dbx 160vu clone

[Andy Peters]

I once got a failure report from Eddie Kramer about a 160X which was distorting. -It turned out that he didn't know how to use it, and didn't know that "peak stop" on the 160 is a waveform distortion function. -Nor did he know which way to turn the control, as it happens...

Keith, I realize that your post is a year old, but I want to point out that neither the dbx 160X, the 160XT nor the 160A have a peak-stop function. (I have one of each. The A sound the best.) The various flavors of the 166, the 266 and the 1066 have the peak-stop feature. On the 166 and 266 it is unusable. It's almost usable on the 1066.

-a

 

[emrr]

The various flavors of the 166, the 266 and the 1066 have the peak-stop feature. On the 166 and 266 it is unusable. It's almost usable on the 1066.

That's what one calls a radio and TV broadcast feature.......

 

[TheGuitarist]

So whats the verdict on this one?

Possible or not possible. Seems like the parts are near impossible to get.

 

[iangomes]

i say possible with some minor tweaks to the design (THAT VCA?). i started up a schematic so I could eventually do a 500 series verison, but then I thought about it and decided that I needed to finish my unrelenting pile of projects (1176s, 175bs, 2254cs... the list goes on) first. Maybe one day I'll come back to it, but not for a while.

 

[SSLtech]

I once got a failure report from Eddie Kramer about a 160X which was distorting. -It turned out that he didn't know how to use it, and didn't know that "peak stop" on the 160 is a waveform distortion function. -Nor did he know which way to turn the control, as it happens...

Keith, I realize that your post is a year old, but I want to point out that neither the dbx 160X, the 160XT nor the 160A have a peak-stop function. (I have one of each. The A sound the best.) The various flavors of the 166, the 266 and the 1066 have the peak-stop feature. On the 166 and 266 it is unusable. It's almost usable on the 1066.

-a

Sadly this is another example of why i detest dbx's nomenclature... I was referring to the 160S which has peak stop and peak stop plus.

Keeping the same number for so many utterly different designs is despicable.

Keith

 

[abechap024]

Mybe if we stripped down the DBX 160vu to its bare bones and made it easier to clone it would catch on. I'm thinking that most of its character is in the 208 rms detector. It doesn't look that hard to build.

Buffered balanced input -> Threshohold -> Sidechain to 208 Rms Detector -> 2181 VCA -  (or optional 202 clone like what was done on the gssl project) -> Make up Gain -> Buffered Balanced Line output

Simple, Easy, one channel. With all the stereo models out there, I don't think anyone would mind have a few mono compressors that can be patched together if needed.
Bass, Kick, Vocals
I'm game to try and figure somthing out. Though i'm  busy with school, mybe this weekend i'll have some time. I am Going to make some simple one Channel Comps, I was thinking of some optical, but this sounds like an excellent project too. I really like the idea of just threshold and gain controls.
AC

 

[abbey road d enfer]

I really don't know why you attribute the 160VU's characteristics to the 208 RMS converter; have you done comparative assessments to support your theory?
The UAD site attributes, rightly IMO, the characteristics of the 160 mainly to its RMS detector, but that would also be true of other dbx products, like the 160A, which used a previous version of the 2252 chip.
It is true that the Blackmer RMS detector gives the whole family of dbx compressors a characteristic behaviour compared to other arrangements (mainly the progressive attack speed), but the 208 does not behave significantly differently than the actual 2252.
As I mentioned in a previous post, the 160VU uses the RMS detector with a single time-constant, as ooposed to the subsequent versions of the 160 family, which used a dynamics-dependant TC.
In addition, there are certainly much more differences between a discrete VCA and a monolithic VCA than between an original RMS and the current IC; these should certainly be taken into account regarding the different sonic signature.

 

[jdiamantis]

I really don't know why you attribute the 160VU's characteristics to the 208 RMS converter; have you done comparative assessments to support your theory?
The UAD site attributes, rightly IMO, the characteristics of the 160 mainly to its RMS detector, but that would also be true of other dbx products, like the 160A, which used a previous version of the 2252 chip.
It is true that the Blackmer RMS detector gives the whole family of dbx compressors a characteristic behaviour compared to other arrangements (mainly the progressive attack speed), but the 208 does not behave significantly differently than the actual 2252...

