Roger Mayer RM57 build?

[RuudNL]

I would remove T4..T6 and increase the gain of IC2A.
The same goes for Q11..T3. This could be replaced with a single OpAmp stage.
We have better OpAmps nowadays...

 

[Vac11]

Yes it is possible of course... I just want to preserve "discrete signal circuit" as is and add balancing und debalancing of compressor in the simplest way from my point of wiev and experience... 

 

[Spencerleehorton]

I see you have added a section after the driver amp?

when i had mine working there was plenty of output, what is the point of this bit that you have added exactly?

 

[Vac11]

Pleas check here:
https://groupdiy.com/index.php?topic=29333.0

 

[Spencerleehorton]

Ah so your going from unbalanced to balanced?
I have a few line balance boards I can use left over from my old ssl 9 k builds.
This will lower noise? But might look at that design as it might me better.

 

[RuudNL]

Best way to achieve input/output balancing is with the THAT line receivers/transmitters.
In that case your inputs and outputs will behave like transformers.
With the 'OpAmp inverter' circuit, you will always get level differences when connecting inputs/outputs balanced/unbalanced.
When using the THAT chips, you can simply ground the unused 'leg' and maintain the same level.

For example: if you use the output unbalanced in the circuit shown, the level is 6 dB lower than when using the balanced output.

 

[Spencerleehorton]

Yeah, I use the that 1646 chips, have tried quite a few different types but the That1646 seem to be the best.
SSM2142 and DRV134 are others.

 

[Vac11]

Best way to achieve input/output balancing is with the THAT line receivers/transmitters.
In that case your inputs and outputs will behave like transformers.
With the 'OpAmp inverter' circuit, you will always get level differences when connecting inputs/outputs balanced/unbalanced.
When using the THAT chips, you can simply ground the unused 'leg' and maintain the same level.

For example: if you use the output unbalanced in the circuit shown, the level is 6 dB lower than when using the balanced output.

So which Thats  could you recommend for input and output balancing to preserve the unity gain?

 

[Spencerleehorton]

That1646

http://www.thatcorp.com/datashts/THAT_1606-1646_Datasheet.pdf

 

[Spencerleehorton]

I got a couple of opa transformers I could use on input and output, will see how they sound, I’ll wire up balanced to unbalanced for input, and opposite for output.
Will use 1:1 ratio.

 

[dirty1_1garry]

any ideas what meter should be? VU or mA?

 

[Spencerleehorton]

Ive used a couple of VU meters from an old GS3000 meter bridge, Fets just arrived so gonna fire up the iron and get soldering!!

Will post a video once ive got it working, fingers crossed!!!

 

[dirty1_1garry]

Ive used a couple of VU meters from an old GS3000 meter bridge, Fets just arrived so gonna fire up the iron and get soldering!!

Will post a video once ive got it working, fingers crossed!!!

Great! Will wait your news!

 

[Spencerleehorton]

Not good news so far , waveform all over the place, seems way to much signal, square wave on output , signs that compression is working but I need to now compare voltages to the spice diagram.

 

[dirty1_1garry]

Not good news so far , waveform all over the place, seems way to much signal, square wave on output , signs that compression is working but I need to now compare voltages to the spice diagram.

let me know if you need my emulation help

 

[Spencerleehorton]

The frustrating thing is that without the fets it was working fine.
I’ve used 22k attack pot, 100k threshold pot, Lorlin with 11 stage resistors going from 470k up to 10meg and 10k output pot with 330R down to gnd and 27k on other pin.
Will take a break then look again as I get the square wave on the 3rd transistor of the preamp.

 

[Spencerleehorton]

ok voltages looks like this in preamp, would be easier if you could change all your values for the same as the original so i can be specific.

Q11
E - 4.2v
B - 4.8v
C - 5.5v

Q12
E - 4.9v
B - 5.5v
C - 17v

Q13
E - 17.56v
B - 17.05v
C - 17.82v

for me something strange is happening as Q11 and Q12 emitter voltage is lower than the base voltage which should be 0.6v higher.
Q13 Emitter is 0.6v lower than the base, which is opposite to the others?

any ideas?

 

[RuudNL]

By the way: doesn't this thing produce a terrible distortion, or is this 'part of the sound'?
I mention this, because in most FET compressors, the gate of the FET is modulated to reduce the distortion.
As far as I see, no attempt is made to reduce distortion in this design.
It might be that distortion is acceptable as long as the voltage over the FET is low, but with a very low  (AC) voltage over the FET,  signal to noise ratio will get worse.

 

[Rob Flinn]

ok voltages looks like this in preamp, would be easier if you could change all your values for the same as the original so i can be specific.

Q11
E - 4.2v
B - 4.8v
C - 5.5v

Q12
E - 4.9v
B - 5.5v
C - 17v

Q13
E - 17.56v
B - 17.05v
C - 17.82v

for me something strange is happening as Q11 and Q12 emitter voltage is lower than the base voltage which should be 0.6v higher.
Q13 Emitter is 0.6v lower than the base, which is opposite to the others?

any ideas?

If Q13 is PNP then this is normal.

 

[Spencerleehorton]

ok, nice stuff there, considerations for mods and improvements then!!!

right, my second version is working how it should so far, getting fairly good signal and compression working how i thought it should.
Fair bit of noise present but i haven't used screened leads yet as waiting for delivery of some better thinner stuff than i've got.
Im going to look at voltages on this working one and redo my other one accordingly to the working one.

So far so good!! it really compresses quite well, haven't heard it yet but thats the next stage.