50w Plexi build -- odd operation

[mort]

This is a fresh 1988 50w plexi build using the chassis, PT and OT from an Ampeg VT-60. The amp is basically what's in the layout but with a 22k bias pot, no impedance selector, and added ppimv. Pretty straight forward. The amp works and sounds good but starts high pitch whining when the volume(s) are raised. If the master is low then you can dime the channel volume and vise versa, but raising both gets the whine going and the whining can vary in pitch. And turning up the unused channel's volume exacerbates it as well.

Here's a list of idle voltages. Note the high cathode voltage on one of the EL34's(1 ohm resistors are installed)..

V1
1. 141
3. 1.07
6. 199
8. 1.94

V2
1. 149
3. 1.17
6. 293
7. 150
8. 152

V3
1. 216.4
2. 27.5
3. 44.8
6. 214.9
7. 28.3
8. 44.8

V4
3. 490
4. 482
5. -34.3
8. .1

V5
3. 500
4. 490
5. -34.8
8. 29.6

[g1]

That would be 30 amps or 900W at that V5 cathode resistor, so either it's wrong value or there is open circuit somewhere.
But even the .1V at V4 cathode seems very high.
Have you tried disconnecting the NFB loop as far as the whine goes?

[mort]

ok the 1 ohm resistor was open. Never had that happen before... anyway, swapped it out and biased down and each got the cathodes down to .035~ each across the 1r. That's still too much, right? 35w per tube? And the plate voltage of course went up and is pushing 520. And that .035 was measured with the bias pot all the way closed so prolly need to get a 50k pot and put in there...

Am I getting this right and should something be done about the high voltage?

[mort]

Have you tried disconnecting the NFB loop as far as the whine goes?

Actually haven't even hooked the NFB up yet. I'll try connecting it and see if that helps.

edit: hooked up the forgotten nfb connection and that cleaned the whining right up. Can't believe I missed that

[mort]

Can someone help me understand the different voltage ratings on this data sheet? 2000v 800v & 450v

http://www.jj-electronic.com/images/.../el34_e34l.pdf

[nosaj]

Can someone help me understand the different voltage ratings on this data sheet? 2000v 800v & 450v

http://www.jj-electronic.com/images/.../el34_e34l.pdf

It looks to me like those are max values that should not be exceeded. EL34 have other uses beside amplifiers. I think I've seen them in power supplies before.

nosaj

P.S. Where in Florida are you?

[pdf64]

ok the 1 ohm resistor was open. Never had that happen before...

It happens, and can be quite useful in limiting tube current; hence I think that 1/2 resistors (with a flame retardant coating) can be beneficial in this application.

...got the cathodes down to .035~ each across the 1r. That's still too much, right? 35w per tube? And the plate voltage of course went up and is pushing 520...

No, 35mV across a 1 ohm resistor indicates that 35mA is flowing.
35mA x 520V indicates that the tube (plate + g2) is dissipating 18.2 watts, so not too bad (a bit cooler at such high HT may be better, depending on the OT impedance).

[mort]

P.S. Where in Florida are you?

between orlando and daytona

[Chuck H]

FWIW, and this may be considered blaspheme to some, but...

Many modern el34's are more tender to HV than their vintage predecessors. I had all my problems with the Rusky tubes. The Shuguang tubes were fine electronically, but a bit too microphinic for a combo type amp and that's pretty unacceptable for a new tube I think. Still, that's what's out there. Marshall type amps have a vague history of blowing screen grids. I think this is because they use a relatively low primary impedance and are often overdriven. When a power tube is clipping there are often high frequency/voltage spikes. Since impedance rises with frequency in an inductor it's possible for plate impedance for these spikes to rise above the screen grid circuit impedance and overdissipate the screens. Then POP! This is one of the major reasons 1K resistors are always recommended for el34's. IMHE the insurance of 1k screen grid resistors isn't sufficient all the time anymore. I would consider putting an additional 1k resistor in series with the choke. This would also "soften" the amp a little in the face of your high-ish volts (for better or worse I suppose). This would also drop the preamp voltages a little. You can bring the preamp voltages back up by reducing the value of one of the 10k resistors in series in the HV rail.

