12at7 clipping

[catalin gramada]

Hello. Have a 12at7 cathode biased at 2V with 470k grid leak. When I applied 1 kc signal I found the input signal it start to be clipped at 1.4Vpk around. Why? How can I get the right conditions to clip the input signal at 2Vpk please. Thanks.

[vintagekiki]

Please upload schematics for 12at7 with voltage and peripheral values.

[eschertron]

what grid stop value are you using? at7s notoriously draw current long before grid voltage reaches cathode voltage.

[catalin gramada]

This


blue- first 12at7 input
yellow - first 12at7 output

[catalin gramada]

Rising the bias voltage it helps to clip the input signal higher , but this in not the point. I choosed the -2V for a reason, I wanna clip the input at 2Vpk aso...

[Helmholtz]

Your schematic shows a 12AX7 at the input?

Do you use a cathode bypass cap?

What is the signal source impedance?

[catalin gramada]

12ax7 have no bypassed cathode resistor. I just checked, bypassing with 100u did not do any difference.
Signal source generator 600 ohm.
12ax7 plate impedance -120k around.(50k- bypassed cathode)
I will connect the generator directly to 12at7 input to see if matter

[Helmholtz]

So this is about the first PI tube?

You might try to wire a diode bypassed with a 1M resistor in series with the PI input.
That would clip your signal at around 2V before the diode, but won't increase grid signal.

[catalin gramada]

Yes first PI tube
Hey Helm, you're right. I drived the 12at7 input from generator and is properly drived. The generator is pretty low 600 ohm, I wonder what output impedance should be necessary to drive my 12at7 without a buffer input please ?
Is clear some current is nicked. Grid stopper or low input impedance source please?

[Helmholtz]

I actually don't understand why you want to get 2V at the grid.

A larger grid stopper will even lower grid voltage.
Best would be a cathode follower after the vol. pot.

[vintagekiki]

This

It is not clear from schematics which diode used in cathode.
If used a DUS or LED diode it makes sense why your PI works the way you described.
If you still stick at diode in the cathode, you must use current regulator diode.

https://www.theengineeringknowledge.com/introduction-to-current-regulator-diode/
Introduction to current regulator diode

http://archive.ampage.org/threads/0/ampdiy/011415/Marshall_LEDs__Diode_clipping-4.html
LED as current source/sink

Select the current regulator diode for desired maximum regulator current and maximum limiting voltage.
https://eu.mouser.com/Semiconductors/Discrete-Semiconductors/Diodes-Rectifiers/Current-Regulator-Diodes/_/N-ax1ml

Current Regulator Diodes uses Mesa Boogie in Mesa D180, Mesa Boogie Simul class 295, Mesa Boogie Mark 2 ... ...

[catalin gramada]

Hey VK It is a 5mm red led. Is not working as ccs but as a low impedance "voltage regulator" instead. Is like a full bypassed resistor. As time it draws through few mA the forward voltage is pretty steady 2.1V with a ac impedance around 5ohm at 12 mA. A common shared resistor in my p-p circuit work the same. Same voltage gain,. I checked the same with common 165 ohm resistor into cathode. Raising the bias voltage with 220 ohm resistor to - 2.6V makes the input signal to start to clip at 1.8Vpk around way before the 2.6V dc set. I replaced the 12ax7 in front with 12au7 8k plate impedance. It still clip before. The only way I can full drive it to bias set was directly from signal generator when input signal was clipped exactly at dc set.

[vintagekiki]

It is a 5mm red led.

Try briefly instead of a diode setting up cathode resistor , something like a Vox V100 to see how the PI behaves then.

[Helmholtz]

Nothing wrong with using an LED or several standard diodes in series at the cathode. That provides a fixed bias supply and low AC impedance.
The latter avoids gain reduction by current feedback as with an unbypassed cathode resistor.

The advantage of a bypassed cathode resistor,though, is better self-stabilizing of the operating point against tube variation, because of DC current NFB.

A current regulating diode would have high AC impedance causing lower gain and other problems with this paraphase PI.
But it might be interesting for an LTPI.

[vintagekiki]

Current regulating diodes provide constant parameters for the selected operating point unlike DUS or LED diodes.
https://eu.mouser.com/Semiconductors/Discrete-Semiconductors/Diodes-Rectifiers/Current-Regulator-Diodes/_/N-ax1ml