Russian Teflon caps

The green ones are k42y-2s. I couldn't tell you what the difference between the silver and the green.

 

That's one of the first accounts I've read of K40-y9 failure.

 

Thanks Greg.

Yes for me too. Never had one failed on me.

 

The one that failed on me was in one of my Dynaco MKIII's and only had maybe 10 hours on it. I was a little surprised for sure and at first did not even consider a capacitor failure. I had bought a whole box of them and since the failure I have not installed anymore just in case.

Gregb

 

Do you test them before installation? Maybe that one was leaking from the beginning and got worse over the break-in period.

 

That’s quite unusual, but certainly not out of the question.

I’m curious, did you test the caps at rated voltage prior to installation?

 

I see Brice and I are thinking along the same lines... and at the same time!

 

I have a Sprague Teleohmic and test all old caps before installing. I have even started checking new caps for value just in case as I had one marked wrong that I installed in a TV horizontal circuit and I lost the last few hairs on my head trying to find that problem.

Gregb

 

Interesting. Your tester is quite good.
I wonder if my aging box would have trigger the failure. I would think so if I let it run for more than 10 hours I guess.
Thanks for reporting your experience Greg.

 

The older style cap testers do sort-of test ESR, but they call it power factor. Its testing for the same thing, but the old PF check was usually done at 120 hz and ESR checks are done at much higher frequencies. ESR is much harder to pass, but for caps that only ever see 120 hz its good enough. Its not too useful for audio frequency coupling or for stuff used in switch mode supplies.

 

I didn't know that. Thanks Gadget.

 

Hi Guys,

I wanted to report on this topic.

- I "cooked" 8 x 0.22uf 600V Teflon FT-3 for over 200 hours in two batches at around 520 VDC, with white noise going through the caps.
- I checked the value and leakage after the cooking: nothing changed. Same value, 0 leakage.
- I Installed them into 4 mono blocs, and respected the polarity of higher voltage in the amps.
I am please to say that it worked !

The Teflon caps had their natural clarity from the beginning and haven't changed since.
No dip dull sound after 40 hours and slowly back up, which is a nerve wrecking experience.
I am listening to them as we speak. Superb sound.

So to my experience, it is very worthwhile aging the caps before than burning your tubes and ears for 200 hours

Cheers,
Brice.

 

Brice, why not share that setup that you built? I'm happy that it worked out for you. I'd be upset to hear something going down hill although from the experiences here it seems that you just have to grin and bear it for some hours and it comes back.

BillWojo

 

Hi Bill,
What I have done is a prototype really. Here is the detail.

- I bought a Chinese PS from Ebay, 12VDC to 1KV for a few mA.
- I bought a white noise generator kit from China also, 12VDC also.
- I had a small plastic box so decided to reuse it. Made ventilation holes.
- I remove the 100k linear mini pot from the HV supply and connected it to a "real" pot fixed in the box.
- I added a 0.1uF 600VCD caps, from the output of the white noise generator
- Added a switch to power the white generator board or not.
- Added a 3.9M resistor and a LED to the HV supply out, to indicate the presence of HV and to load the supply. This is very handy to have.

Here is the schematic and the pics of the boards.
I hope it helps
Brice.

 

the Dynaco MK3 has very high voltages especially if the 5ar4 rectifier has been replaced by a SSR. The initial voltage surge from that set up can be over 600v. If you couplling caps were only rated to 600vdc that may have been exceeded. Any cap with a 600v rating would have a hard time in that situation. I would not blame the cap without more info about the amp and its design.

 

Mullard GZ34 rectifier, no SSR ever in this amp. The caps were a matched set supplied with the amp. The amp is the Latino ST120, what more would you like to know.

Greg

 

It happens now and then with just about everything. I wouldn't panic unless you see a few fail. Film caps will usually support far more voltage than the rating. I don't know if they test them to 2X or more, but I doubt the Dyna would bother any that were up to snuff to begin with. The HV test before use is always a good idea if you want to be really confident. Losses can be specified as dissipation factor, power factor, ESR and a couple other ways. IMO, the losses of a Teflon cap are too low to measure with any common equipment. If you think you see something greater than zero in terms of DF, PF or ESR, it's almost certainly measurement error. I did a page on losses a while back- http://www.conradhoffman.com/cap_losses.htm

 

I agree. It seems that if the Latino ST120, has not been used for over 2 or 3000 hours, ripping out all the coupling caps when just 1 went bad is somewhat of over kill. Unless, you were looking for an excuse to try a different type of cap, in which case that would be a good time to do that.

 

Brice,what kind of white noise amplitude are you able to superimpose on the DC level?

 

Only 1 V at max amplitude, there is a miniature pot on the white noise generator.