Capacitor Discharge Box & Test Fixture for EICO 950

[@rankenfile]

May you have many more years of service at the bench too!

[@marcodoria]

Hi Tony! You sure influence a lot of people, especially by being such an honest and kind human being! Thank you!

[@hestheMaster]

Recently went back and reviewed the video on your EICO 950B from January 8 , 2017. It was very thorough and explained the problem with those 100K 4 Watt pot's and what to replace it with. Just recently bought a Centralab WW-104 100K 5 Watt pot, (sorry last one was $27.00) for a back up in case my original replacement burns out or up. Centralab also made a WN-104 only with a shorter shaft that fits perfectly in the EICO. Someone has those on EBAY. This device is a great idea (I do have separate capacitor discharge device) to save that pot. So I actually bought one of those CD units. Called a Skiliwah HV-S4 discharge pen. Connected to a metal box it has two permanently connected leads with instructions on how to safely discharge a capacitor up to 1000 VDC. The HV-S4 instructions mentioning what is inside and says it discharges through a 100 K 8 W resistor. There are two LEDs that light until capacitor is safely discharge. Leads are non polarity driven actually but either one of LEDs will light until going out indicating the capacitor is completely discharged. There are no switches on the box. All this is in an insulated metal box which is of recent manufacture. This was another great project video Tony.

[@welderfixer]

Tony, Thanks for the education! Good old gear sure needs to treated with TLC. I have so much to learn from experts like you and Mr. Carlson. You guys are great!

[@raymondlewis2055]

Thank you, Tony, for another great video. You are a natural teacher and I have learned a lot from your videos. I wish for you and your family a very happy, prosperous, safe and healthy New Year. 73, AA4EZ

[@jonka1]

Nice idea Tony. If you want to protect the pot from discharge current you need to put a 1000v diode in series with the feed wire from the main unit. Your device will not protect the pot if the operator forgets not to turn the pot down and it is likely only a matter of time before this happens.

[@dzee9481]

Tony I find your video where you describe some of those buggy transistors that either have leakage issues or noisy. You had a list of those transistors in one of your videos, dont remember which one but it would be great if you could bring that list up now and then. I will write down those transistors that are issues. What I am finding difficult is finding the replacement for these transistors. I think you went over a video on how you go about replacing those transistors. This would be a great value on how you go about doing this. Digikey, I know is one of your main source for replacement components. I figure this would be a great video like this one, but it would help some of us that can restore these vintage equipment. I have to say vintage equipment is better than the equipment you find today which is design to be thrown in the garbage or recycle, compared to vintage can be restored and work for another 2 generations, to bad us vintage folks cant be like that. Sorry to bla bla bla but your videos are teaching young people electronics better than what is being taught in school.

[@greengrayradio1394]

Happy 2022, Tony! Nice add-on to the tester. Actually, I built something similar, but used a power NTC as the discharging resistor. Fits in a transparent plug-case which plugs into the DMM. Has a bi-colour LED on top, this shows polarity as well :)

[@ko2fjb]

I have one of these cap testers, never knew why the Voltage was low. I will be building the discharge box. Thank you for this valuable information.

[@ssboot5663]

Your knowledge and patience is amazing!

[@Cavemannspace]

Excellent as always, Tony. I was wondering about doing something like this at one time but didn't follow up on it. (Shame on me.) I think th guys at the radio museum will like this too. Thanks! Great video to start off the new year. Peace, joy, and happiness to you too.

[@mikefinn2101]

Great video and glad you did this. I always learn something new thanks for sharing. Not happy with past negative comments. Some of us learn something and are grateful for time and effort spent on helping others. This is very kind of tony to do this. He knows his stuff. If you don't like his videos don't watch. If you can do better then do your own videos. Be considerate of others who enjoy learning something different. Well done Tony

[@CovertCulvert]

Sure glad that I ran into this video at this time. I'm in the process of restoring a 950B. The voltage pot in the one I got was burned clear through the resistor wire. It looked like it had been dead-shorted, but now I realize it was how the unit was (mis)used that caused the damage. I found one (and only one) of those pots on ebay. Fortunately, after a thorough cleaning the pot measured 103K on the nose, and was linear from end to end. I'll be building your test/discharge box, because I don't want to have to find another one of those inobtainium pots. Thanks for the great videos and all that I've learned from you by watching.

