EL34 Excessive Screen Dissipation

[Colossal]

7W wouldn't be enough, the Trainwreck amps use a 25W resistor and for good reason! I'm sure when I measured my Express the 1K choke resistor was dissipating more than 10W.

That 1k 25W resistor has been discussed many times and always the question of "does that 1k really need to be 25W?" and indeed it does, it gets quite hot!

[Littlewyan]

Just a thought, what frequency should I test the amp with? Is 1Khz best? As in theory, the higher the frequency, the higher the impedance and in turn the higher the screen current draw.

[martin manning]

Are you running into a resistive load? If so, I don't think it matters.

[Littlewyan]

I am, the amp sees roughly 7.2Ohms on my attenuator when used as a load. It was just a thought, I'll stick to 1Khz for now then.

[Littlewyan]

Fixed. Installed 2.2K 7W screen resistors and now at full whack the EL34s only dissipate 12.5W. It's still over spec (8W Max according to the datasheet), but a lot better than 21W! The resistors aren't dissipating too much either, only 3.1W. I also measured the wattage of the amp whilst I had the scope out, it gives out 40W clean and 80W full blast. Measured by sending a 1Khz 100mV signal into the input, scope on the output to find the largest sine wave I could without clipping and then measuring the speaker output with my multimeter. It's not an RMS meter, but given the readings I got I'd say it was close enough.

[martin manning]

Good to hear its been tamed. Did you happen to get quiescent plate and screen voltages after the 2k2 screen resistors were installed?

[Littlewyan]

I did. Plate voltage was 470V and screen voltage was 460V. At full whack the screen node sags down to 418V and the screens themselves down to 336V and pulling 37mA each. I can't remember what the plate voltage sagged down to, but it was between 420 and 440V I think.

[Roe]

perhaps 3k3 works better then

[Littlewyan]

Definitely. I only used a 6.6K OT as it was a spare I had on the shelf. Took it out of my Trainwreck clone after I got hold of the Pacific OT.

[JazzGuitarGimp]

It's not an RMS meter, but given the readings I got I'd say it was close enough.

All you need to do is get the peak to peak voltage on your scope, then divide by 2.828. That gives you the RMS voltage.

Ciao,
Lou

[pdf64]

It's not an RMS meter, but given the readings I got I'd say it was close enough.

All you need to do is get the peak to peak voltage on your scope, then divide by 2.828. That gives you the RMS voltage.

Ciao,
Lou

Isn’t that only valid for a sine wave; and the waveform on a screen grid may be way off that, especially in AB?

[JazzGuitarGimp]

It's not an RMS meter, but given the readings I got I'd say it was close enough.

All you need to do is get the peak to peak voltage on your scope, then divide by 2.828. That gives you the RMS voltage.

Ciao,
Lou

Isn’t that only valid for a sine wave; and the waveform on a screen grid may be way off that, especially in AB?

Yes, you are correct, but I think the OP was talking about measuring the clean output power with a 1KHz sine wave at the input.

[pdf64]

Oh yeah, I missed that, sorry
Actually that reminds me, I need to open up my old scope and try to reset its calibration, as when I tried to use a Vpeak reading from its trace, I realised the trace display was about 10% too high. Bearing in mind that voltage measurement error get squared when using it to derive power.