Testing Vintage Bumble Bee Capacitors for DC Leakage Current 

I have afew 10,000 Mf at 80 volts Electrolytic and must of them are about 1.2 mili amps leaking only my Nichicons are 500 Micro amps

Hi Mark - I agree with you. I am not of the mindset that every capacitor has to be replaced every time someone sees it. Capacitors do go bad and start leaking DC in coupling circuits and do need to be replaced but if they are not blocking DC from one stage to another, it probably doesn't matter if they have a few megohms of leakage and they still work just fine. Not only is this wholesale replacement of capacitors not necessary as some believe, there is a good chance that the skills of the "surgeon" may not be the greatest and the amplifier ends up with a mess under the chassis and even wiring errors made. I believe in replacing bad components and leaving others alone if the equipment is working properly.

I am interested in how they turned out. Were most good or did most lead DC and have to be replaced?

i also test my caps that way...much safer...630v 500v and 250v ones i just stick them in the 220v socket with a diode in series and read DC the voltage with DMM...a new capacitor will spike to lets say 20 or 50v and drop to zero...a leaky one will let dc pass constantly

I looked on your channel and I didn't see the video you mentioned...

That is a good question but I would guess it starts by temperature inside the capacitor operating at a high value (vacuum tube equipment in particular) over a long time and damaging the dielectric. I think it is an accepted parameter that for every 10 degrees C rise in temperature, insulation life is cut in half for many insulating materials. I would think some modern day insulators, Teflon being one of them, the life of the insulation would be greater. In the 70's there was a rage for the "sound" of paper-in-oil capacitors in audio amplifiers. I am talking about the 0.1uF, 0.05uf, 0.47uf, etc. coupling capacitors, not the power supply filter capacitors, and as best I can remember, every one of those amplifiers that I worked on up until around 2015 had bad leaking capacitors. Every capacitor in the amplifier would have to be replaced. There is Internet hysteria about how bad Bumblebee capacitors are. There are bad ones, of course, but I have a lot of equipment that is full of Bumblebee capacitors, including McIntosh amplifiers, Collins Radio equipment and much more, and they work just great. I do not recommend replacing all capacitors every time the bottom is taken off an amplifier as some do. On the other hand, if you were to find a 1960 vintage Fender amplifier that has been in a closet for the last 50 years and you are going to take on the road to make a living, it would be a good idea to replace all of the capacitors. Actually using equipment will make it last much longer than storing it away trying to save it decades for the grandkids. :-) Thanks for your comments.

Hi Ron - I am definitely an orange drop capacitor fan. I tend to buy them 10 or 12 at a time and when I measure them, for one thing, they are amazingly accurate and in addition, even if they are off a fraction of %, they are matched even more amazingly close to one another. They have done a really good job over the years improving these capacitors. I can't see spending $$$ for these other high-end fancy capacitors. Maybe I just don't have the ear for the difference they are supposed to make. BTW - just a standard analog meter like the Triplett or Simpson put in series with the capacitor and HV supply will tell you if there is leakage. Nothing fancy needed there. It is not likely we need to know the Q and D of the capacitor for audio work. Thanks for your comments.

The amount of change in capacitance would depend on how much the capacitor is leaking DC. In many cases, the value of capacitance may not change any significant amount. However, if for example, a coupling capacitors can be read as a 1 megohm resistor on a standard VOM then the plate voltage on the preceding tube which may be in the order of +200 volts will dramatically change the bias voltage on the tube that it is driving which may be in the order of negative 10 volts. In a case like that, it is obvious that the amplifier is not going to work very well at all. With a vintage type analog VOM like the Tripplet or Simpson meters, it is very possible to put the meter on the highest ohm scale and connect a (discharged) capacitor across the meter and watch the meter needle jump to the right and then settle back down to infinite resistance. That is probably adequate enough to evaluate most coupling capacitors. If after leaving the meter connected for maybe 30 seconds and the meter still reads some resistance, the capacitor is leaking DC and should not be used as a coupling capacitor in a vacuum tube circuit. The same procedure can be used on electrolytic capacitors but in many cases the DC resistance may not go all the way back to infinity and the capacitor still be OK. This is more pronounced on the larger capacitance value electrolytic capacitors than smaller values. Truly knowing whether or not a small value coupling capacitor is good or not takes several measurements (capacitance value, leakage, ESR) and most people do not want to go to the trouble and they just change the capacitor. I don't do it that way as I like to keep equipment as original as possible but at the same time, keep it working properly. Hope this helps.

Actually no connections between the two. I make damaged afew electrolytic and they got very leaky and whe I tested ESR and Cacitance....my surprise....must of them test perfect. And some of them show no leaks but very High ESR or low Capacitance.

+Britt Rossman It sounds really interesting but I am not sure how to go about it. Build a simple voltage amplifier with carbon resistors then switch them out with metal film? Something like that? One of my YT buddies did make a video on capacitor distortion that you might like in the link below. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-xfAgOdOrJ1c.html

Hi rb, I have made a couple of videos on the GR 1650-A that you might like. As for external voltages, I don't remember off hand but we do have to be careful with external voltage sources as we can damage the unit. Also, I have rarely ever needed external input except in the case of using maybe 10K or 20K Hertz (at 5 or 10 volts) to easier measure small capacitors. These units and most all of the vintage GR equipment are magnificent instruments so I hope your doesn't need too much work for calibration.

Finished cleanup and calibration of the 1650-A. The resistance levels on the ramp of the CRL rheostat was pretty far off. But definitely adjustable. It took me a while to realize that the internal DC is still required when applying an external frequency. A definite rabbit hole but once I noticed that even in an ext AC setting the DC batteries are still being routed through the circuit I new I needed to put the unit back in the case with the batteries. A beautiful instrument. Didn't really need it since I have a high end Sencore LC101 that goes up to an internal 600 volts and measures henries. As I said I've almost duplicated all you instruments. I just love the audio gear. I hope you come back to audio after you straying over to RF. Ha!

I am indeed back into the audio. I have been working for some time on how to make a home-brew setup to measure transformer L to a reasonable degree of accuracy. I can measure it with my GR instrument as you know so I use it to check my home-brew results. It is pretty tough to deal with inductors above 20 H. I am glad you got your instrument calibrated. I love the GR equipment.

So right.

I make my high voltage electrolytic capacitors by putting ten, 470 uF at 450 volt capacitors in series. This gives me 47 uF at 4500 volts - that some pretty serious high voltage and capacitance. If you used twenty capacitors of this same value, you would have 23.5 uF at 9000 volts... Wow... and the cost is pretty reasonable. You do need to shunt each individual capacitor with a resistor to equalize the voltage across each capacitor so that the high voltage is distributed equally across each one. The value of the resistors are not critical and can be anything between 220k and 1 meg at 2 watts each. This also drains the capacitors when the HV is removed. I mount the capacitors on an insulated board with the terminals pointing up in a convenient and logical manner to connect them in series and put a spot of Goop (buy at Home Depot, etc) on the bottom of the capacitors to hold them to the board. Goop is amazing in that it will hold very well but you can pry the capacitor off the board usually with no damage. That is my solution and it works great. Of course, I am sure you must realize, this builds a HV component that, when charged, will be deadly and can electrocute and kill someone or something that gets across a charged capacitor of this capacitance value and voltage level....

You are absolutely right. Components don't sound, amps don't sound, cables don't sound. Only speakers do. Want better sound, get better speakers. A small signal goes into the amp, a large signal comes out, with or without distortion. Want clear sound, get rid of distortion.