[@peterchambers1868]

Thank you Tony, I just recently purchased a 950, and will restore it...now I know it's weaknesses...I believe I will build your box...Yhanks again. Peter

[@ianide2480]

When I built my own leakage tester I used a switch on the output that shorts the capacitor across a resistor. It maintains its connection to the voltmeter so you can simply watch the needle drop as the cap discharges.

[@w2tty]

This was clear and very informative. Thank you. I will be building one of these.

[@TheGuitologist]

Interesting stuff. As the owner of one of these, I appreciate your videos on them, Tony.

[@xandear1]

Great job Tony! It was awesome remembering good things about your cousin and retribution through fixing this device. I wish a great year to you and I'm waiting for new KZbin videos.

[@craxd1]

I made a discharger for working on HV amplifier power supply circuits similarly, capable of quickly discharging a bank of caps at 3 kV and less. I think I used a large wirewound 100k resistor to do it, along with a switch. It discharged much quicker than waiting for the bleeder resistors across each cap to do it.

[@williamsquires3070]

Hi, Tony. I’d put an extra SPDT switch (labeled “Hi/Low”) to select between two discharge resistors; one for high voltage, and one for low. I’d use a 2k/10W for the “hi” setting, and a 470/5W for the “low” setting. 🤠

[@rrowan327]

Happy New to you and your family. Many thanks for providing excellent electronics videos and thanks to your cousin also.

[@craxd1]

Let me add this. You are very correct about the 4 watt 100K rheostats. One major supplier only carries two brands; one, the old Clarostat brand now owned by Honeywell, and Kyocera AVX. They have no stock of either, and there was no listing for a 100K at that. A 5 watt from another seller was $89. You're lucky to find them at an affordable price. One might try to find a dual 200K, 2 watt, and wire both sections in parallel. How well that would work, I don't know.

[@hernancoronel]

At 5:25 I’d bet there is some KZbinr that could take it apart and fix it no matter how difficult. Thank you for the video!

[@t1d100]

Thanks, Tony. I think this video has helped me discover a dangerous error in my DIY Cap Discharger. I had simply placed a high ohm/high wattage resistor across the probe inputs of a disposable DMM. I now realize that maybe all I did was create a resistor divider with the DMM's internal resistor. So, I will be checking my design. My thinking was that the DMM would just be watching the cap voltage drop. But, that may be wrong; the DMM may be carrying more of the load than I thought? I believe that your Discharger design uses a Make-Before-Break switch. If so, you may want to emphasize that point, for folks that might be using alternate parts.

[@t1d100]

It appears that the DMM of my DIY Tester is situated in the same position as your LED and its protective resistor. I do not have the diodes. I knew that the voltage reading would be incorrect, for my design. But, I could see the voltage drop to zero and that was good enough. Is my design dangerous? IIRC, I used 1M ohm resistors, combined for a couple of watts... Edit: I relieve you of all liability, period.

[@johncoops6897]

Your DMM will have input resistance of many (tens or hundreds of) thousands of ohms in Voltage mode. Just use another meter to measure the resistance, you cannot harm it. Voltage measurements are just that... in a perfect multimeter the meter adds absolutely ZERO load. That's the way that voltages work, you can put 100V across the multimeter and it won't harm it nor draw any appreciable load. Note: 1M resistors will discharge a capacitor VERY slowly. I guess it doesn't matter if you are prepared to wait. The time depends on the capacitor's charge, both voltage and uF of capacity.

