Oscillations problem help

[catalin gramada]

There is a simple test I can do to check interelectrodes capacitance with simple instruments please?

[Helmholtz]

There is a simple test I can do to check interelectrodes capacitance with simple instruments please?

Accurate measurement of capacitances in the 1pF range is tricky, even using lab equipment.

OTOH, effective grid to ground capacitance, which is typically between 50pF and 150pf can be easily measured with a decent LCR meter.

But I think that gm is much more relevant than interelectrode capacitance regarding oscillation.

[catalin gramada]

Well, even with new jj ecc81 I get osscilation again today. I found with 250V B+ it oscillate just at the beginning of the pot track, with 450V B+ it oscillate at the both ends of the track....if have any sense...I think I.ll given up, stuck with grid stopper and stop to mess with...

[catalin gramada]

It is in 150 Mhz range

[Helmholtz]

What about using a real audio tube, i.e. a ECC83? Much less gm but higher voltage gain.

Also you should lower your supply voltage to 330V max.

[Helmholtz]

It is in 150 Mhz range

Thanks, that's something I expected, Because relevant resonant capacitance should be below 100pF and parasitic inductance should be below 100nH.
One cm wire length corresponds to 10nH to 20nH inductance.

[catalin gramada]

Thanks, that's something I expected, Because relevant resonant capacitance should be below 100pF and parasitic inductance should be below 100nH.
One cm wire length corresponds to 10nH to 20nH inductance.

Thanks. What is the best way to stabilise please ? A stopper into the grid seems it do well the job...

[Helmholtz]

Thanks. What is the best way to stabilise please ? A stopper into the grid seems it do well the job...

I already made 2 proposals.

Never a good idea to exceed manufacturers' limits. And using a VHF triode in a typical audio circuit calls for trouble.

But there's nothing wrong with using a grid stopper. That's exactly what they are for: To stop parasitic oscillation.

[catalin gramada]

I already made 2 proposals.

Never a good idea to exceed manufacturers' limits. And using a VHF triode in a typical audio circuit calls for trouble.

But there's nothing wrong with using a grid stopper. That's exactly what they are for: To stop parasitic oscillation.

Yes , Roger that. I have a whole box and nobody seems to use in rf oscillators nowadays.Still it sound very nice in audio circuits. But it have very strange anode characteristics so supplied at high voltage makes it litle bit linear. Still max limits will not be overpassed: 300v plate, (it have 550v absolute limit for 0 current) Is not pretty clear for me why to not be supplied aso please ? I tried to set it for an decent operation point and the excursion will be always within limits.

[Helmholtz]

Yes , Roger that. I have a whole box and nobody seems to use in rf oscillators nowadays.Still it sound very nice in audio circuits. But it have very strange anode characteristics so supplied at high voltage makes it litle bit linear. Still max limits will not be overpassed: 300v plate, (it have 550v absolute limit for 0 current) Is not pretty clear for me why to not be supplied aso please ?

Plate voltage at idle doesn't matter. Plate voltage limit applies to the supply voltage of a gain stage, as signal plate voltage swings up to full supply voltage.
It is not an average but an instantaneous limit.

[catalin gramada]

That I allready pointed : If the plate stay at iddle at 120v and its excursion from my point of interest is 20 - 220v how can endanger the tube suplied at 450v as time it have 300v plate limit ?

[The Dude]

Even if you find a tube that does not oscillate without the grid stopper, I think it's good design to keep it in there. You don't know what tube the next guy is going to put in that location. IMO, it's better practice to design for variables.

[Helmholtz]

That I allready pointed : If the plate stay at iddle at 120v and its excursion from my point of interest is 20 - 220v how can endanger the tube suplied at 450v as time it have 300v plate limit ?

Believe me, the plate voltage limit means that the tube should not be used in circuits where the supply voltage exceeds that limit.
There's a reason why tube charts and examples are only provided up to 300V.
It means that the tube has not been tested/specified for higher supply voltages.

Exceeding the limit may cause to tube to behave unpredictable.

If the tube only needs to swings up to around 300V, why supply it with 470V?

[catalin gramada]

If the tube only needs to swings up to around 300V, why supply it with 470V?

Usually for linearity reason. How many tubes are linear along full supply swing ? In my opinion should be selective in respect with how much excursion it need...then plate voltage limit will be last problem , cause will be never exceeded...except unpredictable issues not related to the datasheet..
why a 12au7 may be used at 500v supply and a 12at7 not please ?

[pdf64]

According to the ‘grid stopper value >= 7/gm’ guideline, 100k may be more than strictly necessary, and being in series with the signal, will be increasing hiss.

So an overly high value of grid stopper seems a non ideal method of limiting bandwidth.

But an audio circuit oscillating at 150MHz is nuts - why on earth does it have any gain up there?
Limiting bandwidth by some means seems good practice - if the window is wide open, something may come flying in that you wish hadn’t!