Logarithmic Compressor?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
simonsez said:
Hai, I just finished download this very good bookhttp://geek.scorpiorising.ca/RDH4.html, quick view and i found this interesting circuit called logarithmic compressor. What this are for? Is there any kind of commercial product of this? Thank you...
Just on deliciously obsolete name for an audio compressor; I suspect the name comes from the fact that, over a specific range, the dB ratio of deltaVout to deltaVin is more or less constant.
 
I see, it's different from varimu comp. control voltage connected to cathode of the vari gain(?) stage instead to the grid on varimu comp. and it is single ended. Anyone tried this?
 
> deliciously obsolete name for an audio compressor

No, no.... look at how the diodes are really connected.

This is a Fuzz Box.

If the output signal (ratio-ed by resistors and transformer) exceeds the diode ON voltage plus the battery voltage, gain drops from about 1,000 to about 1. {EDIT} about 28.

It isn't even "logarithmic" unless the battery voltage is zero; then you get into Child's Law of vacuum diode conduction, except swamped by the large 25K resistances. And instantaneous log-gain is same-as odd-order distortion.

The main use for this is to listen to speech in presence of HUGE noise spikes. Such as communication radio. Set it so speech peaks are lightly clipped, then lightning bursts are clipped to hardly any higher than the speech instead of MUCH higher.

Hey, maybe it IS a good "effect" for today's music. Make R8 R9 a 50K pot, call it a "color box". But not what audio people usually mean by "compressor".
 
Hai PRR, I'm curious how it will sound, it's not a regular distortion box neither compressor. may be it's cool for bass track with wet/dry mix. I will try later this week with EF86 and 12AU7 i have in my hand. Two question..that 22,5v battery paralled with 1k pot...can i feed 24 volt dc from 7824 ic regulator ? 6H6 with 6AL5? I will post the report. Thanks..
 
PRR said:
> deliciously obsolete name for an audio compressor

No, no.... look at how the diodes are really connected.

This is a Fuzz Box.
You're absolutely right. Funny how, by seeing two valves and a dual rectifier, I jumped to the conclusion it was a compressor. I guess my attention span has gone below 5 seconds...
This is an interesting look in the past. I'm a little surprised this thing works only on positive signals.
 
> a little surprised this thing works only on positive signals.

??? No, this is perfectly symmetrical.

index.php


"Opamp" has gain of 1,000, very similar to the two tubes. I didn't have a 6H6 sim, but two diode-wired 12AU7 halves will work the same.
 

Attachments

  • log-comp.gif
    log-comp.gif
    19.1 KB · Views: 1,062
> had not figured out the diodes were back-to-back.

The drawing uses a conventional double-diode symbol, then a lot of criss-cross wires, so it isn't clear what's really happening.

wvcu2e.gif


Start top-right of dual-diode. That plate is connected to other cathode. Other plate is connected through variable-battery to first cathode. It's a back-2-back loop.

It appears that there is a DC voltage in series with one diode; but C2 makes that meaningless. There is some start-up bias-shift but C2-R2 settle quickly.

I should not have said "perfectly symmetrical". The left diode has R10 and battery in series. But the path impedance is up around 12.5K (25K||25K), the battery is a low audio impedance (couple ohms), and worst-case pot impedance is 250 ohms. So worst-case asymmetry is 2%. Perhaps annoying for measurement, trivial for music-trimming, and quite insignificant for the probable speech+static problem it was probably meant to solve.

The 1K is apparently a compromise between asymmetry and battery life. And perhaps an "academic" value which works long enough for the grad-assistant to make some tests, but will flatten the battery in a few hours. A proof of concept, not a Product.

It applies up to 32dB NFB around the amp. There are at least three strong bass-cuts around the loop (C2, C5, C6), plus output winding inductance, and a shelf at C3. I suspect that at full clipping it wants to be an LFO. In the suspected original application, it would get there only on the largest lightning-bursts, very short events. An LFO has a long start-up time. So it may never get into significant motor-boating with "lightning" (real or scratch-a-file fake lightning bursts).

I just can't see how this is better than a simple in-line passive clipper. I suspect it comes from a time when NFB was new and sexy and used everywhere, just as today we use micro-bots to tweet our electric consumption on Twitter (see this month's MAKE magazine).
 
me> I just can't see how this is better than a simple in-line passive clipper.