[@t1d100]

@@johncoops6897 Thanks, good sir. Good info. Yes, for an 1M individual stinger design (meaning a 1M resistor without a multimeter attached,) the wait time is longer. I like the safety and that a slow discharge won't harm the cap. You are correct about the longer discharge time period, but I do not mind the wait. I understand that other folks do not want to wait as long. To each their own... But, I still think that my design is a bit faulty... The multimeter is a freebie from HF... If you don't know what that it is, it is the cheapest, plastic, Asian type. Meaning, if it is destroyed, I won't care much... As you suggested, I measured its resistance. It is just 1K Ohms, in all voltage ranges. Meaning, the equivalent parallel resistance is 999 Ohms. That means that the meter reading is not corrupted much, but the meter resistor is also doing all the work. Correct? I mines well not have bothered installing the 1M resistor, I think. This condition also occurs anytime that someone uses an individual discharge stinger and monitors the voltage with a multimeter. Well, a DMM, I think. I am not sure about what effect an analog voltage meter would have... Its all an interesting consideration...

[@Shnick]

My 950b is a great help testing audio caps in old hifi equipment. I’ve considered putting a digital volt meter face in it to get a correct volt reading

[@timp9848]

Happy New Year Tony. Thanks for another excellent video. As a beginner I have learnt so much from you, thank you! Would it be possible you could do a video for beginners on Solid state Amplifiers on how to diagnose with an oscilloscope and a signal generator, I watched your videos on how to check the output signal with the oscilloscope but I would like to learn more on how to best use the scope probing the amp internally. Thanks again and all the best!

[@SDsailor7]

Yeah good suggestion i would also like to learn how to do that and how to connect the O'scope to the DUT

[@MrDoneboy]

Happy New Year, Tony!

[@erikdenhouter]

8:45 Idea: put a diode with a parallel resistor in series with the wiper to limit the discharge current ?

[@KeritechElectronics]

Hey :) I absolutely enjoyed the video! That Eico tester is quite an interesting device. Compelled me to print and study the schematic & principle of operation on my own, as I want to make a capacitor leakage tester. The voltage divider circuit is formed by R7,P3 and R8 || V2 internal resistance. This schematic is not that messy... at least compared to Heathkit IT28. Now that one is drawn in such a complicated way that I had a hard time studying the signal flow when I analyzed it... Instruction manual with diagrams for each one of the modes helps a lot in understanding the principle of operation. J3 and J4 is only used for reference part in comparator mode. The leakage test uses J1 and J2. Nice test/discharge switch circuit. That's what Eico doesn't have, and Heathkit does :)

[@SDsailor7]

Are you talking sbout the heathkit IM28 vtvm?

[@KeritechElectronics]

@@SDsailor7 not VTVM but a capacitor tester & RC bridge

[@garygranato9164]

those binding posts looked way to good quality to use on a capacitor discharge box. nice vid, happy new year

[@xraytonyb]

They are actually eBay specials, but they look expensive ;)

[@spuncut]

Hi Tony love your videos. How about a diode on the inside of the Eico in series with the "+" to protect the pot if one forgets to discharge. Keep up the good work and happy new year.

[@MarcelHuguenin]

Excellent video! Happy New Year!

[@SDsailor7]

Excellent video! Very informative and entertaining. Thank you

[@tubeDude48]

Great project for the new year!!

[@jamesmdeluca]

Greetings: Suggestion: I would use a momentary switch to default to discharge mode whenever the switch is not being actively held in test mode. The toggle switch version is probably easier to find than the pushbutton version which is my choice. When used as a stand-alone discharging device you need not worry about the switch position, and when used with the leakage tester it is safer as the potential high voltage exposure time is limited. Also, I would be inclined to build this into those leakage testers that do not already have this feature.

[@edwardcollins6895]

Thanks for the tip. I just restored a 950B. I don't see any reason why it wouldn't be equally susceptible to burning out the pot.

[@nedmojo6844]

I'm a little new to this and just got one of these 950B's, so I would like to make this box and use it to dicharge the large caps in valve amps; just a couple of questions (as I try not to blow myself-up): What are the specs of the toggle switch, LED and the 47k resistor? Thanks great video.

