EEVblog

[@hanneshauptmann]

You know what ? I learn more from you than from my engineering teachers all together .... thank you so much !

[@haxensalat]

lol yes

[@voiceforjusticeandproporti5543]

Yep

[@victorchorques4893]

Totally agree

[@fireracerworkshop8251]

I also agree

[@mohammadhushki96]

That's very true! Me too

[@Willam_J]

Dave, I've been an EE for 28 years, but I'm not ashamed to admit that I learn something from almost every one of your videos. I'm already familiar with most of the material, but you always slip something in that benefits me. I really appreciate what you do. After all these years, I like to think I know everything, but your videos keep me humble. Thanks!

[@drillsargentadog]

smashing video man! As an aerospace guy who's interested in getting more into electronics, these videos are a valuable practical resource I can use to help me understand real designs. I also like how you talk about the basic physics reasons for some circuit designs. My original background is in physics and I appreciate engineers who argue from the physics rather than vague and possibly unreliable rules of thumb. Keep up the fantastic work!

[@EEVblog]

drillsargentadog Thanks, glad you liked it.

[@Psycho4Ever666]

+EEVblog Yeah a really great video. :) I have to watch it again, did you mention that when we connect smaller capacitors in parallel instead of using one, we get a statistical chance that some negative deviation compared with the nominal value of the capacitors is compensated by a positive deviation of other capacitors or the other way around, depending on how you look at it. :D And in the worst case we just have the same maximum tolerance, assuming that the big capacitor has the same tolerance that the smaller ones have. Greetings from germany. :)

[@CaptainDangeax]

Good day. Sorry to dig this old video but I need to clarify. I mainly use veroboards because electronics is just a hobby, and I'm French. If I understand your video well, the best way is putting a capacitor close to the output of the power supply (7805 or buck), and a capacitor close to each input of each circuit. Then, if possible, better have a star topology (with the power supply in the center). Did I understand well ?

[@foxyrollouts]

Hes a great lecturer

[@phillyphakename1255]

​@@Psycho4Ever666biggest issue I have with that argument is that there's relatively little deviation between components withing the same batch, but a lot of difference between batches. And if you are manufacturing from the same reel, you probably aren't mixing batches in order to get that statistical regression to the mean.

[@JosipMiller]

These electronics lectures from Dave are the best part of this channel. Theory and practical experience fused in excellent educational presentations.

[@shriramspark]

Hi Dave, greetings from India. I'm currently in my third year of electronics engineering and I have been an ardent follower of you and your video blog for more than 2 years now. I consider you my electronics GURU [A great teacher] and your clear cut explanation, in depth skill in whatever you do have amazed me to the core. I now aspire to become a person like you. I have a lot to say about you, and your blog. I have got a lot of appreciation and accolades for many of my projects for which i have looked up to your videos for its designing, soldering and many other aspects. You are my electronics GOD. Thanks for putting up all your videos. I wish almighty to shower you and your family with loads of blessings and goodness. Hope you read this. If in case you read this, please try replying.*****

[@EEVblog]

Shriram S Thanks, appreciated.

[@calvin-7540]

I spent 2 years studying in Polytechnic taking Aerospace Electronics, but you are the only one that taught me more than the book describesd.

[@calvin-7540]

I have no valid reason not to Subscribe your channel now

[@calvin-7540]

Solid video

[@MrDoneboy]

Many thanks Dave, for the practical knowledge that you teach, which can only be imparted by a true electronics professional like yourself!

[@lloydrmc]

I appreciate the consistent rigor of your real world testing procedures.

[@oldiron1223]

Dave; Go get your PhD and teach. We need thousands more like you to make engineering fun for young people.

[@suprememasteroftheuniverse]

PhD is useless. What do you mean?

[@d95mback]

He is teaching.

[@stanburton6224]

Why would he waste time and money getting a piece of paper that is totally useless? He OBVIOUSLY has every qualification needed to teach the subject, probably MORE than most WITH a PhD, since he has actually been there and done that, instead of wasting his time fucking around in academia.