The difference is not so much in the detector, but in the timing cap and release/current resistor used. In the 160vu circuit, the detector uses a 22uf cap with a 909K timing resistor. In the 2252 data apps, the detector cap is 10uf, with a 2.0M timing resistor. While the release time scaling is identical, the difference is in the current injected into the charging transistor in the RMS module/detector vs capacity. With the larger cap and half the current used in the 208 module, attack will be slower than with the values specified in the 2252 data apps. Add to this the "non-linear" capacitor circuit, and everything beyond the 160vu is an entirely different beast.

jD

 

[guavatone]

I really don't know why you attribute the 160VU's characteristics to the 208 RMS converter; have you done comparative assessments to support your theory?
The UAD site attributes, rightly IMO, the characteristics of the 160 mainly to its RMS detector, but that would also be true of other dbx products, like the 160A, which used a previous version of the 2252 chip.
It is true that the Blackmer RMS detector gives the whole family of dbx compressors a characteristic behaviour compared to other arrangements (mainly the progressive attack speed), but the 208 does not behave significantly differently than the actual 2252...

The difference is not so much in the detector, but in the timing cap and release/current resistor used. In the 160vu circuit, the detector uses a 22uf cap with a 909K timing resistor. In the 2252 data apps, the detector cap is 10uf, with a 2.0M timing resistor. While the release time scaling is identical, the difference is in the current injected into the charging transistor in the RMS module/detector vs capacity. With the larger cap and half the current used in the 208 module, attack will be slower than with the values specified in the 2252 data apps. Add to this the "non-linear" capacitor circuit, and everything beyond the 160vu is an entirely different beast.

jD

JD is that R35 and C 15?

 

[abbey road d enfer]

I really don't know why you attribute the 160VU's characteristics to the 208 RMS converter; have you done comparative assessments to support your theory?
The UAD site attributes, rightly IMO, the characteristics of the 160 mainly to its RMS detector, but that would also be true of other dbx products, like the 160A, which used a previous version of the 2252 chip.
It is true that the Blackmer RMS detector gives the whole family of dbx compressors a characteristic behaviour compared to other arrangements (mainly the progressive attack speed), but the 208 does not behave significantly differently than the actual 2252...

The difference is not so much in the detector, but in the timing cap and release/current resistor used. In the 160vu circuit, the detector uses a 22uf cap with a 909K timing resistor. In the 2252 data apps, the detector cap is 10uf, with a 2.0M timing resistor. While the release time scaling is identical, the difference is in the current injected into the charging transistor in the RMS module/detector vs capacity. With the larger cap and half the current used in the 208 module, attack will be slower than with the values specified in the 2252 data apps. Add to this the "non-linear" capacitor circuit, and everything beyond the 160vu is an entirely different beast.

jD

Please read what I wrote. The 2252 can be made to respond absolutely identically to the 208, which cannot be said of the discrete and monolithic VCA's.

 

[jdiamantis]

I really don't know why you attribute the 160VU's characteristics to the 208 RMS converter; have you done comparative assessments to support your theory?
The UAD site attributes, rightly IMO, the characteristics of the 160 mainly to its RMS detector, but that would also be true of other dbx products, like the 160A, which used a previous version of the 2252 chip.
It is true that the Blackmer RMS detector gives the whole family of dbx compressors a characteristic behaviour compared to other arrangements (mainly the progressive attack speed), but the 208 does not behave significantly differently than the actual 2252...

The difference is not so much in the detector, but in the timing cap and release/current resistor used. In the 160vu circuit, the detector uses a 22uf cap with a 909K timing resistor. In the 2252 data apps, the detector cap is 10uf, with a 2.0M timing resistor. While the release time scaling is identical, the difference is in the current injected into the charging transistor in the RMS module/detector vs capacity. With the larger cap and half the current used in the 208 module, attack will be slower than with the values specified in the 2252 data apps. Add to this the "non-linear" capacitor circuit, and everything beyond the 160vu is an entirely different beast.

jD

Please read what I wrote. The 2252 can be made to respond absolutely identically to the 208, which cannot be said of the discrete and monolithic VCA's.