[@avro66]

I remember while serving in the RAF, the electrical bay, personel would charge up a capacitor and throw it to some unsuspecting indervidual and they would get a large shock, they never made the same mistake again 😂😂

[@gd2329j]

Always keep three or four on the workbench . It keeps the undesirables away !

[@avro66]

Indeed 😂

[@Shnick]

My 950B suffers from this issue. I’m switching it out this week, luckily found one on eBay.

[@NiHaoMike64]

Could the potentiometer be replaced with one of higher value and then a pair of high voltage transistors added as a unity gain class B amplifier?

[@gd2329j]

I just throw a 1 k ( 1 or 2 Watt ) resistor under the pot . Yes it moves the voltage up a bit but it's a low cost fix & works . You cannot move the knob that fast . If you're pushing the tester beyond its design limits ( 22000 uF cap LOL ) the 1k is the weakest link !

[@gizmothewytchdoktor1049]

thank you tony. your work is appreciated to be sure. =dok=

[@terencekaye9948]

Totally Awesome Tony,,, Happy New Year to you. :)

[@robertcunningham1542]

Very good circuit and useful device. However one should use a very heavy-duty switch, with high value capacitors and high voltages the switch contacts can fuse together. A varistor in series with the shunt resistor would help, but would increase the discharge time

[@TMG363]

Nice work and way over my head, but why didn’t you put your voltmeter leads right into the spare receiving banana connectors of the ext box? Green/red wires

[@xraytonyb]

That's what the LED indicator does, so no need for the meter. If the LED goes out, the voltage is at a safe level. You could, however, connect a voltmeter if you wanted, in the case of a really high value low voltage capacitor.

[@TMG363]

@@xraytonyb exactly….when you were measuring that 250v cap you could have been hands free. I really appreciate all you do!

[@CNKayutube]

You used a couple of different styles binding posts/terminals. If you do a Q and A segment, please talk on the styles/ brands. I bought some banana Jack to alligator test connectors, and the banana Jack's last about two insertions, total garbage. So, any feedback much appreciated! Thanks

[@ko2fjb]

This was great, thank you, I will definitely put one of these together for my 950. Just curious, if I wanted to measure the amount of leakage could I simply put a amp meter such as a Simpson 260 in series with the DUT?

[@budboggy4699]

Hi Tony, I've learned a lot from you, thanks. Will you (or anyone) share a source for a comparable replacement 100K 4W pot please? A few years back I inherited several 950s, three of them need the power pot - all need caps.

[@CovertCulvert]

I just saw your post. I found ONE on ebay. It had to be completely disassembled and cleaned of the old grease that had solidified, but now it works just fine. If you still haven't found one, keep an eye on ebay.

[@andymouse]

Great video !...cheers.

[@brucebuckeye]

So if I understand this right, this Eico tests for leakage, which comes from equivalent parallel resistance. And an ESR meter checks for series resistance.

[@xraytonyb]

The Eico can measure capacitance and resistance. In addition, it can measure leakage (parallel resistance) of capacitors as well as power factor, which is similar to series resistance or ESR. The 950 also has a comparator function that lets you compare the device under test to a known reference component. These are pretty versatile units. The leakage tester is also used to form capacitors. Sometimes, new caps that have been sitting around for a while will need to be properly formed before putting into use. The 950 has two levels of current limited voltage up to 500 volts for film/paper caps and electrolytics, when in leakage test mode and this is perfect for re-forming old caps or forming new caps that have been sitting around for a long time. Thanks for the comment!

[@johncoops6897]

They are quite different things, rather than just the type of resistance. ESR is measuring impedance at usually 100 kHz AC. Leakage is more about the amount of DC that the capacitor will pass. ESR is more relevant to modern devices, where you want to shunt high frequencies to ground. Leakage is more relevant in old Tube devices where the caps are primarily to block DC between stages. A dried out electrolytic might have high ESR, but leakage will be Low (or zero). A capacitor could be shorted or excessively leaky, yet an ESR test may well read perfect. I know that is a vast generalisation, my point was that they are different things, and different specs on the datasheet.