[@kevinhevans]

To the comment replies: I think what OP means is many undergrad professors are of awful quality and Dave's teaching ability far exceeds many in academia. But in order to teach in universities, you'd need to have a PhD.

[@stanburton6224]

@@kevinhevans only because academia rules require it. They wish to maintain thier monopoly. Just like Apple refusing to allow third parties to fix their POS's. The PhD in reality is a worthless piece of paper.

[@damnedattendant2536]

I am getting addicted to electronics more and more everyday thanks to your videos ... Thank you sir .. Lot of respect ..

[@briand2614]

Again, another great video. I’m a DIY hack and have managed to repair a lot of my own stuff by replacing capacitors. Keurig coffee machine, Kitchenaid refrigerator, Ryobi tool battery charger, Thermador range, and I’m currently repairing an 85 year old Philco tube radio full of bad caps. So yes your comments about longevity are bang on. Trying to imagine how many appliances and electronics have ended up in the dump due to an inexpensive cap failure is like grasping the size of the universe.

[@MrZetor]

10. Accuracy. You will get a _much_ more accurate (total) capacitance value by using several caps in parallel, since their values follow the Gaussian distribution.

[@davidjackson2115]

Thankyou Ocker. Im working on a power system, but rechange my ideas as when you near any completion "hindsight" kicks in so I go down that path. Lately the idea of multiple caps came to mind as I want to get good power from fast switching and wasnt sure if big cap would charge up fully fast enough, I understand theres a time restraint but gather its more related to resistance and voltage. You piece is exactly what Ive been looking for - its a real gold mine. This opens my mind to building circuits as Im old school {used to f around with the old Holdens} and hate the "throw away society" when you buy inverters etc... and it goes bumg and you dont know what to do so throw it out and get another - Id rather build my own so they last and anything goes wrong you know what to fix straight away. Long life is what I want as dont want to come to depend on something then have it go bung in the middle of the night. And i didnt know you could get caps on a roll like that. No wonder they can make such wonderful electronics so small and so cheap that do work well. But the crafties dont put ripple protection in them so you keep blowing them. Thanks for your good work.

[@astrogirl1usa]

I love Fundamentals Friday, thanks Dave!

[@Belginator]

I have been in electronics for 30 years and you taught me a a lot

[@herbertsusmann986]

This is the kind of practical design stuff they don't teach you in school. Thumbs up for talking about this! This is why new engineers straight from school are often such newbs.

[@ATinyGreenCell]

Your fundamental fridays videos, heck ALL of your videos, should be on a must-watch list for any budding biohacker interested in making their own laboratory hardware. I've learned more from you than most of my intro electronics classes at uni. If I'm ever in Sydney please let me buy you a pint! The least I can do!

[@PhilippeVerdy]

What is missing is why we use an electrolytic capacitor in parallel with a non-electrolytic (ceramic) one. The video does not focus on the reactance of the buitin self current: electrolytic capacitors have high reactance and this plays an influence on high-frequency peaks of currents (notably those coased by digital gates causing noise currents). The response in fact was only studied at a single frequency forgtetting what happens at noisy high-frequency peaks). That's why we find huge electronic capacitors in parallel with microscopic non-eletrolytic capacitors. The latter will not have enough charge to sustain continuous loads, but will react much better against those noises to absorb them (they will be recharged very fast by the charge of the neighbouring big capacitors. A video could compare the response using a large range of frequencies. Notably by using complex numeric charges like a processor running around 1GHz, and to show the effects of the noise starting signicantly around 3GHz, and how a single ceramic capacitor solves the problem compared to the situation where it is missing and all the stabilisation work is done by the electrolytic capacitor. Start by running your processor at low frequency, then increase it (still in its operational range) and you'll see how the power voltage is being affected by the increasing noise (independantly of the frequency of your power source which is typically 50 or 60Hz. The same is true when instead of a AC-DC power converter source you use batteries : the (electrolytic) batteries react also like capacitors and don't like much the jigh frequency noise: adding a non-electrolytic capacitor in parallel allow them to behave better (and also with less heat dissipated by them, those batteries will also have longer life and will be usable for longer time before being too much discharged.

