OPTO specs believable??

[jaddie]

I'm no Ian... but I suspect by "sand" he is referring to silicon, as in solid state electronics..... or not.

JR

Ok, that makes sense in the context of the statement. I do understand that people get silicon, silica, and silicone all jumbled up.

 

[ruffrecords]

Ok, I understand the overall desire. I'm a little puzzled by the desire to keep "sand" (what is that, distortion?) out of the signal path when the distortion characteristics of optos are so poor.

By sand I mean semiconductor

Page 10 of this pdf, for example, details it. Are the new ones better? Because every modern VCA certainly is.
http://denethor.wlu.ca/pc300/optoisolators/analogoptoisolatorintroduction.pdfhttp://denethor.wlu.ca/pc300/optoisolators/analogoptoisolatorintroduction.pdf

That is a very interesting document. Despite lots of searching for documentation on the use of Vatrols I have never come across it. Thank you for posting it

So...operating them with very low audio voltage with lots of makeup gain?
Or maybe that's "color"?

Most of the time, the signal across the opto will be small. For example, say you have a 2:1 compressor and set the threshold to -20dBu. You set the gain make up so that with 0dBu in you get 0dBu out. Since at 0dBu in you are 20dB above the threshold, the compressor should reduce the gain by 10dB and the make up gain will be 10dB. At this point the signal across the opto will be only -10dBu. In a typical implementation there will be threshold and gain make up pots in parallel with the opto so it is not unusual for there to be a 6dB loss with no gain reduction so the actual gain make up would be 16dB and the level across the opto would be -16dBu.

The worst case for Vactrol distortion is when there is no compression at all. Perhaps this is why some people like to run there signal through opto compressors with no compression.

Cheers

Ian

 

[jaddie]

That is a very interesting document. Despite lots of searching for documentation on the use of Vatrols I have never come across it. Thank you for posting it

Back in the day when Vactrols roamed the earth that was easy to find. Part of the Pekin-Elmer required reading.

Most of the time, the signal across the opto will be small. For example, say you have a 2:1 compressor and set the threshold to -20dBu. You set the gain make up so that with 0dBu in you get 0dBu out. Since at 0dBu in you are 20dB above the threshold, the compressor should reduce the gain by 10dB and the make up gain will be 10dB. At this point the signal across the opto will be only -10dBu.

That still puts distortion well above what we would consider "low distortion" in today's world. You know, the one where people argue over which opamp has lower distortion (don't start me up). Obviously, this isn't a "today's world" project though.

In a typical implementation there will be threshold and gain make up pots in parallel with the opto so it is not unusual for there to be a 6dB loss with no gain reduction so the actual gain make up would be 16dB and the level across the opto would be -16dBu.

Of course makeup gain isn't free, and your non-sand thermionic gain devices will add noise in the process. Again, this clearly isn't a "today's world" project, so you're probably recording things onto tape where everything we're talking about is more than masked by its own noise and distortion.

The worst case for Vactrol distortion is when there is no compression at all. Perhaps this is why some people like to run there signal through opto compressors with no compression.

Cheers

Ian

Well, the worst case is the combination of high input levels and high attenuation, static gain or dynamic, where the Vactrol is the dominant signal modifier.
I've attempted to build compressors with Vactrols back when using the real PE VTL5C4/2 was just a matter of ordering up a batch from any supplier. I know, first hand, what their idiosyncrasies are, and tried to work around them. Many people did. But even when they were thick on the ground, they were not the gain control element of choice for pro audio devices. I think there was one I can think of. Everybody else, including all the sought-after classics, ended up using something else. I guess they don't count because of the silicon. I'll give you this: they're pretty simple to use, and the circuits you need around them to accomplish a compressor are fairly simple too.

If you're not making a commercial product, are you aware of the Raytheon Raysistor and the Sigma Data Cell? They are opto couplers too, but with incandescent light sources in them. I found a Raysistor on eBay and bought it as a trophy, as I used to engineer for a studio where someone built a custom console with them as the remote attenuator.

You might want to consider them, because even your Vactrol has a bit of "sand" in it.



Thanks for the explanation. I wish you the best.