Like this one.

2ni1d9k.gif


I've guesstimated the DC voltages. The common cathodes are irrelevant, mostly to save a few parts (though also so you can use the older one-cathode twin-triodes). Back-2-back diodes strung from zero to +1.2V, center at 0.6V. They are non-conducting until audio peaks exceed 0.6V. Either way. Then they act as shorts to ground or to the low impedance cathode network which is much-much less than that 100K series resistor feeding signal from the first stage plate. The C-LC-C pi network filters the hash from brutal clipping, important to narrowband radio transmitters but optional for music-mashing. R1 lets you set the source near clipping level. R2 sets your transmitting level.

Silicon diodes can work the same, and without the added complexity of bias: Si has a built-in ~~0.6V threshold. Around modern boxes, a 10K resistor and two '007 diodes is all you need to brutally clamp your signal near 0dBu.

BTW: the April 1953 printing is missing the connection across the top of the dual-diode. Printing error.
 
PRR said:
me> I just can't see how this is better than a simple in-line passive clipper.

Like this one.

2ni1d9k.gif


I've guesstimated the DC voltages. The common cathodes are irrelevant, mostly to save a few parts (though also so you can use the older one-cathode twin-triodes). Back-2-back diodes strung from zero to +1.2V, center at 0.6V. They are non-conducting until audio peaks exceed 0.6V. Either way. Then they act as shorts to ground or to the low impedance cathode network which is much-much less than that 100K series resistor feeding signal from the first stage plate. The C-LC-C pi network filters the hash from brutal clipping, important to narrowband radio transmitters but optional for music-mashing. R1 lets you set the source near clipping level. R2 sets your transmitting level.

Silicon diodes can work the same, and without the added complexity of bias: Si has a built-in ~~0.6V threshold. Around modern boxes, a 10K resistor and two '007 diodes is all you need to brutally clamp your signal near 0dBu.

BTW: the April 1953 printing is missing the connection across the top of the dual-diode. Printing error.

To complicated.

Standard series diode speech clipper for transmitter(or whatever) use;
http://www.mines.uidaho.edu/~glowbugs/clipper.htm

One could easily build a stereo dirtbox for mastering purposes and sell it for mucho dineros to fools with money. Parts count is low and cheap. Trafo in, eb91/6al5, trafo out, pot to adjust dirt.
It even has a tubes in it, man!! ::)
Any takers? ;)
 
Standard series diode speech clipper for transmitter(or whatever) use;
http://www.mines.uidaho.edu/~glowbugs/clipper.htm

When this clipper distorts what is the dominant harmonic distortion even or odd?????
 
"When this clipper distorts what is the dominant harmonic distortion even or odd?Huh?"

Hard to say just by looking at the schematic.The math is somewhat complex, since each half of the dual diode is depentent on the other half. Bias+source/load impedance matters.

But transmitter design usually calls for symmetrical clipping, so thats my guess. It would however be very easy to make it asymmetrical, by fiddling with the diode´s cathode resistance values.
 
Well I built this circuit on my tube experimenters board and it works really well on mid range and high frequencies - especially vocals and guitar but sounds horrible down the bass end. Now I know why the designer rolled off the bass for the input with the 0.01uf cap.
It would be cool to add a pre-emphasis network before the clipping stage and a de-emphasis after - so you can distort the higher frequencies of a signal while retaining some relative bass purity. Any ideas on how to do this??

 
Bass-cut shelving(under threshold level) ---> limiter ---> bass-boost shelving (freq and Q equivalent to cut but with gain to set independently according to new hi and mid freq level after limiting) ?

just a thought...

Laurent.
 
that's pretty close to what some Marshall amps do with partial cathode bypass on first gain stage (ex. jcm800...) shelving down the lows so the highs distort first, then you can bring bass presence back (less distorted/compressed) in a later stage with the tone stack.

Laurent.
 
http://www.mines.uidaho.edu/~glowbugs/clipper.htm

In this circuit where and how do I implement the pre-emphasis and de-emphasis.
Before the input cap and after the cathode follower? The input cap already represents some kind of Pre-emphasis coupled with the input impedance. What db/oct curve is this 3db per oct??
 
Back
Top