[@martinparnell8990]

Thanks for these videos. I'm approaching the end of my first year in seriously learning electronics for the purpose of designing and making guitar effects pedals, n similar. Your videos have helped me so so much. You are a great teacher.

[@JohnSmith-he5xg]

Great video. Love learning these practical design concerns like taking temperature dissipation over the greater surface area of multiple caps vs a single larger cap

[@Squat5000]

Great video. I wound up using many of these reasons while designing my induction heater. CAP life is difficult to deal with at 75khz and 600amp peak current. Pulsed, 25% duty cycle extends life and reduces heat. Still caught on fire. So I made a modular cap bank that slides in and bolts on. :-P

[@islandhopperstuart]

Great video Dave! A little more about the radiation equation at the end. Specifically, the two temperatures - can and background - are absolute temperatures in K (Kelvin) and not degrees C or F. With a room/background temperature of typically 293K, and the capacitor can higher, the fourth powers of both temperatures are pretty high numbers, and the difference between them exhibits rapid divergence as heat loss through radiation begins to predominate with increasing can temperatures. All good fun stuff.

[@k7adi]

Very nice video, Dave! One thing that was not mentioned (or probably was mentioned but I missed it) was that with more caps you get more self-protection. If one of the caps fails, by having the ESR shooting up, the remaining caps will "protect" it as they will pick up most of the "ESR load". This also will maximize the remaining life (whatever is left) of the faulty capacitor as its "ESR load" will be decreased. (In fact it's more complicated - the concept of "ESR load" is a bit inaccurate) This is in contrast with other situations where a fault in an individual component results in a failure of the whole system. Examples: capacitors or resistors in series. Or a more subtle case is LEDs in parallel, where a LED approaching end-of-life will have a lower breakdown voltage so all current will go through it (mostly) which will accelerate the LED breakdown. Then, then next LED with the lowest voltage takes over, etc.

[@proyectosledar]

Love to see that dummy load again. excellent video Dave. thanks!!!

[@Shaybay922]

I love how practical this is! I thought there was some deep physics explanation, but nope... It basically comes down to time and money efficiency!!

[@Daveyk021]

Wow, I never knew about the heat generated inside a capacitor. That Flir Camera is a wonderful teaching tool.

[@fukyougooglification]

thanks mate, your enthusiasm and depth of knoweledge is a pleasure

[@craxd1]

ESR and longer life are tied to the same thing. ESR causes heating, which will dry out an electrolytic. Parallel them, and the ESR is cut in half. It also increases the power handling ability of the capacitors. All of it together leads to longer life.

[@richcanfield1387]

Dave Gelman...This maybe what you were looking for, if you have not found it yet! "Parallel caps with different capacitance values" Is EEVblog #859...Cheers to anyone searching for it...Bob's your uncle! Enjoy

[@cny02253]

In my experience electrolytic caps in parallel are most often found in RF Power amplifiers to overcome voltage rating limitations and arc-over issues. They are put in parallel or series-parallel, each with its own bleeder resistor. Commonly with 3K to 5K volts running >100 watts across them.

[@m4d3ng]

This video takes me back to my Polytech days. Theory on the whiteboard followed by a practical. Brilliant.

[@lmiddleman]

Lots of good items on the "capacitor specs" list, but it's notably missing an important one...working voltage.

[@EEVblog]

lmiddleman True.

[@anarchy3960]

Maybe temperature as well

[@JoeJ-8282]

Yes, but working voltage isn't necessarily a reason why to parallel capacitors, which is what he's talking about here in this video... Working Voltage definitely IS an important spec to consider when installing a cap, but that spec really has nothing to do with WHY to parallel capacitors.

[@AngDavies]

@@JoeJ-8282 but it does feed into size - a higher working voltage capacitor is going to be (much) physically larger for a given capacitance all else being equal, which might run afoul of your design constraints. On the flip side if one high voltage cap fails short, things are gonna get interesting real quick.

