VCA question

[Martin B. Kantola]

Does anybody have knowledge on the Allison/Valley People EGC-101 VCA and how it compares to the TA-101 (and DBX-202 series)?

Grateful for any small piece of info.

Martin

 

[gyraf]

Look at e.g. the Gyraf/resources page for some info

 

[JohnRoberts]

[quote author="mediatechnology"]Here's the ECG-101 data that was sent to me along with the sample I received close to 30 years ago:

http://www.ka-electronics.com/images/Valley_People_ECG-101.pdf

He cites the Blackmer Multiplier Patent 3714462:

http://www.ka-electronics.com/images/Blackmer_Patent_3714462.pdf

Looks like to me that you could use a THAT 340 transistor array to make an ECG-101.

In the "where are they now" category here's Paul Buff's current website: http://www.paulcbuff.com[/quote]

Yup, Paul escaped the audio biz years ago... more power to him.

If you roll your own VCAs I'd be careful about taking the schematics literally, I think the diodes shown inside the EGC-101s may have been diode connected transistors (probably b-e junctions). Perhaps NPNs in the PNP bases and vice versa.

Yup, THAT arrays would probably work well. Might take 2 per if my speculation is correct about the diodes being transistors.

JR

 

[Martin B. Kantola]

Many thanks for the replies!

Thing is, I don't plan to build these, but bought an old Harrison console which was full of them (to my surprise!).

What I'm interested in is if it's worthwhile to replace them, because some people seem to have very good things to say about the slightly newer TA-101. Wonder how they improved compared to the ECG-version?

Looking at the Ben Duncan doc and the data sheet it's starting to look like a device I might enjoy the sound of. Control voltage thumps are not a big deal in slow VCA faders, and I can even stay away from mutes...

If the distortion is mostly 2nd order, and crossover distortion minimized, I might just keep these.

Martin

 

[JohnRoberts]

[quote author="Martin B. Kantola"]Many thanks for the replies!

Thing is, I don't plan to build these, but bought an old Harrison console which was full of them (to my surprise!).

What I'm interested in is if it's worthwhile to replace them, because some people seem to have very good things to say about the slightly newer TA-101. Wonder how they improved compared to the ECG-version?

Looking at the Ben Duncan doc and the data sheet it's starting to look like a device I might enjoy the sound of. Control voltage thumps are not a big deal in slow VCA faders, and I can even stay away from mutes...

If the distortion is mostly 2nd order, and crossover distortion minimized, I might just keep these.

Martin[/quote]

If you use decent opamps around them and trim out distortion I wouldn't expect them to color the sound that much. In compressor application there is often CV modulations going on from dynamic range manipulations that will IMO dominate sound character.

If you don't have access to a distortion analyzer you might be able to trim and/or compare VCAs with a simple null circuit. It will be useful to look at the null product with a scope and/or listen to it. In fact listening to the null without boost will reveal what coloration's are in fact caused by the VCA in proper perspective, while I always boosted the distortion products to hear them better.

JR

 

[Martin B. Kantola]

Thanks John!

When you say decent OP-amps, are there any special requirements to consider? The data paper suggests this kind of circuitry:

I can measure THD, will look into it as soon as I've powered up the mixer for the first time.

Martin

 

[JohnRoberts]

[quote author="Martin B. Kantola"]Thanks John!

When you say decent OP-amps, are there any special requirements to consider? The data paper suggests this kind of circuitry:

I can measure THD, will look into it as soon as I've powered up the mixer for the first time.

Martin[/quote]
Paul Buff probably used TL07x bifets which are fine as long as not driving low Z, I'd also make sure CC1 & 2 are decent film dielectrics. If you use other than stock opamps make sure the VCA is stable as there is voltage gain in feedback path shunted by CC1 and C to ground at input of OA2 that must be stabilized by CC2.

Harrison IIRC was also in or close to Nashville so they probably worked closely with Paul and did it right but it doesn't hurt to check.

Note: in the documentation posted by Wayne there is a distortion trim you might consider if THD noticeable, but it may not be trivial to cut into a tight PCB.

JR

 

[recnsci]

[quote author="Martin B. Kantola"]

If the distortion is mostly 2nd order, and crossover distortion minimized, I might just keep these.

[/quote]

These are class A afair, no crossover anything. And THD residual
should be as good as it gets in solid state world. CV thumping will depend
on how matched are transistors. IIRC class B dbx (and now That)
variety will outperform these quite a bit in noise department.

urosh

 

[Jim Zuehsow]

I have a whole rack full of GainBrain II and Kepex II and they still sound good to me and get lots of use to this day. The older ones use the EGC101 while the later ones use the TA101. I can't tell any difference in the sound. I also use a pair of 440 limiters on the 2 buss output and they use the TA101 and sound just fine. I also have 24 channels of dbx 316 noise reduction which uses the 202 dbx module. It too sounds fine. Used the right way all this old 70's and 80's gear does just fine, and many hits were cut using this same stuff. Are the newer THAT chips cleaner and better? I suppose, but I kinda like the stuff I have and am used to it and it's sound, but hey, that's just me. One thing I did do with all my old gear is to update the opamps to more up to date ones.
Back in the day, Paul Buff and B&B audio were having a pissing contest about which company had the better product. B&B had their 1537 attenuator chip, which is in a couple of Aphex CX1 compressor/limiters I have racked up. The mud was slinging fast and furious in the pages of RE/P. It was great fun watching this debate go on. My old Harrison 4032 has a bunch of Valley People and dbx vca's in it since I didn't add automation all at one time. It also has 24 Jensen transformer mic pres, and the remainder are transformerless using the TransAmp LZ pres. I don't recall very many people going with the B&B voltage controlled attenuator mod to replace VCA automation. When they became part of Aphex, I think the dropped the attenuator scheme and went to VCA.

