varimu idea

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

5v333

Well-known member
Joined
Jun 30, 2013
Messages
631
Location
Gothenburg
i have been thinking if this could work.

classic varimu pushpull/symetric circuit.
without the output stage and its PP OT.
connect NFB to the cathodes via transfomer windings for lower distortion.
use the same feedback winding (buffered with high voltage fets) for sidechain circuitry.

in my head the circuit should use multiple tubes for better balance and to get lower plate resistance/output Z.

if it would work it would be great for the wallet and for simplicity...

thoughts?
 

Attachments

  • varimu.pdf
    11 KB · Views: 55
I don't know enough about the characteristics of FET's, but the source resistance to the rectifier is of vital importance as it affects the  fastest attack time possible.

For example, the source resistance of a Fairchild is around 100 ohms and this allows its sub millisecond performance.  An Altec 436 will only allow about 50mS attack.

A bridge rectifier will have a greater voltage drop than a pair of diodes.

DaveP
 
hi Dave!

perhaps we could use highvoltage bipolar transistors instead if a pair of fets wouldnt do well.


is there sufficiant dc to be generated from this kind of circuit is one of my two big questions. i guess the amplification from the tubes would be a little less than 10, and then negative feedback would eat most of that so  we would see unity gain (if we want) at the OT sec and also the feedback winding.
if we inject one volt ac at the grids we would get a maximum of 1.41V DC from the side chain (minus diode drops). how far do you get with this?
of course with higher input signal we can generate higher DC signal...
maybe we can make the feedback not eat all the gain away, leaving both a little more signal to make DC from and also have a little gain on the output that can be attenuated with a makeup pot.


the other question is about the negative feedback. will it have any problem with the bias voltage changing all the time? will it compensate and defeat the whole purpose?
 
The choice of tubes would be very important.

You would need tubes that can be run at higher current and they would need to be "semi-remote" types.

A tube like 6SK7 needs to go down to -70V for full gain loss, but a 6BZ6 can operate OK with -5 to -10V.

I think feedback would try to compensate for the gain loss, so it's a no-no.

There have been older designs that have used cathode followers for isolation and current gain,  these could operate from the plates so the voltage would be much higher.  a twin triode like a 12AU7 or 12AT7 would be fine.  The low output Z from a CF would give a faster attack time.

Best
DaveP
 
5v333 said:
i have been thinking if this could work.

classic varimu pushpull/symetric circuit.
without the output stage and its PP OT.
connect NFB to the cathodes via transfomer windings for lower distortion.
Remember that using NFB linearizes gain, which is contrary to what you want in a gain cell. You would end up with the NFB fighting against the control voltage to maintain gain.
That's the reason why NFB is never applied around the gain cell, although it would be tempting in order to decrease distortion of the gain cell.
 
Yeah now i realise that when compression kicks in, the amount of feedback gets reduced in the same time so that less negative voltage feedback takes place...

In the trashcan it goes.
 
This is about as simple as it gets from 1954:-
2utru49.jpg

DaveP
 
Someone here built that one a long time ago, there used to be pictures.  Very low output as a single stage.  NBC had a very similar earlier 3 tube circuit for 'supplementary gain control'. 
 
EmRR said:
Someone here built that one a long time ago, there used to be pictures.  Very low output as a single stage.  NBC had a very similar earlier 3 tube circuit for 'supplementary gain control'.

Hm, I wonder how low. What is that from? I want to snag the schematic and maybe toy with it.

All the best,
 
It was posted by cannikin on this thread, Doug has a good memory ;D

https://groupdiy.com/index.php?topic=27226.msg390187#msg390187

DaveP
 
It's just a simple rotary three-way switch, not drawn in the conventional way.

1st pos. n/c  (limiting)

2nd pos.  balance ac

3rd pos.  grounded (off)

DaveP

 
Ok thanks!

So the thread mentions the balancing switch makes it a linear amplifier with no limiting???  What is it doing? A type of feedback?? Is it still a limiter in this mode???

What is that 6.3v...... where does it come from? Is that related to the heater supply????
 
> the thread mentions the balancing switch makes it a linear amplifier with no limiting???

What? Where?

(That's what the "Off" position does.)

"Balanced" injects a signal into both sides. When balanced, this cancels. For cheapness the signal is just 50/60Hz heater power.
 
PRR said:
> the thread mentions the balancing switch makes it a linear amplifier with no limiting???

What? Where?

(That's what the "Off" position does.)

https://groupdiy.com/index.php?topic=27226.0

The first post but maybe I misunderstood

PRR said:
"Balanced" injects a signal into both sides. When balanced, this cancels. For cheapness the signal is just 50/60Hz heater power.

Ok.... . I just don't get the balancing part yet ..... 

It is meant for some kind of calibration ???  Meaning is it a temporary setting????
 
The main problem with vari-mu types of compressors is "Thump".

This is when a large signal arrives that needs a lot of reduction and consequent voltage change in the vari-mu tubes .

If they are not balanced, then a very low frequency thump can be heard when this happens, so a means of balancing is provided.

If they are in perfect balance then the +ve and -ve parts of the signal cancel in push pull, this occurs in the interstage TX.

A low voltage ac signal (derived from the heater supply) is switched to the input in the 2nd switch position marked balance, so that  you can adjust the tubes until the ac hum is inaudible, the tubes are then in balance.  The same facility is provided in the RCA  BA-6A

When the tubes are balanced, you turn the switch back to Limiting.

DaveP
 
new try!

i removed the whole negative  feedback thing and made a new threshold circuit.

sidechain is fed back from half of the OT sec as before. as the sidechain is fed back after the PP OT, the sidechain is free from even harmonics and should offer a cleaner compression signature.

the threshold pot injects positive voltage through the feedback windings to back bias the diodes so they conduct at higher or lower signal.

i wonder if the sidechain circuit will load the OT in some strange way or even start modulations when retfication is going on and off.

your thoughts once again :)
 

Attachments

  • varimu.pdf
    11.3 KB · Views: 19
There is not enough gain in a 1-stage amplifier to give substantial compression.

What are you trying to do?? Save a buck? I would assume many of the cheaper legacy designs were as cheap as the designer felt he could get away with; and that he probably had more think-time on the issues than you do.

Plagiarize!!
 
Back
Top