[@davecc0000]

"Electrolytic Caps 101". Great vid Dave. Chock full of years of practical experience. Suggest making "part 2" next addressing questions raised here in the YT comments and in the forum.

[@EEVblog]

davecc0000 Yeah probably needs a part 2. I forgot to include a few things I originally had in mind. That's the problem with not keeping a list before I go record a video.

[@davecc0000]

EEVblog Yeah, terrible. (c; You cover 99 of 100 things in the category of e. c.'s. better than any educator I've seen. It's not due to failure on your part I ask for part 2 but your subscribers' insatiable desire for more knowledge!

[@jerry2357]

This is a really interesting video. I have one quibble, though. For the temperatures and conditions you are talking about, convective heat transfer (either natural or forced convection) will be considerably more significant than radiative heat transfer, so your Stefan-Boltzmann equation is not the one to use. I would use heat dissipation Q=UAΔT, where the overall heat transfer coefficient U depends on the air flow (in the case of natural convection the air flow depends on the temperature difference ΔT).

[@jimthannum7151]

I echo all the previous comments excellent video, you present it such a practical and fundamentally useful way........far better than my EE Professor who would write formulas as a stream of consciousness with his right hand across the board as he erased with his left hand. Yep, gave away my age it was a blackboard!

[@PankajKumar-zr3tv]

I feel blessed that I found your channel

[@RaisingAwesome]

Great video. I binge watch you on 2X and feel like Neo when I'm done. This video explains why my 1991 Mitsubishi Mirage would eat ECUs (aka brain box for the engine) every three years. After the third one went, I finally opened up the case versus submitting it as a core exchange. I found dielectric had spewed out of a large capacitor. I swapped the capacitor out and got a few more years out of it. In hindsight, I should have rigged up two capacitors in parallel to fix it forever.

[@Stelios.Posantzis]

Great video series that exhaustively explains a neat selection of problems from a purely practical perspective. Great for honing those basic electronic skills and for refreshing areas you could be taking for granted otherwise.

[@Jedda73]

That was a great video Dave. I first looked at the time and thought to myself "half an hour!, I'll just watch a bit till I'm bored", and before I knew it you were apologising for making a 30 minute vid.

[@00Skyfox]

Another way cost can factor in is if the product is manually constructed. If the builder has to solder in numerous parallel caps instead of one larger one, that adds to the costs of both time and money and may call for replacing such parallel circuits with one larger component. It's definitely something to keep in mind.

[@KX36]

An obscure reason regarding ESR; DC-DC converters have LC filters, typically rather high Q/low ESR as there is a high ripple current. LC filters resonate and you can damp this with an RC snubber in parallel with the capacitor. If you can make something close to this RC snubber out of a single cheap capacitor with about the right ESR rather than expensive low ESR cap and a resistor, that's a win. So you often see a cheap cap in parallel with a low ESR cap in the output of these. You might see this on a second stage filter outside the control loop for example. ESR is my main reason as DC-DC converters / SMPS are my thing. Converters I build have specific requirements for capacitance and ESR for transient response and ripple voltage etc and low voltage, low capacitance caps have much too high an ESR so I might split a cap into 3 to 5 parts for example to meet those specs. I also use expensive solid electrolyte caps from nichicon when I can, but that's impractical in mass produced consumer items.

[@steventrompet3151]

Good video, great topic. Shows many aspects, and the huge discrepantion between the newbees theory in class and the oldguys implementation in industry.

[@paulbizard3493]

Thanks, very nice video. Very instructive. (Note: the capacitor surfaces A1 and A2 are swapped 14:17)

[@kjamison5951]

A great video, Dave. Well explained with diagrams and your usual delivery. That's what makes it so enjoyable! I'll be looking for that video explaining the non-electrolytic caps. Good reason for making #742 after #740. Some great comments on your twitter feed! Keep her lit, mate!

[@bdon7825]

EEVblog Hi Dave. What is the purpose of the diode and how is it placed into the circuitry between the caps and the scope?

