6V6GTA plate dissipation in cathode bias push pull amp

[pchilson]

Hi,

I've been doing some caluclations and either I'm doing something wrong or something is amiss in the amp.

Amp - 1958 Ampeg M-12 Mercury
5Y3GT - strong
2 x 6V6GTAs - test strong
3 x 6SL7GTs - V1 is new, v2 tests weak but is the tremolo tube, V3 - PI tests about %80 of 100. All new preamp tubes coming this week.

Schematic calls for filter caps of 20, 20, 20. Cap can installed is 20, 20, 40. Lots of noise/hum out of this cap can. I temp installed 20, 20, 20 and now it is dead quiet.

Now the power tubes cathode resistor, schematic calls for 250R. It physically measures 256R. Close enough.

My plate dissipation calculations go thusly:

Cathode resistor = 256R
Voltage drop across cathode resistor = 34.2 vdc
34.2 / 256 = 0.1335 / 2 = 0.0667
Plate voltage = 347 - 34.2 = 312.8
312.8 x 0.0667 = 20.8 watts idle plate dissipation - yikes?

The 6V6s are not redplating at all. The amp sounds quite good but there is no clean headroom at all.

Am I doing this correctly or no? If so, what should I do? Anything?

Thanks

[pchilson]

Another question. I am going to order a new cap can for it this week. Should I go with the 20, 20, 20 as specified by the schematic or go with 20, 20, 40 as per what is in it now? Pros, cons?
I know the 5Y3GT spec for filter input cap is 20mfd but we all see this exceeded alot. What would you do?

[Bruce / Mission Amps]

I've had really bad luck with the multi-caps from CE Dist... I can't take many more chances on using them anymore so I use discrete components when ever possible.. zero failures since I've been doing that.
Yes, that idle current-power is way too high... but the other values seem OK... which leads me to believe your "2 x 6V6GTAs - test strong", are really going soft and need to be replaced.
Also, check your screen voltage.
Cathode biased amps seem to respond better when the screen voltage is a little less then the plate voltage... maybe 5 to 10 v less.

[pchilson]

Ok, I did some more measurements/calculations with the RCA 6V6s and some EH 6V6s I have on hand.

With the RCA tubes:

Plate = 338vdc
screen = 333vdc
cathode = 256R
voltage drop = 33.7vdc
33.7 / 256 = .1316 / 2 = .0658
338 - 33.7 = 304.3
304.3 x .0658 = 20.02 Idle plate dissipation

With EH 6V6s:

Plate = 337vdc
screen = 324vdc
cathode = 256R
voltage drop = 33.2vdc
33.2 / 256 = .1296 / 2 = .0648
337 - 33.2 = 303.8
303.8 x .0648 = 19.68 Idle plate dissipation

The only thing really different is the screen vdc.

Any ideas? What direction to move in?

[pchilson]

There is no one else is making a cap can like CE is there? I've used a few and didn't have any trouble with them but not on the volume you probably use. There seems to be a problem even getting them at present and has been for sometime. Are they having some type of manufacturing issue? I'm using discrete caps right now but would like to have a can just for "correctness" sake in the future.

[Rhodesplyr]

Most Leslie amps use the 30-30-30-10 capacitor, and I hear similar reports from other Leslie techs--that there's a higher failure rate with CE Dist. cans than with discrete caps under the chassis. More specifically, there is a higher failure rate with Leslie amps that use solid-state rectifiers. Apparently, the can caps don't like the voltage spike/inrush current of SS diodes. And, half the time, CE Dist. is out of stock of the particular can I want.

Still, there are instances where there are few other options due to lack of chassis space. In those cases, if possible, I'll use a discrete capacitor for the first filter stage and the can for the following stages to hedge my bets.

If you look closely at the labels of the can capacitors, you'll notice that most of them are rated 55 degrees C. You can get discrete caps with 105C ratings and better specs for less money than the cans.

What I **really wish** someone would make is a DIY twist-lok can capacitor kit. You'd wire up discrete capacitors yourself, solder them to a twist-lok base and then slide or screw a can down over them. I've pulled the old cans apart and re-stuffed a couple, but I wish there was an easier and prettier way to do it.

[MWJB]

" I'm using discrete caps right now but would like to have a can just for "correctness" sake in the future. " Manufacturers used cap cans because they could buy in bulk & keep costs down, get the cans made to order. Using discrete caps is the better option.

Are you measuring that 256 ohms from the 6V6 pin 8?

Have you replaced the cathode bypass cap?

[pchilson]

" I'm using discrete caps right now but would like to have a can just for "correctness" sake in the future. " Manufacturers used cap cans because they could buy in bulk & keep costs down, get the cans made to order. Using discrete caps is the better option.

Are you measuring that 256 ohms from the 6V6 pin 8?

Have you replaced the cathode bypass cap?

I lifted one leg of the cathode resistor and measured across it to get the 256R value.
I replaced the cathode bypass cap with a 22uF.

Something is funny with the board. I've replied twice to posts and they don't show up... This is my second time replying to MWJBs questions and my first reply to Bruce didn't showup either. Had to repost it as I am this one.

[tubeswell]

What I would want to know is - how can you be dropping 34V across a ~250R cathode resistor with a plate voltage of only ~350?

Did you try a different cathode resistor?

[jmaf]

There is no one else is making a cap can like CE is there? I've used a few and didn't have any trouble with them but not on the volume you probably use. There seems to be a problem even getting them at present and has been for sometime. Are they having some type of manufacturing issue? I'm using discrete caps right now but would like to have a can just for "correctness" sake in the future.

I use JJ multisection cans, have not ever had a failure. I also give them 50 volts headroom above unloaded voltage, if the voltage without any tubes in is 450, I use 500 volts. If it's 500 VDC, I use a 2 x 350 volt totem pole.

[pchilson]

What I would want to know is - how can you be dropping 34V across a ~250R cathode resistor with a plate voltage of only ~350?

Did you try a different cathode resistor?

I just tried a brand new Dale 300R in its place (only size I had on hand) and the voltage drop change was negligible. Down from 34.2vdc to 32.3vdc.

[pchilson]

I use JJ multisection cans, have not ever had a failure. I also give them 50 volts headroom above unloaded voltage, if the voltage without any tubes in is 450, I use 500 volts. If it's 500 VDC, I use a 2 x 350 volt totem pole.

Well the price sure seems right!!! But what do they look like? Are they gonna fit in visually with a 1958 Ampeg?

20 X 20 X 20 X 40 X 500 Volts
Price: $15.00
Description: 20 X 20 X 20 X 40 X 500 Volt Multisection

[jmaf]

Well the price sure seems right!!! But what do they look like? Are they gonna fit in visually with a 1958 Ampeg?

I have used the 40 + 20 + 20 + 20 and the 50 + 50 ones. They sound great so I guess they're fast enough and don't leak. And both fit in the LCR clamps on marshalls for example, I guess they're standard for current day cans. But for the '58 I can't say, never had the privilege of repairing one

They're black with red lettering. I've bought them here in the past: JJ Electronics and they have nice pictures there too, click on them on the details page.

Testing one here hotlinked: http://ep.yimg.com/ca/I/yhst-8476489..._2128_42666738

[pchilson]

Anyone tried the Weber cap cans?
They have 20, 20, 20, 20 and a 20, 20, 20, 40 both at 450vdc.

Should I get the 40 or 20 for the first stage? Will 40 harm the 5Y3 or not really?

[tubeswell]

Try getting the screen voltage down to 50-60V below the plate voltage with a bigger dropping resistor between the plate and screen supply nodes (i.e.: to be more like a 5E3 setup) and see what difference that makes