KT88 amp

[hylaphone]

Have you tried altering your load resistance?
And that GEC datasheet seems overly optimistic to me. Plugging your numbers into Nick's calculator gives around 60W with your values and 70V p-p signal applied.
Not 70W, but more than 50 anyway. I had it set for -46V bias and didn't re-try with the lower bias number (-41V).

What is Nick's calculator?!?

[Helmholtz]

What is Nick's calculator?!?

See here:

http://bmamps.com/ivds.html

[hylaphone]

Doesn't seem to make sense.
Something is missing. Maybe oscillation involved?

Can you post scope pictures of grid and output signals? And don't forget the scope settings.

Here are some.
1. Output max clean signal. 1V/div, 10:1 probe. 1kHz
2. Output max signal clipping about 75W, same settings.
3. Grid max clean signal, same settings.
4. Grid max signal clipping, 2V/div

I see no signs of oscillation, my scope goes up to 100mHz.

What's interesting to me is how the signal compresses, but I'm able to feather it back up... go slow enough and I can almost hit 60W clean, but this isn't the kind of dynamic I want in a bass amp

[hylaphone]

See here:

http://bmamps.com/ivds.html

Wow cool! Thanks --

[hylaphone]

Have you tried altering your load resistance?
And that GEC datasheet seems overly optimistic to me. Plugging your numbers into Nick's calculator gives around 60W with your values and 70V p-p signal applied.
Not 70W, but more than 50 anyway. I had it set for -46V bias and didn't re-try with the lower bias number (-41V).

I've not tried changing load resistance. I could halve it using a different tap/same load but 2.5K seems like an awfully high load for this amp?

[g1]

I've not tried changing load resistance. I could halve it using a different tap/same load but 2.5K seems like an awfully high load for this amp?

Yes, halving it would be too much I think.
But just to be sure, the tap arrangement on this model OT is not so simple. Are you using one of the wiring schematics from the attached datasheet?

[g1]

I'm not clear on the amplitude of those grid waveforms. 60Vp-p unclipped, 120Vp-p clipped?
Something odd there.

[hylaphone]

Yes, halving it would be too much I think.
But just to be sure, the tap arrangement on this model OT is not so simple. Are you using one of the wiring schematics from the attached datasheet?

Sorry, should have specified it's the "easy wire" version. Just measured the secondaries and it is wired correctly.

[hylaphone]

I'm not clear on the amplitude of those grid waveforms. 60Vp-p unclipped, 120Vp-p clipped?
Something odd there.

Indeed. Here is a shot at full clip. 2V/div, 10:1 probe.
Top trace is one of the grids, bottom trace is across the dummy load.

[Helmholtz]

Here are some.
1. Output max clean signal. 1V/div, 10:1 probe. 1kHz
2. Output max signal clipping about 75W, same settings.
3. Grid max clean signal, same settings.
4. Grid max signal clipping, 2V/div

I see no signs of oscillation, my scope goes up to 100mHz.

What's interesting to me is how the signal compresses, but I'm able to feather it back up... go slow enough and I can almost hit 60W clean, but this isn't the kind of dynamic I want in a bass amp

1. Means 52Vpp or 42W into an 8R load.
4. Assuming AC coupling, the positive going grid signal gets clipped above 20V. Looks like grid conduction but way too early.

Do both KT88 grid signals look the same?
Do you have another set of KT88s to try?
If not you may temporarily try a set of 6L6s

[hylaphone]

1. Means 52Vpp or 42W into an 8R load.
4. Assuming AC coupling, the positive going grid signal gets clipped above 20V. Looks like grid conduction but way too early.

Do both KT88 grid signals look the same?
Do you have another set of KT88s to try?
If not you may temporarily try a set of 6L6s

Here is a shot of both grids with a different pair of KT88. Second trace inverted for comparison.
Same settings, top trace is the grid of the "top" KT88 on the schem, driven by the non inverting half of the PI. 220K grid leaks are still in place.
Sorry for the blurriness, it's hard to photograph a trace!

[Helmholtz]

Here is a shot of both grids with a different pair of KT88. Second trace inverted for comparison.
Same settings, top trace is the grid of the "top" KT88 on the schem, driven by the non inverting half of the PI. 220K grid leaks are still in place.
Sorry for the blurriness, it's hard to photograph a trace!

Not nice, because signals are not balanced and look triangular instead of sinusoidal - but no top side clipping above 20V like in your picture #4 above.
Did you use same scope settings?
Can you confirm that different KT88s cause different grid signals?
Do you see the PI output imbalance and triangular waveshape also with KT88s out?

[hylaphone]

Not nice, because signals are not balanced and look triangular instead of sinusoidal - but no top side clipping above 20V like in your picture #4 above.
Did you use same scope settings?
Can you confirm that different KT88s cause different grid signals?
Do you see the PI output imbalance and triangular waveshape also with KT88s out?

Ok -- new day, new measurements. Got a little lost in the flurry yesterday.

The different sets of KT88 show identical behavior.

Here are pics with the KT88 removed. First pic is max clean signal, 2V/div, 10:1 probe. This shows 130Vpp from the top half of the PI, and 150Vpp from the bottom, inverting half.

Second pic is full clipping with KT88 removed. Both sides around 160Vpp.

[hylaphone]

With the KT88s installed, more careful measurements show it's worse than I thought.

520V on the anodes, 360V on the screens, -41V on the grids. 220K grid resistors.

First pic is the the grids at max clean signal. 1V/div, 10:1 probes. Top trace is the non inverting side, 35Vpp. Bottom trace is the inverting side, 45Vpp. This produces 35W output.

Second pic is full clip, 2V/div. Non inverting side 120Vpp, inverting side 90Vpp.

This makes around 70W output.

What do these waveforms suggest?

[Helmholtz]

First make sure that both scope channel knobs are in arrested CAL positon.

Ok, now adjust the PI input signal to get around 100Vpp at both PI outputs. (Does balance improve?)

Then lift one end of the NFB resistor and reinsert power tubes. What happens to the grid signals?