[@CH_Pechiar]

good video! The bench demo at the end complements a lot.

[@achemachew]

I wondered the same and now know why this occurs Dave! Thanks for yet another great video! Cheers!

[@daveblane6442]

Dave, you are an effing genius! and a fine teacher! Bob's your uncle!

[@MartenElectric]

Brilliant introduction to electronic design. Please make more videos like this.

[@gamccoy]

Very interesting. I appreciate the time that went into this. Good engineers always leave a safe margin.

[@samba3403]

Control the question to the easiest possible answer. Way to go there Dave.

[@jp_engineer]

Thank you... Dave So great class. I wish I had also this kind of class in college.

[@yosefmacgruber1920]

Good job on putting the 10 capacitors close together on the breadboard, as on an actual circuit board, they likely would also be crowded close together, and there could be some restrictions upon air flow as well.

[@GiancarloAllasia]

Hi Dave!!! Always useful, I really love your videos!!! I only want to point out that nearly 13:40 when you explain capacitor dissipation, A1 and A2 are switched in formulas. Apart this, great vlog!!

[@lasersbee]

Another GREAT tutorial video. Very enlightening and informative. Thanks again Dave. Love the demo. A picture is worth 1000 words and gets easily stuck in your head ;)

[@bollllllleeeee]

Just wanted to thank you good sir. You and some of your peers on KZbin, has taken electronics out of the realm of black flipping magic and in to the realm of this is easy. By easy i mean most of my components survive an encounter with the on button. You are awesome!

[@chriswouse7713]

cheers, EEVblog (Dave), I've never considered the power dissipated in the ESR before, Will be bearing in mind for future applications!

[@fieldsofomagh]

A lot of relevant information to digest. Could you please pass the Rennies .Esr in parallel was an eye opener.The amount of variables that interconnect in a simple design project is mind boggling. I liked the video and the test bit at the end.

[@EEVblog]

Tom OConnor I should have measured the ESR and shown that in practice, I forgot!

[@whitcwa]

Reason 10 to use multiple caps is if you have a series inductor between them (Pi filter). I know they aren't actually in parallel then, but they may look like they are in parallel if you don't follow the PCB traces.

[@NOnEveReallyDys]

Please do more fundamental fridays. Best videos thanks! :)

[@jonw0224]

Nice video. Very practical. I always enjoy the Fundamental Friday videos!

[@MrBanzoid]

Most informative Dave, thanks. I'll remember this vid next time I'm tempted to just drop a capacitor into a design.

[@deanrubine2955]

Caps fail short so aren't particularly redundant in parallel. It's a tradeoff where the reduced stress on a cap reducing the probability of that cap failing is weighed against the probability of system failure due to a single cap failing in a parallel array.

[@TCBEperformance]

Correct me if I'm wrong Dave, in power supplies on the rectifier smoothing with 1 large capacitor will take longer to charge especially with the ESR factor. So with 1 large capacitor, it might get more ripple under high load compared with 4 capacitors to the same value as 1. reg. joe

[@davidgriffin79]

Really like this. I was into electronics before I decided I liked mathematics more and got my first degree in maths. Now that I do mechanical engineering as a profession it's good to get back into electronics as a hobby. I follow this guy with an open Mathcad sheet and have just constructed a sheet that calculates effective capacitance and effective ESR for a multiple capacitor power supply.

[@MrZetor]

10. Precision I don't know if there are too many discrete-component applications in which the capacitor value(s) must be as close to the design specs as possible, but at least in the IC domain, using parallel caps is pretty much the only way you can get reasonable yield. I don't do analog layout myself, and don't know whether the cap value distribution (due to process variations etc.) is Gaussian or something else, but it is rather obvious that the more parallel caps you use, the closer to the required/specified capacitance you will get.

[@MrHBSoftware]

precision is never a thing with electrolytics...TYPICALLY you can get away with a 1000% bigger capacitor or a 20% smaller one... they are typically for filtering or decoupling so no precise precision involved

[@markharder3676]

Thanks, Dave. Good question. Good answers. Didn't mind the time at all.