JM2C

[J M Fahey]

Can someone help me understand the different voltage ratings on this data sheet? 2000v 800v & 450v

http://www.jj-electronic.com/images/.../el34_e34l.pdf

Ua 800V is the classic "Plate voltage rating" found in most datasheets, it means "Power supply V+ must not exceed 800V" .
FWIW Philips published a 2 x EL34 , 100W RMS amplifier which was very popular here, which used 800V +V .

Now what many forget is that in push pull amps the output transformer works like an electricl see saw, when one plate saturates and almost reaches ground, the other reaches 2 X +V , some 1600V here.

And that with a resistive load; with a speaker which is inductive it may get way higher.

Hence the other (rarely seen elsewhere) voltag rating:
Ua0 2000V means voltage which must not be exceeded, even if tube is passing ZERO current, or it will arc inside.

Ug2 450V is the maximum *screen* voltage not to be exceeded.

The 100W I mentioned was split rail: 800V plate but 400V screen.

As I said, very popular in Argentina, when I was young every Club seemed to have one, very justified because it was loud, inexpensive, and mainly because because Philips offered, besides the circuit like anybody else, *full* winding details for primary and output transformers, which were relatively simple, I built and sold a lot of them.
Only "problem" was that original OT only offered a 500 ohm tap, because it was meant for large places and long distance line transformer speakers, I recalculated it for 4 and 8 ohm output.

[Chuck H]

Ah, yes, the data sheet... It's just a cartoon. That is, it resembles real life in a playful way but you can't take it literally or you might get hurt. Ok, maybe not that bad. But we have it on pretty good authority from one of our members that most of the rusky tubes are made by completely non technically trained labor remote from the brander/seller. That is, none of the factory employees have to have any clue how a tube works. only how to operate the machinery. No QC is done at the factory. At least that's what we were told. Any testing for relative goodness must be really poor because I've gotten MANY duds brand new from on line tube vendors (which apparently don't all test anymore either). I can't say that JJ tubes haven't been tested accurately to see that they actually perform to the spec sheet, but I can say that I've seen them pop within spec'd parameters many times. And isn't it interesting that with very little variance any new spec sheets look like regurgitated information from vintage spec sheets? The same member that offered the info on the rusky tube plant also indicated that there is nothing "wrong" with the healthy tubes leaving the factory, but they aren't really the actual tubes their model designates. That is, a new spec sheet should be made to indicate the actual specs of those tubes and perhaps even a new model number so that people know it's quite specifically NOT the model it's marked as.

I've been through five sets of rusky el34's from three different, known vendors in the last five years and only one lived in a healthy environment. I've been through three sets of Shuguang el34's from two of those same vendors and they all worked fine, but all were all microphonic.

I have to own up the fact that when I test my builds I eventually get to the "crank 'em up and see" part. Because I KNOW that will happen to them. Admittedly it's not protocol for vendors or even guitar amp tech repairs to test in this manner (though some smart techs actually do). So am I abusing the tubes? Nope. Such a test is typical of the environment in which they will be used. Some vendors will expect you to justify that. Mostly because they're assholes getting away with what they CAN instead of doing what they SHOULD. That's greedy and evil and they should be peeled. Rant over

YMMV

[eschertron]

That's greedy and evil and they should be peeled. Rant over

I read that as "they should be peed over"

[Chuck H]

I read that as "they should be peed over"

Close enough

[Randall]

I hear the Ruskis are actually into that in some cases.

[Justin Thomas]

I have to own up the fact that when I test my builds I eventually get to the "crank 'em up and see" part. Because I KNOW that will happen to them. Admittedly it's not protocol for vendors or even guitar amp tech repairs to test in this manner (though some smart techs actually do).

Yay! I'm smart!
Maybe my amps should come with a warning label: "Not Suitable For Bedroom Or Apartment Installation!"
Or, "No Polite-Society-Friendly Circuits Inside. Please Refer Usage To Proper Guitar Players."