Dynaco ST-70 Filter Capacitor

[Sal Brisindi]

Tonight I decided to replace the filter can with a new 4 section capacitor can I purchased on ebay. I am glad I did. 2 sections (20uf each) were open but the other 2 sections capacitance was higher than rated. I did not have time to check the leakage current of the other 2 sections (30uf and 20uf) but will do so tomorrow and let you know what I find.

Regards,
Sal Brisindi
http://www.tuberadios.com

[Shannon Parks]

Hi Sal, and welcome to the forums.

I have been able to get pretty remarkable leakage current results after reforming these quad caps and I recommend everyone to do so - even with a new cap. With all four sections strapped together I can get 500uA or less at 525V. See if yours is around that.

The trick in reforming is to keeping the current low - even lower than the 10mA some sites might recommend. Otherwise the heat impedes the process and you can't get to the best results. I set my 525V supply to 'constant current' mode at 1mA for all four sections connected together. Works like a dream. Maybe I should offer a free reforming service?

[Sal Brisindi]

Hi Shannon,
Thanks, this is a great forum to belong to. I didn't even think to check my new replacement capacitor to see the amount of leakage current it has or to reform it before I installed it. Its soldered in place now, I didn't do the smoke test yet as it was late last night and to early now... :-)

My original can is definitely bad as 2 sections were open. I took photos of the replacement so I can put it on my website. The photo of the can you will see crud on the bottom, it leaked.

Regards,
Sal Brisindi
http://www.tuberadios.com

[Shannon Parks]

I didn't even think to check my new replacement capacitor to see the amount of leakage current it has or to reform it before I installed it. Its soldered in place now, I didn't do the smoke test yet as it was late last night and to early now... :-)

You could still do this pretty simply. Just pull all the tubes except the 5AR4 - which you actually could also pull and add a single 1N4007 to get the max voltage for reforming. Then take a ~500V cap out of the junk box for the filter section in this jig (it's leakage current doesn't affect the test), and then run a ~200k resistor from it to a jumpered section quad cap. Once the voltage drop across the 200k drops to 100V, then switch to a 100k til the drop is 50V, then a 20k til the drop is 10V. Viola! Really easy to do this, and it gives piece of mind that the cap is at peak condition.

[dcriner]

...but the other 2 sections capacitance was higher than rated.

Hi, Sal. In the presence of some leakage current, many capacitance checkers get confused, and report the capacitance value higher than it actually is.

Another possibility is that the higher-than-rated value is within the original tolerance of the cap. 20% tolerance is common.

[Sal Brisindi]

Thanks for the info on the leakage affecting my capacitor meter.

Like I said earlier, the old quad can was bad, 2 sections were open. Of the other 2 sections, 1 had about 20ma of leakage, the other section was good.

Shannon, thanks for the tip, last night I was eager to try out my ST-70 and figured the new cap should be better than the old quad capacitor so I decided not to check it as I was given a pair of Dynaco A-25 speakers and wanted to hear them for the first time.

Well, all I can say is WOW! The amp sounds absolutely awesome with the Dynaco A-25's. I no longer have the low level hum and the highs were high and the lows were low! Those speakers rock! The amp definitely needed that recap, now all I have to do is replace the bias selinium rectifier and the 2 50uf capacitors. I didn't even try my new SED EL34's in it yet.

Regards,
Sal Brisindi