01A question

[mcandmar]

I think i answered my own question, i have been testing the amplifier with RMAA and seeing identical gain and measurements on both channels with the gyrators set to 3ma, and the bias voltages the same. Plates were reading 123v & 116v.

Ale if your reading this i need help, i think i have damaged one of my Gyrator boards.

I had both set to 3ma (measured across the 10ohm resistor R5), i went back to tweak it and one channel read 8ma. Didn't understand why so i quickly wound the adjustment down to 3ma again, and now i am seeing 75v on the plate which makes no sense. Double checked the bias & filament voltages are correct, and tested the tube which is fine.

Any idea what is going on? I suspect i may have damaged the BF862 in J2?

[nashbap]

Originally Posted by euro21

Anode voltage and bias voltage determines operating point (Ua, Ia at -Vg) ... at specified tube and at given condition.

If the tube changing (another tube, or aging), the operating point also changing.

There are no two identical operating point of two tubes, but if it not too far -relative to each other-, it's irrelevant.

The important thing is tubes gain -preferably- similarity at near operating points.

Do you prioritize matching gain over similarity in THD spectra or try to get similar THD spectra and then adjust the gain elsewhere, for example, at the voltage divider just before the SF?
Thanks . nash

[Rod Coleman]

Originally Posted by mcandmar

Theory question: I am using Ales Gyrator boards and Coleman filament regulators with cathode bias. What is more important to match, bias voltage or plate voltage?

With preamps, where the output swing is only a few volts, the DC Anode voltage needs only to be in the right range to yield a suitable bias level. So this is least important for tight control. The triode gm varies with anode current, and this is more relevant for good performance. So I would say aim for a matched anode current - and you can adjust the anode voltage and/or the bias voltage to get there, so long as they don't end up being wildly different.

[euro21]

Originally Posted by nashbap

Do you prioritize matching gain over similarity in THD spectra or try to get similar THD spectra and then adjust the gain elsewhere

If you measured enough large batch of -nearly- thousand years old tubes, you will realise, that selecting two "identical" tube is a real nightmare, it's almost "Mission Impossible". :-)

The traders/vendors measuring "pairs" with tube testers.
The average -good- tube tester measures tube at given operating point where gm near -or not so near- to datasheet's value in this point.

I shows, that in this point (changing from one tester to another, no "fix" point of this tube) the tube condition near to required value or not ("good"-"not good", "%" etc.).

It says nothing from distortions, which is a function of given curves (Ua-Ia at given -Vg) to relative datasheet curves.

If you a lucky, and own a curve tracer (uTracer, eTracer etc.) which shows curves, you can compare curves from tube to tube, relative to datasheet's curves.

It' a "real" pairing, but slow, labour-intensive, costly (do you have few bunch of hundred years old tubes?).

Instead of this, the gm measuring at given operating point is accepted.

After this you can do THD spectrum analysis, but I say in advance that you will not two "identical" spectra, even measuring tight "pairs".

BTW some years ago I measured a few bunch of 801/801a tubes.
The difference of THD spectra was larger (better-worse), than measuring -good, but not excellent- 801a/VT-62 tube versus 25/VT-25!

[nashbap]

Thanks very much for this detailed response. Coming from Mosfets it shows how little I know about tubes. Will be fun!
nash