Tube compressor circuit adapted to tube preamp

Help Support GroupDIY:

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
Hello, everyone,

For my v76m project which I'm slowly navigating through the information, I am thinking about the possibility of incorporating a tube compressor after the first gain stage. I am looking through a number of designs and I think I would like to explore the Federal AM864 limiter using a 6av6 tube. The one main thing the AM864 differs from the V76m is that it has a center tap output transformer, which I don't know is necessary for gain reduction in the audio path. I looks like it is just being used for plate voltage for the V3 output amplifier but I am not sure.

It looks like I could take a tap off of the plate on EF804s on the second stage to feed the grid of 6av6 that is used for limiting on the AM864. From there I don't know how to get the signal from the 6av6 back into the audio path for gain reduction. Could I just add the voltage reduction signal back before the grid of the E83F on the V76m after the 25nF cap?

Thanks!

Paul
 

Attachments

  • Federal AM-864.jpeg
    Federal AM-864.jpeg
    87.6 KB · Views: 92

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
Potato Cakes said:
The one main thing the AM864 differs from the V76m is that it has a center tap output transformer, which I don't know is necessary for gain reduction in the audio path. I looks like it is just being used for plate voltage for the V3 output amplifier but I am not sure.
The output stage is a push-pull with differential inputs, which has the property of reducing significantly common-mode signals. That comes extremely useful for filtering out the sudden plate voltage variations of the input stage, when it's submitted to the gain control voltage.
You just can't avoid a differential stage here.

It looks like I could take a tap off of the plate on EF804s on the second stage to feed the grid of 6av6 that is used for limiting on the AM864.
For correct operation, the signal must be presented in balanced form to both grids of the gain cell (the 6AU6).

From there I don't know how to get the signal from the 6av6 back into the audio path for gain reduction. Could I just add the voltage reduction signal back before the grid of the E83F on the V76m after the 25nF cap?
You are so confused about how these things work I can't answer that question.
You must understand how this compressot works before trying to frankenstein it.
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
abbey road d enfer said:
The output stage is a push-pull with differential inputs, which has the property of reducing significantly common-mode signals. That comes extremely useful for filtering out the sudden plate voltage variations of the input stage, when it's submitted to the gain control voltage.
You just can't avoid a differential stage here.

  For correct operation, the signal must be presented in balanced form to both grids of the gain cell (the 6AU6).
You are so confused about how these things work I can't answer that question.
You must understand how this compressot works before trying to frankenstein it.

I suppose I should looking at other tube compressor circuits then. That and obviously I have more reading to do.

Thanks!

Paul
 

wlinart

Well-known member
Joined
Jan 19, 2012
Messages
165
Location
Belgium
abbey road d enfer said:
You must understand how this compressot works before trying to frankenstein it.
Not that i understand how this compressor works, but just wanted to add if anyone is trying to understand how it works, the original manual is amazing for that. There's a very deep explanation in there about everything (including a chapter on how to destroy it to avoid it falling into the enemy's hands  :eek: )
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
I've been looking around at different schematics and doing a bit more reading. First, reason I was trying to find a way to chop up parts from other compressor circuits and splice them into a tube mic preamp was my concern was getting a power supply to handle the heater current needed for eight tubes, but I think I will be more than fine with several PSUs designed for dual mono/stereo tube compressors, especially if I run the heaters at 12.6V. But instead of looking to make a unit with independent pre amp and compressor sections, I am looking to insert this between the first and second stages of a V76m as I also wish to add a passive EQ to make a channel strip. I found this schematic for a tube limiter circuit that is built around a pair of ECC83s, which seems more what I'm after. I would have to do some more reading to see if things like attack/release/etc can be implemented with this schematic.

Thanks!

Paul
 

Attachments

  • 12AX7 compressor.jpeg
    12AX7 compressor.jpeg
    16.2 KB · Views: 69

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
I've been looking around at different schematics and doing a bit more reading. First, reason I was trying to find a way to chop up parts from other compressor circuits and splice them into a tube mic preamp was my concern was getting a power supply to handle the heater current needed for eight tubes, but I think I will be more than fine with several PSUs designed for dual mono/stereo tube compressors, especially if I run the heaters at 12.6V. But instead of looking to make a unit with independent pre amp and compressor sections, I am looking to insert this between the first and second stages of a V76m as I also wish to add a passive EQ to make a channel strip. I found this schematic for a tube limiter circuit that is built around a pair of ECC83s, which seems more what I'm after. I would have to do some more reading to see if things like attack/release/etc can be implemented with this schematic.
Before committing yourself to this, I suggest you breadboard the thing and not put too much expectation.
This is the kind of limiter that was addressing the ham market, where fidelity is not a primary goal.
Time constants are governed by the cathode capacitor of the 2nd triode and I cannot imagine how to make attack and release variable independantly.
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
I'm still a ways from doing any testing as I'm continuing to look at designs and reading more related material to get a better understanding what is taking place with tube compressors. I have a very general idea but starting to work from a very limited amount of knowledge is what usually gets me in trouble.