 

[ruffrecords]

@jaddie

In today's world, very many people are seeking colour so I think from that point of view it is very much a today's world project.

Unfortunately, VCAs are not noiseless either. The THAT2180 has an output noise of -88dBV at 15dB of gain. My tube gain make up amp is perhaps 6dB worse than that under the same conditions.

I also came across the Raytheon device. I must admit I was seriously tempted to try one

Cheers

Ian

 

[gyraf]

back when I was allowed to use Optos, the VTL5C3 was favorite - up to around 1VPP i couldn't detect distortion with the AP

very different with the new ones

 

[ruffrecords]

back when I was allowed to use Optos, the VTL5C3 was favorite - up to around 1VPP i couldn't detect distortion with the AP

very different with the new ones

I always thought it interesting that PE had several different opto resistive 'material types' but never actually said anything about what they actually were. We all assume they are CdS based but are/were they?

Cheers

Ian

 

[gyraf]

I think there are slightly different formulas - the above pdf http://denethor.wlu.ca/pc300/optoisolators/analogoptoisolatorintroduction.pdf mentions five different "Material Types" in the table on P41/LastPage

Each specific material type represents a tradeoff between several characteristics.

 

[musipol]

Here's a question for Jaddie. What do you mean, "Vactrol has a bit of sand in it" ? Is there SiO2 or Si used in the Vactrol? I don't know, as I have never been involved in the manufacture of those particular devices.

 

[L´Andratté]

Here's a question for Jaddie. What do you mean, "Vactrol has a bit of sand in it" ? Is there SiO2 or Si used in the Vactrol? I don't know, as I have never been involved in the manufacture of those particular devices.

The LEDs in Vactrols are silicon based...incandescent light bulbs obviously not...

 

[jaddie]

Unfortunately, VCAs are not noiseless either. The THAT2180 has an output noise of -88dBV at 15dB of gain. My tube gain make up amp is perhaps 6dB worse than that under the same conditions.

Never said they were noiseless, but under proper applications the 2180 has a total dynamic range of 120dB. You pick how you apply it. It's a question of how you weigh the advantages vs disadvantages. Predictable control law, perfect channel tracking, low distortion, extremely wide control range, noise...you get to choose your poison, what you want to optimize. With other control elements, a lot of that is chosen for you, compromises already locked in and you're stuck with them. For example, try a stereo FET based dynamics proc. Fun for young and all, depending on your definition of "fun". Or a variable-mu version if you hate silicon FETs. It's an on-going tweak-fest now, and forever.

 

[JohnRoberts]

Never said they were noiseless, but under proper applications the 2180 has a total dynamic range of 120dB. You pick how you apply it. It's a question of how you weigh the advantages vs disadvantages. Predictable control law, perfect channel tracking, low distortion, extremely wide control range, noise...you get to choose your poison, what you want to optimize. With other control elements, a lot of that is chosen for you, compromises already locked in and you're stuck with them. For example, try a stereo FET based dynamics proc. Fun for young and all, depending on your definition of "fun". Or a variable-mu version if you hate silicon FETs. It's an on-going tweak-fest now, and forever.

The modern THAT corp VCA technology is quite good.

Everything has noise.

JR

 

[jaddie]

"Noise" levels only make sense in context of total system dynamic range. If the system is limited by a single device with a high noise floor (like any analog recording system), or a device operating too close to maximum output, then the total system DR has been assigned a range. Individual parts will add signficantly to that noise floor only when they are within a dB or so of the system noise, or above it. Noise 10dB below system noise doesn't change the total much.

I don't know what the end use here is, so a noise discussion can't really go anywhere.

 

[ruffrecords]

Never said they were noiseless, but under proper applications the 2180 has a total dynamic range of 120dB. You pick how you apply it. It's a question of how you weigh the advantages vs disadvantages. Predictable control law, perfect channel tracking, low distortion, extremely wide control range, noise...you get to choose your poison, what you want to optimize. With other control elements, a lot of that is chosen for you, compromises already locked in and you're stuck with them. For example, try a stereo FET based dynamics proc. Fun for young and all, depending on your definition of "fun". Or a variable-mu version if you hate silicon FETs. It's an on-going tweak-fest now, and forever.