[@Steven-pv1xr]

Nice presentation lots of energy and interest way to go

[@optimusone1985]

Even Man is designed with two capacitors in parallel adequately positioned for better heat dissipation and a reliable system for redundancy to keep your load happy in case of a partial malfunction.

[@anwarshiekh7142]

Did you miss that the inductance goes down when in parallel

[@larrybud]

I love practical production advise like this. People think companies have unlimited resources, and that customers will pay unlimited amounts for devices, but it's just not true. When I worked in automotive, they would have entire programs designs to save literal pennies on fasteners, for example, because they use MILLIONS of a specific fastener. Adds up to real dollars, and whether the vehicle is profitable or not. Same goes with production electronics.

[@frankvilla1624]

can you show us how to bleed / discharge capacitors ? please

[@S13Reborn]

If I had an electronics teacher like you at college, possibly I would have changed to electronic engineering at the time. We had a nasty terrible man... THANKS FOR YOUR WORK!

[@joaquins90]

Hey Dave, nice video, as usual showing real world cases and explaining why they are done! I'm not so sure about the last bit, you only included radiation, but ignored convection which in this case is probably higher than radiation. Thanks for all this videos! JS

[@johnsmithaha]

32:24 stephan boltzman constant: "some weird ass funny number you learnt in physics" hahahahaha

[@davidgriffin79]

BTW A1 and A2 are the wrong way round for the surface area of the capacitor: A2=pi*r^2 and A1=2*pi*r*l (=pi*D*L). The total surface area will be 2*A2+A1 since you have to include the bottom of the can, unless you assume perfect insulation at the can base due to the circuit board:)

[@tainer7467]

I think the formula for radiated thermal energy should have (Tcap - Tamb)^4 instead of (Tcap^4 - Tamb^4)

[@islandhopperstuart]

No, Dave's formula is correct. By the way, values of T are in K (Kelvin) i.e. absolute temperatures.

[@joiseystud]

would like to see a video on why back to back electrolytic caps can be used as non polarized.

Thank you for curing my insomnia. I haven slept this well in ages.

[@BeyondDuctTapeFixItRight]

That was an excellent demonstration! I also come to discover your FLIP camera is incredibly sensitive. Picking up heat reflected from your hand. I wouldn't have expected that.

[@jethro4520]

Thanks for your videos! A primmer on surface mount pcb assembly for small to moderate volume would be very helpful

[@ChipGuy]

Thanks Dave, that video is going to be one of those that will be important to a lot of engineers. Basically timeless. And you skipped #741 as suggested, awww ;) Looking forward to that one now.

[@robfenwitch7403]

Chip Guy Vids I'm looking forward to #7400...

[@destroyer2012]

Hey Dave when will you be printing out degrees

[@conoba]

A lot he said goes for capacitors in general. Like parralleling a high capacitance bariumtitanate ceramic with aluminiumoxide ceramic one with good high frequency ESR.

[@AlienDelon]

watch at half-speed to get the Drunk Dave

[@martinkuliza]

hehe... tried it..... IT'S PRETTY FUNNY

[@martinkuliza]

@Another Dick LOL... ok i'll try that at another time and if i can put my 2 cents in watch any video of Phil Anselmo (Pantera) when he tells a story except.... Phil already sounds drunk so..... Watch is on 1.5x speed hehe it brings phil back to normal you tube doesn't have a setting that makes phil sound drunk when you tube decided on what speeds were going to be low and high they took a page from phil's book and made the lowest setting based on phils normal talking voice LMFAO then from there i was simply a matter of double that speed, then double it again (was Normal playing speed) then double again and double again Phil is the standard LMFAO

[@martinkuliza]

@Another Dick also watch this scene of CENTRAL INTELLIGENCE kzbin.info/www/bejne/gmbYpaVso8uajZY 1:08 doesn't the princpal look EXACTLY like Dave with white hair ? hehe

[@freemanz4051]

.75 turns me on.