[@srtamplification]

Tony, when you turn the voltage down to 0, doesn't that take the wiper to ground and therefor not have the discharge of the capacitor go through the pot?

[@xraytonyb]

When you set the voltage on the leakage test mode, the cap will charge up to the voltage that the wiper is set to. When you rapidly reduce the voltage control, there will be a voltage difference between the wiper of the pot (connected to the cap) and the resistive element of the pot itself. This voltage difference can be large enough to overload the pot and open the thin resistive winding. I've seen it several times myself on 950's that I've repaired. This is especially a concern with caps that have high capacitance.

[@srtamplification]

@@xraytonyb It just looks to me that the cap under test gets connected between J1 and J2 and then gets discharged back to ground through R11 or R12 depending on which leakage setting you have. Are you saying that the ground at that point can get elevated up to a +500 Volts from the cap discharging giving a 1000V differential across the pot? I would think that the discharge voltage is already referenced from the pot where you would never have more than a 500V differential across that pot. I get what you are saying if that is true though that's like 10W across the pot at that point albeit for a short period of time.

[@xraytonyb]

Think of it this way; You have a high value cap charged up to 350VDC. You turn the control down quickly. The resistance between the wiper and ground is suddenly very low resistance, but the voltage between the wiper (with the cap on it) and ground is still very high (cap is still charged up, at least partially). This pot is a 4 watt component with 2.5 watts already being dissipated across it by the supply. It won't take much to pop the nichrome wire element in the resistor. On all of the failed controls I've seen on these, they always open about 1/8 to 1/4 of a turn up from full CCW. The one in this video was no exception. A much higher wattage rheostat (10 - 20 watts) would most likely be able to take the short current spike, but not these ones. The problem is that there are no high wattage potentiometers rated at 100k ohms.

[@srtamplification]

@@xraytonyb I totally understand what you are saying, but you have other resistors in the circuit that the voltage will be dropped across. Looking at it I believe that voltage drop will look like a negative voltage on the ground side of the pot which could be as much as -500V. That would be 0V difference across pot. The way I'm seeing it, you would never have more than 500V across the pot. It would be interesting to see an analog meter or scope across the pot during discharge. New video?

[@gd2329j]

@@xraytonyb High wattage potentiometers are low wattage at the ends ! As are all potentiometers .

[@chriscimino7854]

Yesterday I saw an ad that advertised radio for sale one dollar volume stuck on high. Thought I can't turn that down

[@johncoops6897]

That joke should not be a loud here.

[@chriscimino7854]

@@johncoops6897 everybody else was ok with it

[@johncoops6897]

@@chriscimino7854 /whooooosh

[@peterchambers1868]

Tony do you know the difference between a 950 and 950B?

[@xraytonyb]

Here is a link on the antique radio forum. The 950b is a slightly improved model, but they are mostly the same. There was also a 950A. antiqueradios.com/forums/viewtopic.php?t=234534

[@troy6882]

Drop gate resistance across the test??A Director drop bypass??Nice😃😄😁😆😅 5 Smiles.

[@davidlong1786]

You say that by using a 10 Meg input DMM it will load the tester and give a lower reading than actual, so how did the manufacturer measure the output voltage when they calibrated it? Show us how to make sure the output voltage is as claimed please.

[@davidlong1786]

Please address this question.

