Vari mu compressor with 6V6 tubes

[bluebird]

Here is a sweep I took of my poor man 660 hack with eight 6BA6's

input level = +10dbu

GREEN = freq sweep no compression (+10 added gain from compressor)
YELLOW = THD+N no compression

RED = freq sweep 20db of compression
PURPLE = THD+N 20db compression

 

[bluebird]

So not that great at -20db GR. The big sweep up in THD around 30hz is my side chain pooping out.

IMHO to get really down in the -20db range with a single gain stage (660 topology) your gonna need remote cutoff tubes in numbers.

I WAS WRONG

After looking at the sweeps I went back to the unit I have the two 6V6's in and started messing around some more. I found another set of old 6V6's (more Philips) that produced close to the same specs as the 6BA6's but more gain was available.

I think with some more tweaking, perhaps another couple transformer winds and a better sidechain transformer, I can get the GR down to -20 and have it be usable.

I just don't have many 6L6's like I had 6V6's laying around. I am going to have to look into current off the shelf 6L6's and see if I can find a brand that has consistent performance in this type of circuit. I'll try some sovtek 6L6's or maybe  I'll just drop the cash on the Tung Sol 5881's.

Hey thanks for leading me down another vari mu path!

 

[Heikki]

I limited the max gain reduction to about 15dB in the sidechain because at more than 15dB gain reduction around +10 dBu output was where the distortion became audible when listening music through the compressor. At 0dBu output with 20dB gain reduction with 6L6 tubes the distortion should be tolerable and maybe even better than Farichild.

In the Fairchild 670 manual there is IM distortion vs level of gain reduction graph at various output levels (I'm assuming time constant 4 was on when they measured those IM levels). In a Fairchild distortion start to rise fairly quickly after 16dB of gain reduction even at 0dBu output. At +8dBu output the rise in distortion is very quick which is similar to my compressor.

I'll have to look if I can find my old notes where I made measurements and constructed tables just like in that German pdf. If I remember correctly my compressor always had lower distortion than Fairchild 670.

 

[bluebird]

Interesting, I never made IM measurements, only THD+N. Since the default signal output of the AP is +4dbu thats probably what I was making the -20db GR measurements at. I'm not going to argue with the original manual, I'm just not familiar with IM distortion and what is and what is not audible. I know what THD sounds like coming up on 5%.

I'm not sure if 4:1 means pos 4. It might mean 4:1 ratio? Could be those measurements are pos 1. In which case Pos 4 would have significantly better figures.  Seems more useful to make the measurements at worse case Pos 1. Then the user can assume better specs at all other settings.

I'll make new measurements on the units here at work when I have some time.

 

[Heikki]

4:1 is the ratio of signals in IMD measurement. If fast release would have been use in those measurement I would expect IMD to be a lot worse. I'm just guessing they used slow release to make the measurements look better.

 

[bluebird]

Ahh I see. I'm not sure if companies were like that back then. I was talking to a friend about it this morning. I don't know how much competition there was for peak reduction compressors made for cutting back then. I would think they would care about the technician knowing the limitations of the equipment more than marketing. You know, like its supposed to be

 

[emrr]

Historically I would expect slower release times.....as I supposed you would to.  Especially for measurement quotes. 

PRR and others way back reiterated the technical argument for a 50K transformer or choke load on a single 6386 for lighter AC loading and reduced DC swing, which has always seemed a benchmark to me.  The Fairchild repro iron is also 50K-60K for use with a 4 tubes.  Yet, we see plenty of instances of 20K resistive loading in 6386 circuits too.  All capable of 20+ dB with useful sound and slight distortion profile.  All thoughts in this paragraph relating mainly to typical low level operation with make-up amp stages, versus the single stage at hand, though I'd think offhand lighter loading would prove useful at elevated single stage levels as well. 

Watching with interest. 

 

[bluebird]

Well with the two power pentodes, a lower ratio transformer is definitely the key here.  And a resistive load of some sort is needed to tune the performance in deeper reduction. Something under 600 ohms.

