Preamp with 2 compressor circuits, design discussion.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

ELS

Active member
Joined
Mar 15, 2023
Messages
43
Location
Intergalactic space
1679615431927.png
I rethought my preamp design and came up with this.
I didn't want to stress the power transformer I wanted to use too much so I changed the 1st compressor detection circuit to only be half-wave detecting, not full wave, this cut down 3 tubes to just 1 (although if I went for sand rectifiers I could've used just 2 with the original design)
Then I found out why classic preamp designs almost always use very low plate voltages, because it reduces noise, low anode voltage + low cathode current + then if you reduce heater power for some tubes it results in even less noise in these operating conditions... So I changed the power supply circuit so it's more adequate with this in mind.
I'm thinking of using an 02B shunt tube, but if not I put the altered resistor values on top for without the tube.

I want this preamp to work for a ribbon mic too, might be too little gain for that tho.
The 6N1P for the detection circuit isn't the most ideal but I have like 20 of them so I put that in there.

what do you think?
 
I would advise you to rethink your LED driver circuit for the opto compressor. The cathode follower driving them has a standing current of no more than 3mA and this is therefore the absolute maximum current that it can drive into the LEDs which is probably too low for this application. I have used an SRPP stage running at ab out 5mA for this purpose in the past. It has two advantages. First being a true push pull circuit it can dive 10mA into and out of the LEDs and secondly the stage provides gain. Having said that, driving an opto LED directly with audio is fraught with problems because the LED current does modulate the opto resistance despite the relatively long time contsants of the optoresistor. The result is high levels of distortion in the signal - I have measured in excess of 5% harmonic distortion at 1KHz with 20dB of gain reduction. You can avoid this by rectifying the signal in the normal way before applying it to the LEDs.

Cheers

Ian
 
I would advise you to rethink your LED driver circuit for the opto compressor. The cathode follower driving them has a standing current of no more than 3mA and this is therefore the absolute maximum current that it can drive into the LEDs which is probably too low for this application. I have used an SRPP stage running at ab out 5mA for this purpose in the past. It has two advantages. First being a true push pull circuit it can dive 10mA into and out of the LEDs and secondly the stage provides gain. Having said that, driving an opto LED directly with audio is fraught with problems because the LED current does modulate the opto resistance despite the relatively long time contsants of the optoresistor. The result is high levels of distortion in the signal - I have measured in excess of 5% harmonic distortion at 1KHz with 20dB of gain reduction. You can avoid this by rectifying the signal in the normal way before applying it to the LEDs.

Cheers

Ian
Well I don't need a strobe light for the LDR, ~50lumens should be fine
SRPP would need another tube.
 
Well I don't need a strobe light for the LDR, ~50lumens should be fine
SRPP would need another tube.
Where did that figure come from? It would probably help if you said which opto you plan to use. Most need around 10mA to obtain sufficient gain reduction.

SRPP would need a different tube not an extra one. Use a double triode instead of a 6F1P.

Cheers

Ian
 
Where did that figure come from? It would probably help if you said which opto you plan to use. Most need around 10mA to obtain sufficient gain reduction.

SRPP would need a different tube not an extra one. Use a double triode instead of a 6F1P.

Cheers

Ian
Double triode wont give me enough gain.
I'm not using an opto-coupler. I plan to use LED's zip tied to an LDR.
LED's at low currents manage around 200Lumens/watt, so at around 1/4watt it would give 50lumens.
This part of the circuit I plan to tweak a lot anyway, I'm most worried about the actual preamp part of it, so it doesnt sound like shit hopefully.
 
Double triode wont give me enough gain.

SRPP will give you 25dB of gain which is plenty for a comp side chain.
I'm not using an opto-coupler. I plan to use LED's zip tied to an LDR.
LED's at low currents manage around 200Lumens/watt, so at around 1/4watt it would give 50lumens.
Typical LED drops 1.75V which for 0.25 watts means approximately 140mA.

Cheers

Ian
 
Last edited:
what do you think?
I would question the use of G3-controlled vari-mu in a feed-forward configuration. because of the hyperbolic gain law, vari-mu gain cells are usually used in a feedback configuration, which regulates the compression slope.
I know there have been a few examples of feed-forward vari-mu compressors, but they are not common, mostly because they require thorough maintenance while tube characteristics change over time.
You have increased the problem by providing a DC path for the control voltage.
I think you should be expecting a number of experimentations for straightening out all possible issues.
 
I would question the use of G3-controlled vari-mu in a feed-forward configuration. because of the hyperbolic gain law, vari-mu gain cells are usually used in a feedback configuration, which regulates the compression slope.
I know there have been a few examples of feed-forward vari-mu compressors, but they are not common, mostly because they require thorough maintenance while tube characteristics change over time.
You have increased the problem by providing a DC path for the control voltage.
I think you should be expecting a number of experimentations for straightening out all possible issues.
How do you get compression ratios bellow infinity:1 then? if it's a feedback design, by using G3 to drop the gain, it will have nearly infinite:1 ratio.
by taking the signal from before the compression stage it allows me to reduce the ratio with the 2nd pot adjustment.
Tube drift wont be a very big issue, mostly it will affect the range of the ratio control, if the tube gets weak, the lowest ratio will be even lower, and the max ratio will start to become active always, the "-5" line should be adjusted so it's slightly bellow the voltage that the CF+diode start conducting.
Also my design allows for negative ratio. where the louder the input gets, the more limited the output is, not very useful but certainly could be fun.
 
Last edited:
Back
Top