[@xraytonyb]

It can be calculated using Ohms law, Kirchhoff's law and Thevenin’s Theorem. R7 and P3 are in series, forming a voltage divider. If P3 is turned all the way up and there is -520V at the input of R7, you will have -503 volts at the wiper of P3. This would be max voltage. In paper/mica mode, the voltage is fed through R9 (270k), then the device under test (your 10Meg input impedance meter in this example) and R10 paralleled with R12 (which equals 450k) to ground. If you take the 10 Meg input impedance of the meter and place it in series with the total resistance of the resistors (which is 27k+450k = 720k ohms), you will have a series current of 503 volts / 10720000 ohms = about 47 microamps. If V=IR, then 47 microamps X 10000000 ohms (which is the impedance of your meter) = 470 volts. This means that your meter will read 470 volts and will act a a leakage current of 47 microamps. This would be a fairly high leakage current for a modern film or mica cap. Maximum short circuit current is 503 volts/720k = .69 mA. In electrolytic mode, the 270k resistor is bypassed, allowing for a short circuit current of 1.1mA. Placing your 10 meg meter in the circuit would cause it to read about 481 volts, still lower than the 503 volt input. Measuring from the wiper of P3 to ground and NOT between J2 and J1 will show the actual input voltage before the current limiting circuit. Don't be surprised if it reads something different than -503V, as the mains voltage and rectifier tube will influence the voltage. In addition, you must verify the accuracy of the resistors, as they will affect the circuit and your readings. Hopefully, this all makes sense.

[@psionl0]

It occurs to me that a transistor between the pot and the output would also protect the pot from high capacitor discharge currents.

[@RapperBC]

(raises hand) "Uhmm... I was told there would be no math?" {8^>) (actually that math was useful and necessary; thank you)

[@hoobsgroove]

what's the point you might as well get the carlson's lab capacitor leakage tester and prediction tester. works much better than those old machines and more sensitive.

[@georgejobin1744]

You can put the real voltages into the cap plus you can reform caps with this unit also you don't need to make a curcuit board and solder smd componits that's hard for a beginner that's the point

[@hoobsgroove]

@@georgejobin1744 I'm not talking about his device he made not a good idea to reform caps you don't need high voltage to measure leakage

[@xraytonyb]

Thanks for the comment! First, this is a piece of vintage test equipment that belongs to my 85 year old cousin. I repaired it for him, as it has sentimental value to him and it still works very well for its intended purpose. Second is that this unit has the ability to form capacitors. Many people mistakenly think that forming capacitors is just a practice used to squeeze out a bit more life from a component that should really be replaced. This is not the case. Brand new capacitors can sometimes need to be formed before you can use them. If you watch my video series on the Fisher X-101-C restoration, you will see an example of this. I purchased a brand new can capacitor from CE manufacturing and it shorted out the power supply. Luckily, I had it on the current limited "dim bulb" supply. After applying a current limited voltage to the cap and slowly increasing it to it working voltage, it re-formed and has been working perfectly ever since. This is more common than people realize. The 950 works very well in this capacity (pun intended). Last, leakage is not the only factor that can determine a capacitor's condition. Capacitance, ESR (or power factor, as this unit tests) and dissipation factor can also affect the performance. The 950 can test all but Dissipation factor and Φ, at least to some degree (pun intended again). Super high leakage resistance testing is important for certain types of caps for certain applications, but is not as important as some of the other factors for other types of caps and applications. Thanks again for the comment!

[@hoobsgroove]

@@xraytonyb it might be ok reforming for a new capacitor but it's not really wise to use it on an old capacitors because they've decay!!! and could cause serious damage and injury a short way down the line, it is poor practice what's the point they're not particularly expensive anyway

[@gd2329j]

@@georgejobin1744 The carlson's leakage tester is just that a leakage tester ! It's not a working voltage or breakdown tester . PS You don't need to make a circuit board or use smd construction .

[@DeadKoby]

With BIG caps you've gotta go slow......... at the time that this cap checker was a thing, most tube circuits had 100uf or smaller as the biggest capacity, at really high voltages.

[@jdmccorful]

Nice fix for the 950. Wish you could send me the file for schematic, can't screen print from my tablet. Thanks for the look.

[@Gengh13]

Volume down+power button simultaneously pressed for a second usually takes a screenshot on most Android devices.