I've also looked at Ian Thompson's Poor Man Tube Comp/Limiter for ideas. As is, I think it would work as far as what I am wanting to accomplish with this compressor (attack and release controls are desired but not necessary), but I would have to work a good way to implement the make up gain. Ideally, I would want to use the second stage of the v76m as make up gain, but I don't know if that can be accomplished by doing some alterations to the +5dB switch or if I will have to change some of the values at the EF804s of the second stage to increase available and then use a potentiometer for increasing gain post compressor. I realize there are penalties to pay for either course of action but to what extent I am not certain.

Thanks!

Paul
 

Attachments

  • PM Tube Comp:Lim.jpeg
    PM Tube Comp:Lim.jpeg
    184 KB · Views: 63

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
Ideally, I would want to use the second stage of the v76m as make up gain, but I don't know if that can be accomplished by doing some alterations to the +5dB switch
What +5dB switch? No such thing in a V76.
or if I will have to change some of the values at the EF804s of the second stage to increase available
What for?
and then use a potentiometer for increasing gain post compressor.
IMO, you don't need to seriously alter the gain, which is, if my calculations are right, 9 or 19dB.

Ian' s compressor can be inserted between the 2nd EF804s and the 3rd. You need to connect the output of the 2nd EF804s to the input of the potentiometer (no xfmr).
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
The v76m has a stepped +5dB switch in 1dB increments as well as a 1dB fine gain. The v76 does not.

Inserting the compressor between the 2nd and 3rd EF804s is what I had in mind. And if there is sufficient gain at the 2nd stage as is then I may see about using that as make up gain. I does seem I'm at the point of needing to start prototyping.

Thanks!

Paul
 

Attachments

  • V76m.png
    V76m.png
    921.8 KB · Views: 48

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
That is what I intended to do. This prompted my question/pondering regarding the possibility of needing to mess with the gain at that third EF804s, which it seems that I may not have to do so.

Thanks!

Paul
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
Found one more thing that may be closer to what I'm after (bottom half of attachment). It's a simplified side chain circuit using a 6AL5. I have seen this type of implementation in several of the push pull style compressors I've considered for this particular project. Since I will already have a two gain stages, all I am needing something to generate -DCV (the pentode show will probably also have to be kept). I can then add a control for make up gain going to the second stage. I would have to figure out how to do gain reduction metering. I would also have to figure out what I'm missing and why I'm probably way off on this particular option.

I do like Ian's simple comp/limiter, but basing circuits around the VTL5C/3s might not be the best idea as those are particular photocells are becoming harder to find. Maybe it could be modified to work with a T4B type photo cell.

Thanks!

Paul
 

Attachments

  • 6al5 comp 1.gif
    6al5 comp 1.gif
    33.2 KB · Views: 51

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
BA25 is an AGC, meaning the attack time is quite longer than needed for signal processing.
I believe a good start would be the Altec 436/EMI RS124.
You would need to address the issue of producing the two polarity-reversed versions of the input signal.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
It is very similar to the schemo you posted earlier. It was one of the "cut corners" I mentioned.
It relies on both halves of the balanced gain cell being strongly coupled via the common cathode resistor. However the balance is not perfect, resulting in thumps when the grid voltage is varied quickly, which is the case in a vari-mu comp. In the Hammond amp, it results in a slight unbalance of the positive and negative peaks, which is not a big issue, since it's somewhat compensated for in the subsequent stage, and anyway it adds some euphonic 2nd-order distortion.
 

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
1,877
Location
Nashville, TN
The RS124 is one have be been looking at particularly which led me to the unbalanced 6AL5 schematic. As you pointed out the thing I don't know how to address is the inverted signal as I will have an unbalanced at the input. The simplistic schematic to me looked like the RS124 without the B6C8 tube section and configured for unbalanced. Perhaps I would just have to add a transformer to create two signals issue and keep the B6C8. I don't think I can use simple semiconductors like diodes to invert AC polarity in an unbalanced signal that uses the chassis as the reference.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
I don't think I can use simple semiconductors like diodes to invert AC polarity in an unbalanced signal that uses the chassis as the reference.
You can't. You can use a xfmr, but you must make sure the preceding stage is capable of driving it correctly. Using a dual triode is possible but it must have an adjustable balance trim, in order to compensate differences between sections.
 
Top