 

[Jim Zuehsow]

The history of VCAs from THAT Corp. is interesting, and they do pretty much start at the beginning. They did leave out a few of the players along the way. There was the VCA from Melcor which was meant for console automation. Curtis Semi made a vca chip that was used mostly in electronic music synths. Along the same lines, SSSM had vca chips. They eventually became just SSM and were bought by Precision Monolithics, which in turn was bought by Analog Devices. Variations on the Gilbert cell were the common method until Blackmer's Gain Cell. Some of the early 2 quadrant multipliers and transconductance amplifiers have all been tried as vca's and worked to some degree. They each brought their own set of problems and their "character" to the signal.
The modern chips from THAT are very good at what they do and great progress has been made, but the older devices weren't that bad, and perhaps provide some of what we like about recordings from back in the 70's.

 

[Jim Zuehsow]

I've got all the old PMI books and the data sheets from CEM too. I still have some spare Curtis chips that went into an Oberhiem OB-8. I also have a bunch of the SSM chips which I used to make some compressors with. A couple tubes of SSM2016 mic pre chips too, which I may use one of these days.
ADI had a compressor in the application book which was very good. It used to be online also, but got taken down. I will scan that and post it one of these days.

 

[SSLtech]

[quote author="mediatechnology"]I'm about to post an entire SSM data catalog from the pre-PMI era. If interested, watch for it in an upcoming thread.[/quote]
I have a full catalog from... 1989 I think... is that of any interest?

Keith

 

[kvintus]

[quote author="Martin B. Kantola"]Many thanks for the replies!

Thing is, I don't plan to build these, but bought an old Harrison console which was full of them (to my surprise!).



Martin[/quote]


Hej Martin!
So you bought the Harrison you talked about when we last spoke! Congrats! I have to come and check it out sometime when I come to Vasa.

/Anders (jeppis)

 

[SSLtech]

Yes, and I want to know about the difference between the rev.2 and the rev.3 modules... I have the paperwork differences, but I want to know how they SOUND...

Keith

 

[Martin B. Kantola]

Making good progress on the console, it's so much fun to bring this one back to life. The sound of the EQ is indeed to die for!

Regarding the difference between Revs, got both version of modules here, so I can report on that later, but they have different VCA's so not 100% compatible, probably need to have the VCA at fixed unity gain for testing.


Which brings me to my question, would anybody know the control voltage for unity gain?

It's 0 Volts for the DBX 202 modules, but the ECG101 module is very different. The kind people at Harrison are trying hard to help me out. Got the schematic for the ECG101 as it's used in the console.

Martin

 

[SSLtech]

Sorry Wayne, -I missed your question...

The Version 2 and version 3 I'm referring to is not a VCA version, but a Harrison module version. I have some version 2's here, and I know that Martin has some version 2's there. He also has version 3's, and we share a mutual and highly regarded produer and engineer friend who owns a version 3 console, and who swears by the version 3's. I've just not heard the rev.3 modules myself.

Martin,

Are the VCAs current controlled? if so then the voltage for 0dB is going to be dependant on the resistance in the control line, which should be high-precision and low noise.

Are you going to be able to make it over here for that project in 4 weeks or so? I'm really looking forward to the idea of doing a side-by-side comparison!

Keith

 

[Martin B. Kantola]

Here's how the VCA is used in the Harrison. What I'm trying to figure out is simply which control voltage gives unity gain?

Now, unfortunately I don't have the schematic for the rest of the circuit, would need to know how the input and output are configured I guess. Could trace it from the modules...

Maybe I'll just measure input and output (of the whole thing) and set the voltage until I get 0dB?

Martin

 

[JohnRoberts]

From that schematic it's a bit of a trick question.. unity current gain for the VCA should be 0V DC at the control input Vcon. But the input resistor, feeding Iin is not shown. So if the input resistor feeding Iin happens to also be 5.1K then unity voltage gain will be at 0V, otherwise scale it for the ratio of Rin/5.11K.

It looks like a positive control voltage at Vcon will command attenuation.

JR

 

[Martin B. Kantola]

Thank you so much for the valuable help!

This will help me to sort out what's going on with that channel strip, at the moment there is far too much attenuation with the VCA fader set at 0dB. The control voltage is a sum of several sources since there are VCA groups and a group master fader, but at least I know now what the voltage sum should be.

Martin

 

[gyraf]

Re: the harrison schematic bit:

just in case you hadn't seen it, this is a current-in voltage-out setup - so you will need to add an appropriate resistor at the input to "scale" the gain correctly

Jakob E.