Datasheets only specify maximum resistance value but I'm not using a resistor, my capacitor is directly connected and fixed bias is through a resistor
I want to know because I want to calculate for exact high pass cutoff for my speakers.
Heres my schematic.
Oh and while we are talking about it, this equation for high pass filter effect of coupling caps is correct?
C=10/(2*pi*grid_resistor*cutoff_freq)
In the case of my pre the grid resistance is 1 meg plus 15k correct?
Schematic here:
I want to know because I want to calculate for exact high pass cutoff for my speakers.
Heres my schematic.
An externally hosted image should be here but it was not working when we last tested it.
Oh and while we are talking about it, this equation for high pass filter effect of coupling caps is correct?
C=10/(2*pi*grid_resistor*cutoff_freq)
In the case of my pre the grid resistance is 1 meg plus 15k correct?
An externally hosted image should be here but it was not working when we last tested it.
Schematic here:
Last edited:
The grid resistance is undefined by definition, and at dc is only limited by leakage currents, so the input impedance is really the resistive component of your grid bias resistor appearing in parallel with the reactive component produced by miller capacitance, strays and inter-electrode capacitances.
IIRC that should be 1/......
Preamp: The input impedance of the follower is extremely high because the 1M grid bias resistor is bootstrapped by the AC voltage dropped across that 15K resistor, as a swag it could be equivalent to 10M or more - depends on the losses in the follower.
Power Amp: Note that in your power amplifier the schade style feedback is going to defeat your desired lf cut off, this should be done outside of the feedback loop, best really at the input of your amplifier. Also that 2.6K is far too low a load for a 12AT7, something on the order of 30K would give much better results.
I recommend using spice to double check one's calculations, and measurement of the working circuit to verify it works as expected.
IIRC that should be 1/......
Preamp: The input impedance of the follower is extremely high because the 1M grid bias resistor is bootstrapped by the AC voltage dropped across that 15K resistor, as a swag it could be equivalent to 10M or more - depends on the losses in the follower.
Power Amp: Note that in your power amplifier the schade style feedback is going to defeat your desired lf cut off, this should be done outside of the feedback loop, best really at the input of your amplifier. Also that 2.6K is far too low a load for a 12AT7, something on the order of 30K would give much better results.
I recommend using spice to double check one's calculations, and measurement of the working circuit to verify it works as expected.
Last edited:
So best to select interstage capacitors such that they are pretty much transparent to the signal and either use a input cap or the preamp coupling cap to create my filter?
perhaps seeing the rest of the bias supply design is in order, I think it might already have needed resistance:
I'm still learning about tube amps, these forums have been helping me alot
perhaps seeing the rest of the bias supply design is in order, I think it might already have needed resistance:
An externally hosted image should be here but it was not working when we last tested it.
I'm still learning about tube amps, these forums have been helping me alot
The formula you need is
C=1/(2*pi*relevant_resistance*cutoff_freq)
where relevant_resistance will typically be
grid_leak_resistor + output_impedance_of_previous_stage
and the latter is typically
anode_impedance in parallel with anode_load_resistor
The factor of 10 you had is sometimes used when you don't want the coupling or cathode decoupling to set the LF rolloff. If within a feedback loop then it all changes anyway, so you need to do some reading on feedback theory.
C=1/(2*pi*relevant_resistance*cutoff_freq)
where relevant_resistance will typically be
grid_leak_resistor + output_impedance_of_previous_stage
and the latter is typically
anode_impedance in parallel with anode_load_resistor
The factor of 10 you had is sometimes used when you don't want the coupling or cathode decoupling to set the LF rolloff. If within a feedback loop then it all changes anyway, so you need to do some reading on feedback theory.
This circuit is working? Without a plate resistor at 12AT7? hmmm... very curious...
this looks vaguely similar to the RH807 with paralleled Output tubes.
Last edited:
This is correct
I only referenced that diagram to show bias supply design
Looks like ill be trying to put together a sim model of my amp when I get some free time. I know less then I thought it seems
so right now I'm using too small a resistor, but what does that do compared to the correct one, just attempting to understand. Should the bias and cap be tied in before this resistor or its it right to directly connect cap
for my tube amp in general, my buddy nudged me in the direction of a beginning with his schematic and reference to a pre circuit and has mostly let me figure things out as I go
Last edited:
Having a grid leak of 470R, fed via 0.33uF from a 12AT7 anode (with 300R cathode degeneration) would give you rather low forward gain and rather high distortion! However, you are partially saved by the 22K pot. Your total grid leak resistance will be somewhere around 6K, depending on the bias setting. This is still far too low, but not quite so disastrous. The open-loop LF rolloff due to the 0.33uF cap will be somewhere around 20Hz, so OK, but you are throwing away gain and creating distortion by loading the 12AT7 so much.
You said it!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.