Calculating plate to ground capacitor for LPF in tube output stage

Shop deals on ebay
Advertise with us
shop deals on amazon
GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
T

tardishead

Well-known member
Joined
Aug 11, 2004
Messages
625
Location
Sussex, UK
I'm a little confused on this
Cutoff Frequency Formula F=1/2PiRC
In a tube output stage single ended before the output transformer I want to put in a LPF
My figure for R. Is this the tube stage's output impedance (reciprocal load resistance and plate impedance). Does the transformers reflected impedance have an effect? Does the resistance of the windings have an effect?
And what about for a cathode follower before transformer?

 
tardishead said:
I'm a little confused on this
Cutoff Frequency Formula F=1/2PiRC
In a tube output stage single ended before the output transformer I want to put in a LPF
My figure for R. Is this the tube stage's output impedance (reciprocal load resistance and plate impedance).
Click to expand...
Yes. But the reactance and parasitic capacitance of the transformer will alter the result.


Does the transformers reflected impedance have an effect? Does the resistance of the windings have an effect?
Click to expand...
They're all part of the stage's output Z.


And what about for a cathode follower before transformer?
Click to expand...
The cath-follower has a low output Z, so you would need a much bigger cap; the tube will probably not have enough drive capability at HF, so you'll have slew-limiting and HF distortion.

Anyway, applying passive filtering to a power stage is not a good idea; why don't you apply it to the prededing stage?
 
If there is no NFB then you can apply the LPF at the grid of the output stage. R is driving stage plate resistance in parallel with plate load resistor.

If there is NFB then why not do it in the NFB loop?

Cheers

Ian
 
Ok thanks
So the circuit I've been mulling over is the Neumann Sm69 tube version which gets round the treble pre emphasis of the capsule by adding a plate to ground capacitor before the output transformer. I'm trying to work out a figure for the output Z.
Tube is Ac701 and the cap is 470pf. For an output impedance of 25000 I calculate a frequency of 13000hz As the 3db point. This seems a little high . To make the frequency lower the output Z would be higher and I did not think that would be the case with a triode such as Ac701 and the resistance of the transformer windings.
Schematic attached below
And then I thought how could it be done with a cathode follower mic such as a Sony c37a or AKG C12a?
 

Attachments

  • Neumann-SM69.pdf
    167.2 KB · Views: 16
tardishead said:
Ok thanks
So the circuit I've been mulling over is the Neumann Sm69 tube version which gets round the treble pre emphasis of the capsule by adding a plate to ground capacitor before the output transformer. I'm trying to work out a figure for the output Z.
Tube is Ac701 and the cap is 470pf. For an output impedance of 25000 I calculate a frequency of 13000hz As the 3db point. This seems a little high . To make the frequency lower the output Z would be higher and I did not think that would be the case with a triode such as Ac701 and the resistance of the transformer windings.
Schematic attached below
And then I thought how could it be done with a cathode follower mic such as a Sony c37a or AKG C12a?
Click to expand...
Internal resistance of AC701 is about 8 kohms, so the actual output Z should be around 5k.
 
So LPF frequency for 5k with 470pf is 68khz. Whats that for? RF? Not to tame the capsule?
There is no feedback winding on the transformer for HF deemphasis from the K67 capsule.
On the SM69 manual there is a rise of 3db for the top end but not the 6 or more that the K67 throws up.
???
 
tardishead said:
So LPF frequency for 5k with 470pf is 68khz. Whats that for? RF? Not to tame the capsule?
There is no feedback winding on the transformer for HF deemphasis from the K67 capsule.
On the SM69 manual there is a rise of 3db for the top end but not the 6 or more that the K67 throws up.
???
Click to expand...
In fact, the figure I had for transconductance of AC701 is for a much higher current(4mA) than in the sm69 circuit. With about 0.4mA plate current, transconductance is bound to be about 3-4 times lower, so it is not impossible that the actual cut-off frequency was somewhere around 15kHz. Actually, there are too much unknowns (internal resistance, xfmr reflected Z) for any sensible calculation.
 
Right so I think I understand. Correct me if wrong
The tube is run at much lower current so plate resistance can be a lot higher than when operated at its "optimum" settings?
But i still don't get it
Even if the plate resistance is 30k and the output tx is a 12:1 ie 30k parallel - resistance is still only 15k. Add the primary resistance say 2k and you end up with a 3db point of 20khz. What am I missing here? Are tube microphone output transformers much higher resistance? If so how much? I guess they are made of much thinner wire and many turns in order to get the inductance.
 
tardishead said:
The tube is run at much lower current so plate resistance can be a lot higher than when operated at its "optimum" settings?
Click to expand...
Correct.


But i still don't get it
Even if the plate resistance is 30k and the output tx is a 12:1 ie 30k parallel - resistance is still only 15k.
Click to expand...
That would be true if the output was loaded with 200 ohms, but it is customary to load the mic with 1.5-2k, so the reflected impedance is probably about 100-200k. Actually, the loss resistance and the stray capacitance are probably dominant there. It is a big unknown since these figures are seldom published.
 
But some tube mic pres are true 200ohm devices no?
So that LPF is gonna be varied quite a bit if you hop between different preamps?
 
tardishead said:
But some tube mic pres are true 200ohm devices no?
Click to expand...
No; ever since the 1930's, designers recognized the need to bridge mics (not match).


So that LPF is gonna be varied quite a bit if you hop between different preamps?
Click to expand...
True; loading the mic with too low impedance will shift the roll-off towards ultrasonics. Too high Z results in increased HF resonance. So much that some designers create mic pres with very high input Z, because it generally results in a flattering sound.
 
I was kind of wondering about this regarding tube tracers. If your tracing a tube current with  a set plate voltage and a range of grid bias, it is only making the measurement against the tube plate resistance.  So the tube will behave differently for the same plate voltage and grid bias with a different load that a circuit will provide right? So unless the tube tracer has a selectable load, the measurement is only kind of useful?
 
bluebird said:
I was kind of wondering about this regarding tube tracers. If your tracing a tube current with  a set plate voltage and a range of grid bias, it is only making the measurement against the tube plate resistance.  So the tube will behave differently for the same plate voltage and grid bias with a different load that a circuit will provide right? So unless the tube tracer has a selectable load, the measurement is only kind of useful?
Click to expand...
Tube tracers measure current in precisely contained conditions. One has to learn how putting the tube in  a real circuit affects the operation.
 

Similar threads

S
6V6 Output Transformer NFB resistor mod / value
Replies
14
Views
670
ruffrecords
ruffrecords
S
Plate Reverb Circuit Feedback
Replies
18
Views
3K
Philbo King
Philbo King
V
Design an analog K-weighting filter
Replies
3
Views
2K
mbn
mbn
H
Edit: Haufe T14 is NOT ringing with the 6s6b tube (false measurement, my setup was not correct, sorry)
Replies
13
Views
1K
kingkorg
K
heatwalk
Fender 5E1 Preamp Output
Replies
3
Views
1K
heatwalk
heatwalk

Latest posts

Back
Top