I did read what you wrote. Did you read what I wrote? (actually, I did write in error; the timing resistor injects double the current into the 208, not half. Still, this just affects threshold, not attack time; that is determined by the charge current on the cap via the RMS module, mostly during small changes in current/level). Yes, they can respond identically. However, in my experience the dynamic character of the timing parts contributes highly to the sound of the earlier units.

The discrete VCA's, and other parts add to that character, but the dynamic feel can be reproduced with the 2252 and the 22uf cap and 909kohm resistor combination (it won't be exactly the same, due to the difference in dynamic resistance characteristics of the charging transistors in the 208 vs the 2252, but will be close). Adjustments for threshold vs input signal would have to be re-calculated with the new components to achieve similar results to the new standard parts.

To answer the parts question, on my 160 schematic, the parts in question are R35 and C15.


jD

 

[abbey road d enfer]

I really don't know why you attribute the 160VU's characteristics to the 208 RMS converter; have you done comparative assessments to support your theory?
The UAD site attributes, rightly IMO, the characteristics of the 160 mainly to its RMS detector, but that would also be true of other dbx products, like the 160A, which used a previous version of the 2252 chip.
It is true that the Blackmer RMS detector gives the whole family of dbx compressors a characteristic behaviour compared to other arrangements (mainly the progressive attack speed), but the 208 does not behave significantly differently than the actual 2252...

The difference is not so much in the detector, but in the timing cap and release/current resistor used. In the 160vu circuit, the detector uses a 22uf cap with a 909K timing resistor. In the 2252 data apps, the detector cap is 10uf, with a 2.0M timing resistor. While the release time scaling is identical, the difference is in the current injected into the charging transistor in the RMS module/detector vs capacity. With the larger cap and half the current used in the 208 module, attack will be slower than with the values specified in the 2252 data apps. Add to this the "non-linear" capacitor circuit, and everything beyond the 160vu is an entirely different beast.

jD

Please read what I wrote. The 2252 can be made to respond absolutely identically to the 208, which cannot be said of the discrete and monolithic VCA's.

I did read what you wrote. Did you read what I wrote? (actually, I did write in error; the timing resistor injects double the current into the 208, not half. Still, this just affects threshold, not attack time; that is determined by the charge current on the cap via the RMS module, mostly during small changes in current/level). Yes, they can respond identically. However, in my experience the dynamic character of the timing parts contributes highly to the sound of the earlier units.

The discrete VCA's, and other parts add to that character, but the dynamic feel can be reproduced with the 2252 and the 22uf cap and 909kohm resistor combination (it won't be exactly the same, due to the difference in dynamic resistance characteristics of the charging transistors in the 208 vs the 2252, but will be close). Adjustments for threshold vs input signal would have to be re-calculated with the new components to achieve similar results to the new standard parts.

To answer the parts question, on my 160 schematic, the parts in question are R35 and C15.


jD

I was answering someone who considered that it was impossible to duplicate the behaviour of the 160VU without duplicating the 208. I say that the 2252 paired with the right components can be made to have exactly the same behaviour as the 208 in a 160VU.

 

[matthias]

why not rebuild the 208 module?

the components are still available

1x ca3083
1x bc453c
+ some resistors and caps

 

[abbey road d enfer]

Improve it with a 12ax7  ;D

 

[SSLtech]

Back to mono!  :-*

 

[iangomes]

why not rebuild the 208 module?

the components are still available

1x ca3083
1x bc453c
+ some resistors and caps

man was I ever struggling with figuring out what that ca3083 was... thanks!
I kind of moved on from this project but if people are interested, I could probably pull something together (I'd prefer 51x format, but that would mean a different power supply from the original).

 

[TheGuitarist]

personally i'm interested, in a racked version, or 51x, i'm buying one now

 

[abechap024]

Im interested!

 

[Morning_Star]

I'm interested.  Rack format.

 

[shot]

Yeah!
Very interesting!
rack format would be my choice