Attenuator High Frequency schematic

[kingkorg]

They look ok to me

Can you explain how this works, i can't wrap my mind around it. Wouldn't signal be filtered by 200Meg and 10nF? And signal also halved by the following voltage divider?

 

[vmanj]

View attachment 92214
How about this one？

I have a fixed bias and the cathode is connected to ground.
Unfortunately, your option does not work.

 

[oscarhuang]

Can you explain how this works, i can't wrap my mind around it. Wouldn't signal be filtered by 200Meg and 10nF? And signal also halved by the following voltage divider?

I also can't see how this circuit attenuates high frequencies.

 

[kingkorg]

I also can't see how this circuit attenuates high frequencies.

I guess feed the inverted signal from the plate to the capsule, but either i'm missing something or Khron might have been too quick on the trigger this time.

 

[oscarhuang]

I guess feed the inverted signal from the plate to the capsule, but either i'm missing something or Khron might have been too quick on the trigger this time.

But the 50M and 0.01uf capacitors seem to erase the signal.

 

[vmanj]

This circuit is very similar to the c12. Can it attenuate high frequencies?

I converted Cad Trion 8000 to ef86 tube with fixed bias.
I made it closer to the u67 circuit, but with a drain transformer (no feedback).
I also made several adapters from (7pin) EF95 to (9pin) EF86, 12A..7, E88CC to use different tubes.
I did the capsule connection as in u47 (only cardioid).
I liked the sound of the 12AT7 the most, so this circuit became similar to the c12 microphone.

 

[Khron]

Can you explain how this works, i can't wrap my mind around it. Wouldn't signal be filtered by 200Meg and 10nF? And signal also halved by the following voltage divider?

I guess i was indeed a bit quick to approve - i had ONLY looked at confirming the connections of the components i had suggested in my first reply in this thread, and my extension of that in post # 14, but NOT in conjunction with the "M49 option" (which first appeared in post #11 in this thread).

My bad...

 

[vmanj]

So without NFB (requires a xfmr), I only have the M49 option?

 

[Khron]

I'm gonna have to call a time out here - there seems to have been a chain of misunderstandings that have thoroughly muddied the waters here, and i'm much to blame

1) @kingkorg 's question in post #21 made me kneejerk-react (late at night) and think that he was referring to the split cathode resistor in the OP's schematics being some form of feedback in itself. Technically it might be (albeit positive feedback, since the cathode signal is in phase with the signal coming into the grid), but especially with the 50meg/10nF filter there, that's merely a grid bias network. He was actually referring to, i'm guessing, the +60V capsule bias, which i had never mentioned or touched upon.

2) He might have misunderstood something first (not guaranteed), since the schematic is drawn in a slightly odd way. Usually the anode resistor is vertical (conventionally), so it's clearer to see the current flow between the positive supply voltage (usually towards the top of the schematic) and the negative supply voltage, or ground (which is the case here), towards the bottom.

3) With my approval in post #17, i was referring to the two variants of schematic alterations the OP attached to post #15.

Just to clear things up even further, i'm attaching a quick simulation of a plain gain stage, one with a capacitor in parallel with the drain resistor, and one with a series resistor & capacitor in parallel with the plate resistor. I hope the node and plot names are clear enough.

Note the monotonically dropping response of the red trace (out3 in red), with the parallel capacitor; the shelf response of the series resistor-cap (out2 in blue); both compared to the unaltered gain stage (out1 in light green).

 

[kingkorg]

I still don't get how this works, how does hf get attenuated by bypassing anode resistor? I've never seen this arrangement, so it's kind of revelation to me.

 

[Khron]

I could be thoroughly wrong on this one, but an explanation that my simplistic brain could come up with and understand would be:

Since the gain is roughly defined by anode resistor divided by cathode resistor, tacking a capacitor in parallel with the anode resistor reduces the combined resistance as the frequency goes up, thus reducing the total gain of the signal.

Technically it's "impedance" though, i know...

Conversely, adding a capacitor in parallel with the cathode resistor (further) boosts high frequencies. But in such gain stages, the capacitor is often of high enough value that the corner frequency is in the single-digit or low-double-digit Hz range.

 

[kingkorg]

Thanks! It never occured to me you could do ot this way.

 

[Khron]

That was revealed to me years ago on the micbuilders yahoo group, when i asked about mods to the Schoeps circuit in order to tame the top end in K67 capsules (as in the Alctron HSMC001, my first "real" mic i had bought after lurking there for a while).

 

[Gus]

The cap across the plate resistor as drawn is very close to a cap plate to ground. This is assuming the power supply has a low Z output.

Run the sims again with the B+ power supply having an increase in output resistance.

 

[Khron]

Most tube mic power supplies i've seen have a last filter capacitor either right before the connection to the plate, or as the last element in the power supply itself. I'm ever so slightly skeptical of (worst case) a couple ohms of 10-15ft of cable resistance would make that drastic of a difference. Or would it really?

 

[vmanj]

With my microphone, a standard Chinese power supply.
At the end of the circuit there is a filter capacitor, a zener and a 1 meg resistor to discharge the filter capacitors.
Could this interfere with the Attenuator High Frequency filter (plate resistor cap bypass ) in some way?

 

[vmanj]

- Everything works great !

I like this option much more than the M49 version.
Under my primer (Cad Trion 8000) I have approximately the same resistor and capacitor values as in the image.

 

[Gus]

Most tube mic power supplies i've seen have a last filter capacitor either right before the connection to the plate, or as the last element in the power supply itself. I'm ever so slightly skeptical of (worst case) a couple ohms of 10-15ft of cable resistance would make that drastic of a difference. Or would it really?

Yes there is a last RC stage bypass cap(often in the microphone) to be more of a AC "ground".
I suggested running your sims with added non bypassed series resistance to "see" what happens as a learning experience.
A few ohms of cable resistance will not matter, capacitance might.

 

[Khron]

Series resistance in the supply voltage, I'm assuming? And how much would be reasonable / realistic?

 

[vmanj]

I tried different options for reducing high frequencies without feedback.
Everything works.
But I want to experiment with feedback the same way.

I drew a diagram more convenient for viewing.

If I understand correctly, then in my circuit there is only one option - is it the third winding of the transformer?