I don't doubt THAT parts are extremely good. I was just responding to you comment that "your non-sand thermionic gain devices will add noise in the process". They do and so do THAT devices.

Cheers,

Ian

 

[jaddie]

I don't doubt THAT parts are extremely good. I was just responding to you comment that "your non-sand thermionic gain devices will add noise in the process". They do and so do THAT devices.

Cheers,

Ian

Your response seems pedantic. I've never said it's a question of noise or not. It's a question of how much, and the total dynamic range, at least, it is to me. I did quote the total dynamic range of a VCA, which I'm quite confident is greater than a Vactrol/Valve topololgy. Again, that may not be (and hopefully is not) the goal of the design.

Like I said...without context, we can't have this "noise" discussion. Everything generates noise (we seem to agree), it's a question of degree and context (we're nowhere close on that one). Your design goal clearly has emphasized the lack of silicon, aparently because every type or impliementation of the element does something bad, without explanation. I don't really need one. It does have the appearance of another out-of-context parameter evaluation, and there we clearly have a philosphical problem. That's fine, you do you.

 

[JohnRoberts]

Your response seems pedantic.

Pedantic is an insulting word that describes someone who annoys others by correcting small errors, caring too much about minor details, or emphasizing their own expertise. It is often used to describe someone who is narrowly, stodgily, and often ostentatiously learned in some narrow or boring subject

IMO Ian is not the pedantic one.

JR

 

[jaddie]

IMO Ian is not the pedantic one.

JR

Point taken. Apologies, all.

 

[ruffrecords]

Your response seems pedantic. I've never said it's a question of noise or not. It's a question of how much, and the total dynamic range, at least, it is to me. I did quote the total dynamic range of a VCA, which I'm quite confident is greater than a Vactrol/Valve topololgy. Again, that may not be (and hopefully is not) the goal of the design.

I am sure you are right and I am sorry you feel I was being pedantic.

Like I said...without context, we can't have this "noise" discussion. Everything generates noise (we seem to agree), it's a question of degree and context (we're nowhere close on that one). Your design goal clearly has emphasized the lack of silicon, aparently because every type or impliementation of the element does something bad, without explanation. I don't really need one. It does have the appearance of another out-of-context parameter evaluation, and there we clearly have a philosphical problem. That's fine, you do you.

This thread is about whether or not you can believe the published Vactrol specs. You mentioned noise and I responded. Can we please return to the original topic.

Cheers

Ian

 

[cyrano]

IMO Ian is not the pedantic one.

Agreed.

BTW, I now realise where you get all these insults, John, that I fail to see.

I just checked the word "pedantic" in around a dozen languages. Only Merriam Webster sees it as "insulting". All the others have many definitions, but only find the user of the pedantic style as being cocky, arrogant or whatever, but not as trying to insult anyone. The etymology of the word doesn't suggest it either, as it comes from "tutor".

 

[JohnRoberts]

Agreed.

oops

BTW, I now realise where you get all these insults, John, that I fail to see.

"broken windows policy" if we can reduce micro aggressions we will have less macro's to deal with.

I just checked the word "pedantic" in around a dozen languages. Only Merriam Webster sees it as "insulting". All the others have many definitions, but only find the user of the pedantic style as being cocky, arrogant or whatever, but not as trying to insult anyone. The etymology of the word doesn't suggest it either, as it comes from "tutor".

Yes, the "insult" in that definition seems a little odd. I am a chronic pedant but do not do it with any intention to insult others.

This definition shift to call more behaviors insulting is arguably divisive. "Hey you have just been insulted, how does that make you feel".
=====

To get almost on topic, one point about VCAs that I haven't seen inspected in this thread is that there is a tradeoff between residual noise and distortion related to operating current. The latest generation VCA provides access to this quiescent current but does not advise deviating from a nominal amount (and I am not either). So why did I mention this... Perhaps to be pedantic.

JR

 

[jaddie]

I am sure you are right and I am sorry you feel I was being pedantic.

I’m wrong, I accept John’s correction, guilty as charged.

But no insult was intended.

And there I go again!