 

[Habibi]

I may be missing something blindingly obvious here - but I do not see any sidechain rectification in the OPs schematic. What am I missing?

 

[scott2000]

Would be nice to hear this particular unit's side chain process explained in simple terms.

 

[PRR]

> side chain process explained in simple terms.

What part? All of it?

The time-constant parts are mostly not shown. Re-draw with those included and sleep on it.

Rectification-- Just look at Q2. What happens at no-signal? What happens when R7 goes positive? What happens when R7 goes negative? Then Q3's action should be apparent. And the need for "cathodyne" Q1. Q4 is a buffer.

 

[alexc]

One things I am sure of ...  fc660 style done super cheap  ....  with 8x 6K4P tubes ... and an Edcor 8K:600 10W op traffo ....  works very well at the 20dB GR  mark 

And driving directly into a 6L6 monobloc (also an Edcor)  and hifi speaker, the sound is magnificent  8)

.....

I would like to try the  'Heikki side chain'  in a 6L6  GR amp      ....    going straight to the speaker appeals to me ...  and the 6L6 is a real fav of mine, sound wise.

I think 6 to 12db GR at decent listening  volumes would be about right  ....  even parallel push-pull 6L6 is doable  8)

I find the modern tung-sol 7581a with it's extra couple of watts for  'low-thd headroom'  is killer for the price (and heater requirements).  Can't help but wonder what 4 off would do in a ppp GR amp driving a big ass Tannoy !

... anyone with pcb ?           

 

[capacitor]

Hey all! I know it's been a few years since the last post. The attachments no longer appear and I was curious how this is implemented. Does anyone have a copy of the schematic (or a link to where this is discussed further)? This looks appealing!

 

[scott2000]

Could possibly try waybackmachine for some bits of the info until Heikki or others chime in.

Also, not sure if you've seen this yet....

https://groupdiy.com/threads/transformerless-vari-mu-compressor.75663/
https://groupdiy.com/threads/transformerless-vari-mu-compressor-build-thread.78043/

 

[capacitor]

I was just going to edit my post to say I tried the Wayback Machine but without success

Found the schematic at http://www.ghr.fi/supercontrol.pdf and am looking it over. Really cool. I was looking at the transformerless Vari-Mu earlier which looks fantastic. I'm thinking of this unit as more of a color device.

It looks like when I navigate to www.ghr.fi, the server automatically redirects the browser to Variable Mutual Contuctance Compressor (edit: from www (dot) ghr.fi to www (dot) ghr.fi/varimu). I can see the main site page for just a tantalizing instant, which probably has links to the SuperControl info proper. I did pick up a directory that has other stuff in it.

I'll reach out to Heikki as well.

 

[Heikki]

Good that you found the schematic. I think all the info there is on Super Control is in this thread. There was only a webpage which had only some specs and nothing else.

I have designed new version and new boards for this but I haven't built it yet.

Schematic
https://www.ghr.fi/proaudio/6L60/6l60.pdfBOM
https://www.ghr.fi/proaudio/6L60/6l60.ods

B+ power supply
https://www.ghr.fi/proaudio/6L60/660ps.pdfBOM
https://www.ghr.fi/proaudio/6L60/660ps.ods
I can send you the boards for this new version for free if you are brave enough to try untested desing.


Edit: layout
https://ghr.fi/proaudio/6L60/layout.pdf

 

[capacitor]

I am, and I would be glad to build the updated design and provide feedback and measurements. I've done enough tube work to not put myself (or the parts) in danger, haha!

Edit: just sent a PM but let me know if I should send contact info another way. Thank you!!

 

[Tubetec]

I remember reading somewhere , the grid volts required to drive a 6v6 to full power if applied to the grid of a 6L6 gives a similar output power at a very much reduced distortion level .

 

[capacitor]

Imho, brand/model/type tube rolling, in order to get a certain effect, is where things could get really interesting with this design. These could be signature pieces in someone's studio.

 

[capacitor]

Just a quick update. I have the boards in hand for the 6L6 version of the SuperControl. I'll finish collecting parts and be able to work on it in earnest after the holiday. The boards look fantastic. Thank you, @Heikki !