, but with rohs I can at most buy myself a better conscience, and even this is unfortunately not proven whether it will do anything.

[@jayytee8062]

Hi, i have a 1Kp 50volt (+/_20%) small cap that looks identical in size and appearance to a resistor. It's from a williams pinball machine 1991 cpu board. What is a modern equivalent replacement for this please?

[@JDW-]

What is "Kp"? Can you post a link to a photo?

[@jayytee8062]

@@JDW- 1Kp= 1 kilo pico farads = 1000 pico farads. Or 0.001uf I'm not sure how to post a photo. But as i mentioned it is identical to a resistor with the color code stripes. Some people commonly mistake them for resistors.

[@JDW-]

Thank you for explaining. I am unfamiliar with pinball machine circuitry, nor do I know if it is in the power supply or something else. If you cannot provide a link to a photo of that capacitor, perhaps you can determine if it is ceramic, electrolytic, film, mylar, tantalum, etc? By knowing the type of capacitor it is and how it is used in your circuit, one can then more intelligently choose a suitable replacement. Picking a completely different capacitor type, even if the capacitance and voltage rating are the same, isn't the best practice, hence my question to you today.

[@tenminutetokyo2643]

Console5 is awesome.

[@JDW-]

Console5 offers very low cost kits, but it's important to keep in mind that they don't have kits for everything, nor can you buy any capacitor you like from them individually. That is why there is still reason to purchase capacitors from places like Mouser or DigiKey.

[@tenminutetokyo2643]

@@JDW- Agreed. I bought their Atari 2600 Jr. Refresh kit and it worked great. That was how I discovered the tiny square film caps. But I added some more new parts from Mouser too like a TRS-1 2450 and new chips. I'm going to try their Colecovision kit next.

[@Doonit_hard_way_since_65]

I would have to disagree on your analysis of AC line filter caps aka "Rifa caps"/ Metalized film caps on the AC line should be an automatic replace in most older gear. Affectionately known as "the cap of death" in older audio gear, a fail to short puts AC line on the chassis. Those European made caps from the mid 70's-on aged terribly, catastrophic failures (aka "capocalypse") are more common than one would want. If you are lucky they just short, smoke, and quit, if not... well, they were the reason we now have X and Y specifications for safety caps. That old junk never had polarized cords, which further increased the chances of getting a surprise from an electrified chassis. Good video, appreciated much.

[@JDW-]

Although I was giving general advice, I largely come from the standpoint of a vintage Apple Macintosh enthusiast. And I know for a fact how those RIFA capacitors fail on those machines. Most of the time they turn into black goo, and it’s not a situation you have AC constantly being shorted. The worst case of them failing in a Mac means you remove the gooey mess and swap it out with your preferred replacement cap. Some people just remove them and leave them removed all together, which will allow the Mac to keep working, but I don’t really recommend that because the caps were there for good reason. You can replace them with another metallized paper RIFA cap, or a ceramic safety capacitor if you prefer.

[@phillipweaver]

can you go up in voltage. I dont mean by a lot

[@anullhandle]

@phillipweaver in general its ok to go up in voltage rating. A modern replacement might even be the same size for a higher voltage rating. A 35 volt replacement for a 6.3 volt is not a lot for an electrolytic for example.

[@JDW-]

Yes, you can choose a replacement capacitor with a higher voltage rating, either a little bit or a lot. But keep in mind the capacitor with a higher rating may be physically larger than the original capacitor and therefore may not fit. So you need to measure the lead spacing and measure what the maximum diameter of your replacement capacitor can be before you purchase that replacement.