Low output on JTM45 build

[tubby]

Maybe your Mullards are no longer like new?

Could be, at least they test as new on a tube tester with regard to emission and transconductance.

Just to be absolutely sure, I've just tested the amp once more with brand new JJ EL34s at 62% bias:
17,1 Vrms at the onset of clipping; 8R output into an 8R load. ~36 Wrms
17,6 Vrms at the onset of clipping; 16R output into an 8R load ~39 Wrms

Then another testrun with brand new JJ 6L6GCs at 60% bias:
17 Vrms at the onset of clipping; 8R output into an 8R load ~36 Wrms
18 Vrms at the onset of clipping; 16R output into an 8R load ~40Wrms

And once more with brand new JJ 6V6GTs at 70% bias:
13,8 Vrms at the onset of clipping; 8R output into an 8R load ~24Wrms(which is about the same power output I get with the Shuguang KT66s)

So the power output for El34s, 6L6GCs and 6V6GTs at the onset of clipping with 8R output into 8R load could be seen as ballpark.

However, what still doesn't make sense to me is the low power output of the KT66s with the 6,6k primary (~14Vrms=24.5Wrms) and
the unexpectedly small rise of power output of all of the above mentioned tubes (except 6V6GTs) with halved impedance / 3,3k primary.

Any more ideas?

@helmholtz: Checked the PI plate cap, it's actually a 100p.
I will scope out the grid signals with a new pair of 6L6GCs as suggested.
Checked the probe, calibration is fine, attenuator is set at 10x on probe and scope.
Confirmed scope measurements for output power with a TrueRMS DMM.

[pdf64]

Maybe early russian made TungSols? No longer available, though.

I don’t recall the Tung Sol brand being used by Mike Matthews for new production prior to 2000?
I bought these GTs around the early-mid 90s.

[Helmholtz]

I don’t recall the Tung Sol brand being used by Mike Matthews for new production prior to 2000?
I bought these GTs around the early-mid 90s.

I assume that means rather not.

Actually I meant the tube and not the label.
Until recently e.g. TAD offered a russian TungSol KT66 and I figured that there probably were'nt that many different russian KT66s.
(I happily own a few MO-V/GECs and will compare with Shuguangs.)

[g1]

I don’t recall the Tung Sol brand being used by Mike Matthews for new production prior to 2000?
I bought these GTs around the early-mid 90s.

Could they be early labeling of what they started calling GT-KT66HP, which were supposedly Saratov?

[tubby]

Attached are some the measurements. The file name displays the respective measurement & paramters.

[Helmholtz]

Attached are some the measurements, description is in the respective file name.

Pictures not showing.

[tubby]

With new Shuguang KT66 tubes. Bias at 70%. No NFB. Measurement points in the captions.

[tubby]

With brand new JJ 6L6GC tubes. Bias at 60%. No NFB. Measurement points in the captions.

[tubby]

WIth brand new JJ EL34 tubes. Bias at 60%. No NFB. Measurement points in the captions.

[Helmholtz]

I'd say signals look normal.
PI output signals are large enough.
Single sided clipping is probably due to PI imbalance.

So I think your main problem is the high Vsat of your KT66s, which makes the output clip well before the grid signals clip.
In other words, the actual tubes' plate characteristics don't agree with the datasheet plate curves.

[g1]

Not that it makes any difference, but it's strange how that scope is doing pk-pk to rms conversions. A squared off waveform should give rms more than .707 of pk, not less.

[Helmholtz]

Not that it makes any difference, but it's strange how that scope is doing pk-pk to rms conversions. A squared off waveform should give rms more than .707 of pk, not less.

Are you sure RMS values are too low?
Some readouts are blurry and leave room for interpretation.

[tubby]

I'd say signals look normal.
PI output signals are large enough.
Single sided clipping is probably due to PI imbalance.

So I think your main problem is the high Vsat of your KT66s, which makes the output clip well before the grid signals clip.
In other words, the actual tubes' plate characteristics don't agree with the datasheet plate curves.

Yes, on point!

My build seems to be fine after all and we found out the hard way (from all the measurements taken) that at least
contemporary Shuguang and JJ KT66s (these are the only KT66s I could test) seem to have Vsat issues and/or
don't match the original KT66 datasheets by a bigger margin than usual for modern production tubes.

As a first time KT66 user that came unexpected and I don't think many people are aware of this.
Personally, I've heard about it the first time here on this forum thanks to "Helmholtz".

Would be really interesting to compare the russian Gold Lions.
Then we would know about all KT66s in production at the moment:
Gold Lion and Tung-Sols from Russia, Shuguang/Valve Art/TAD/Tube Town from China, JJ from Czech Republic.
Did I forget any re-brands?

And of course a comparison to NOS GEC / M.O. KT66s would be in order. But who can afford those...

Thanks everyone for all the input, help and bearing with me through 80+ posts!

[Helmholtz]

As a first time KT66 user that came unexpected and I don't think many people are aware of this.
Personally, I've heard about it the first time here on this forum thanks to "Helmholtz".

I strongly recommend to read Prof. Martin Zollner's book "Physik der Elektrogitarre" (aka POTEG), especially Chapter 10.5.9 ff.
Also check the GITEC e.V. website.

[tubby]

I strongly recommend to read Prof. Martin Zollner's book "Physik der Elektrogitarre" (aka POTEG), especially Chapter 10.5.9 ff.
Also check the GITEC e.V. website.

Thanks for the hint.Very good reading, recommended!

Already on pg. 10-134 Zollner writes about the JTM45:
"Fitted with the TAD [Shuguang China KT 66 tubes], it struggles to climb over the 30W mark even when overdriven (and loaded with the nominal impedance),
while with the Tung Sol-tubes, it is o.k. to call it a “30-W-amp”. Only when overdriven, and with a mismatched load, it gets close to 45W."

Fits in perfectly with the findings in this thread.

Makes you wonder how most JTM45 kit manufacturers (and Marshall Amplification, for the JTM45 reissue) are claiming between 30 and 45W CLEAN for their amps,
even though they're fitted with the same Shuguang tubes.

Also, since I measured approx the same output power with Shug & JJ KT66, it seems the Russian KT66s are the best option on the market at the moment
concerning matching specs & power output compared to original GECs and the datasheets for those tubes. Will try those next once I need to change tubes!