Optical Comp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

analag

Well-known member
Joined
Apr 23, 2005
Messages
1,942
Location
Mars
Removed:

Let me get reckless and share my compressor again. This started out as an LA2A clone, and has evolved into what you see here. I recently threw the XSM10K:600 into the output and found it to be an excellent match. The way the Vactrol is drawn is exactly how I have it working and it works very well. Compression goes from smooth to as tight as you want yet no pumping or breathing. This is the only compressor I use for laying vocal tracks.
I will remove this schematic in 3 days.
10_08A.jpg


"Hey CJ " can you see the SS regulator? Zero hum, I rebuilt the tube section many times without touching that 7 year old PSU. BTW this pic was taken before the audio trannies went in.
Digital friendly tube device?
icon_wink.gif


analag
 
nice.. ..but dont you need a resistor in series with the LED so that it can "dim" when the voltage increases/decreases?..not just turn on and off...( LEDs dim with current, the voltage drop is constant..)

johan
 
[quote author="Johan"]nice.. ..but dont you need a resistor in series with the LED so that it can "dim" when the voltage increases/decreases?..not just turn on and off...( LEDs dim with current, the voltage drop is constant..)

johan[/quote]

Nope. This was a experiment that worked and I left it the way it was. Gain reduction is totally controllable. It seems to defy reason in the way it work?

[quote author="clint"]Which type of Vactrol did you use ? (as far as it's critical)[/quote]

VTL5C4/2 with the gain reduction meter setup taken from the LA2A.

analag
 
Wow! I've been waiting for this based on what you said in the one bottle thread. But it's not quite that same one, is it?

Hmm, maybe time to start gathering parts. (after paying some bills first)

I'm going to try to do the "tubesistor" stage in this, the one you posted earlier.
 
seems to defy reason

Not really. The cathode follower will source more or less current into the LED depending on the signal level, whether or not a resistor is added in series with it. The purpose of the usual resistor is to limit the maximum current through the LED. An LED will work without it but may end up fried.
 
[quote author="NewYorkDave"] An LED will work without it but may end up fried.[/quote]

Dave I have tried to "fry" the LED but to no avail. The CCS plate loaded stereo version lives in my mixer main buss doing light compression but clamps the explosions in a way that I have yet to hear an SS compressor do. This thing can stop a kick and a snare like you wouldn't believe, all without a pump or a breathe, due to the feedback attenuation technique. Technical speculation will argue that it should be slow, but I record to digital format and I have no problem when recording close to digital zero.

analag
 
There is no defiance of logic here.

The issue with the LED is not so much forward current, since the hollow state drive will limit that (unless it were a supertube like the 6H30). Pulsed I's of 50mA are probably quite safe, and you're not likely to get even that much.

The bigger issue is the reverse breakdown voltage of the LED, since this has a huge effect on the way the coupling capacitor drive works, and affects the compression characteristic markedly. When we had our earlier acrimonious exchange about this, PRR stepped in and asserted that LEDs broke down around 6V IIRC. I wrote to Vactec and tried to get them to say, and was stonewalled, but their spec, again IIRC, says don't apply more than 3V. That doesn't mean they break, and clearly they do not, since Rowan has had this beast working for a while. If you had large currents in reverse then you would exceed the dissipation limit, but clearly that's not the case here---your reverse currents are going to be equal on average to the forward currents, due to the single 0.1 coupling cap drive. There may be a long-term degradation mechanism, but if the time constant is ten years, who cares.

I measured some comparable Vactrols and found their LED Vbrs to indeed be around 6-7V. Some other LEDs I've measured are as high as ~30V, and based on the Vactrol spec some may be as low as ~3V, although that is probably the LED manufacturer's CYA spec.

The point is, this voltage has a huge effect on the precise way that the compressor works, as would be revealed by simulation. So the word to the wise would be, measure your optos and know your Vbr. To get Rowan's performance we need to know what his parts break down at.
 
why not just add inverse oriented *blue* led parallell to the vac - and mount it on the faceplate?
 
[quote author="tv"]why not just add inverse oriented *blue* led parallell to the vac - and mount it on the faceplate?[/quote]

If three volts or so is what you want, that would be o.k.

But, it is likely that you want the breakdown voltage to be substantial---exactly how much we don't know, since analag has not told us what his parts are doing. It's not an issue of protection but intrinsic to how the machine works. My point is this is not a well-specified parameter, hence perilous for manufacturing. Fine for DIY though, where we have all the time in the world to measure parts, and we don't have a deadline to ship x units to Rumplestiltskin so we can get paid and make payroll next month.

A circuit could be readily designed to simulate one with a given LED reverse breakdown voltage, while not depending on the actual voltage at all.
 
then, it's a blue led plus just-the-right-one zener in series. vac is safe, and there's a sexy *compression* indicator on the faceplate.
 
[quote author="tv"]then, it's a blue led plus just-the-right-one zener in series. vac is safe, and there's a sexy *compression* indicator on the faceplate.[/quote]

But what is the blue LED's breakdown voltage? (EDIT) If it's larger than the forward voltage of the Vactrol one plus the zener forward drop of about 0.6V you're o.k., and that's likely. If it weren't, then you add another signal diode in series with the zener-LED to reverse block, and take its forward drop into account. And you lose your indicator function, sadly.

You then have the issue of what the impedance of the new forward-conducting string is, compared to the old opto LED breakdown impedance. Probably not a serious difference, but something to keep in mind.

And you also have to now know that the Vactrol LED breakdown is greater than the zener-blue LED voltage.
 
hi, :grin:

The schemo of your comp is very nice and what's you said about how it's working and the sound is interesting.

I'm in the way with that kind of comp but with a SRPP stage on end.But I'm not familiar with The Opto side chain.. So,

How much is the GR Max that's you can reach in this design?

What is the function of the 500K pot in front of U5?

Thank's for your reply and if you send a new schemo that's will be very nice.
:thumb:

Best regards,Bruno.
 
[quote author="b2"]What is the function of the 500K pot in front of U5? [/quote]

I'll try. It's the "treshold". ie, controlling how much you want the opto envelope detector to react to the input signal.
 
And you lose your indicator function, sadly.
Jep that was my afterthought but have already posted. If vacs break at a couple of volts, a blue may work. If vac break at higher voltage then led's redundant I guess - except for eye candy.
 
[quote author="tv"]
And you lose your indicator function, sadly.
Jep that was my afterthought but have already posted. If vacs break at a couple of volts, a blue may work. If vac break at higher voltage then led's redundant I guess - except for eye candy.[/quote]

We need data.
 
Hi, :grin:

Thank's for the explanation of the pot (500k), that's nice for me if it's the threshold cause I want to plan a comp with a Vin,a Vout ,and just a minimum of pots(no release,no attack,..)and may be a Attenuator pots on the Output.


Thank's a lot. :thumb:

Bruno.
 
Speaking of which,

how would one go about doing more than one time constant on this?

Adjusting the cap/resistor values going to the opto element?
 

Latest posts

Back
Top