[@martinkuliza]

nah man....0.25 is the money shot hehe 0.50 speed is dave after 3 beers 0.25 speed is dave after 8 beers LOL

[@aaronandannelogan]

At 14:10, A1 and A2 appear to be swapped.

[@Roxor128]

I've just learnt about something new to keep an eye out for next time I pull something apart.

[@mirkomueller3412]

Fantastic lecture !!!!!!!! Would not wanted to miss it. Would give it 10 Thumbs up if i could. Thank you so much.

[@Mr.Leeroy]

ESR is proportionally less with height of the ECap to the point where it is required to choose higher capacitance (exactly because it is bigger physically) that you need e.g. in DC-DC converters just to meet ripple current spec due to ESR.

[@MatthewSuffidy]

The sum of the little caps may be giving off as much energy, but it may be easier to dissipate the larger output area...

[@raindropsrising7662]

Thanks for the sharing and clear explanation. Love it.

[@inthenameofjustice8811]

Brilliant. Even I could follow most of that.

[@dalenassar9152]

Dave, At about 10:40 you state that the ESR of two identical 'lytic caps connected in parallel, gives the total ESR as half of that of the two identical ESR's. I thought this for many, many years, until I saw a discussion from MIT that this is NOT the case in parallel 'lytisc! Note that the two ERS resistors are not actually connected in parallel!! T creates trouble and complicates things. I will have to go back and try to find what the actual conclusion was--or maybe you can help here?

These are of course fundamentals of electronics engineering, but still excellent presentation!!

[@Bodragon]

I wouldn't ever have thought that the higher voltage caps have a longer lifespan. No way! That's just crazy! There's always another surprise just around the corner!

[@steveseifer6784]

Nice instructional videos, Dave

[@T3sl4]

Note that radiation is a very small part of dissipation at room temperature, even for passive convection cooling. Black versus shiny heatsinks only improve by a few percentage points. But hey, if you need those few bits to get your design margin for that cheap consumer product, it's worth it (and the black anodize or paint might look better, or improve corrosion resistance). If you expand the differences (oh snap, that needs Calculus!), the radiation temperature difference can be expressed as a conductivity dependent on absolute temperature. Instead of (T1^4 - T2^4), use (4*T^3 * dT), where T is the absolute temperature of one or the other object (the point being, dT is small enough that the distinction doesn't matter), and dT is the difference. Convection of course is notoriously hard to calculate (convection, fluid flow, ventilation, air density, boundary layers, oh my!), but a sufficiently crude rule of thumb is 0.0013 W/(cm^2*K). So a project box with 100 cm^2 surface area might be expected to have a thermal conductance of 0.13 W/K, or a thermal resistance of 7.7 K/W (for temp differences, K == C so no worries).

[@ciphernemo]

Nice job on the video. I'm not sure #9 of peak currents is that much of a justification, since you can always make caps and traces bigger. If you can't, then that goes back to other reasons from #2 through #6. The biggest reasons in high-quantity manufactured electronics is usually #2 through #6 (ESR, Cost, Reuse, and Product Config) and how they can be worked into #1 for common capacitance values. As for longer life and redundancy, this is usually only worried about in top-end products or mil-spec stuff, not so much in consumer electronics meant to last 5 years or less. For example I've seen high-end PC motherboards with redundant caps and even very reliable, solid caps, but those are there to be touted in the marketing and keep that brand's reputation relatively higher than other brands.

[@EEVblog]

ciphernemo Yeah, just trying to think of every possible reason to show that potentially that's possible.

[@nickb2245]

ciphernemo Actually I was just reading a Xilinx app note where they did almost exactly that. www.xilinx.com/support/documentation/application_notes/xapp623.pdf

[@snaprollinpitts]

Hey Dave, that was an excellent explanation, and great demo of what happens, thanks, you're obviously Da Man!!!

[@EvilLOON]

Great video m8. Love your edu-videos. Still learning everything I can from you and your people on the forums. I have posted a few times with some strange questions. Overall everyone is really cool. Cheers for educating the world.