Single Knob Compressor (Schematic enclosed).

..

THAT circuits will work for input and output, output will drive a transformer no problem. 

This could be really cool to make, kind of like a hardware version of Massey's CT4

Yeah, that is similar to my intentions. The idea came abut after using the Sonalksis TBK-3, though it is more of an effects compressor than anything. I'm still a little uncertain if the gain reduction meter will work, but I'm in the process of laying out a pcb and hopefully a prototype will form in the next week or two. The aim is for it to be under $100 per channel.

Maybe someone is interested in the Yamaha single knob compressor, currently used in most of Yamaha's mixing desks. See attached circuit diagram of a single channel. It is a low cost compressor.

I have been looking at this compressor schemo lately:
http://sound.westhost.com/project67.htm

A FET, TL072, a few caps and resistors...  looks very cheap to build. I am planning to proto board it soon...

Hello,
sorry for bad English ;-)

I`m new here in "diy". I`m looking for electric schematic for one-knob-compressor using in some Behringer-Mixers, like XENYX Q502USB, QX1002USB etc. Knows somebody this schematic? I can not found it in the www... Nothing... Can you say me, where I can find...?

best regards
Maxi22

The original one knob compressor was pioneered by DBX engineers decades before they became known as THAT corp.  I would start with that corp app notes

http://www.thatcorp.com/Design_Notes.shtml

Keep in mind that simple does not mean it will sound good, but the THAT design notes should be respectable.

JR

Thanks, JohnRoberts.
I know this THAT-Documents.
I need the compressor not for recording, but only for talkback-mic. The problem with the THAT-schematics for me is:
it is difficult to get the IC´s. The 218x-family is possible, but the others are not possible / difficult to get here. So I´m looking for the Behringer-schematic, because there is the 074 used...


Maxi22

Maxi22 said:

Thanks, JohnRoberts.
I know this THAT-Documents.
I need the compressor not for recording, but only for talkback-mic. The problem with the THAT-schematics for me is:
it is difficult to get the IC´s.

Click to expand...

Where are you located? There are vendors that ship almost anywhere...

The 218x-family is possible, but the others are not possible / difficult to get here.

Click to expand...

For a compressor, you need one VCA, one RMS detector and a few opamps. For opamps, almost any generic opamp woulmd do, TL0, 5532, even 1458.  VCA, the 218x is ok. Now the RMS was the 2252, but it's not produced anymore; you can probably find some from vendor's stock. But you should use a 4301/4305; are you saying the 4301 is not available for you?

So I´m looking for the Behringer-schematic, because there is the 074 used...

Click to expand...

Behringer use the same chips than THAT. Years ago they used NEC chips that were the same as dbx chips, now they use THAT chips.

Thanks for your answer. I have found a distributor for 4301 --> mouser.com.
Are you sure - Behringer is using the same IC´s like THAT? I have opened the XENYX Q502USB. On the platine there I can find a 14-PIN-SMD-IC "074" for the compressor-section. Is this not a 4x-op-amp 071?

Maxi22

Maxi22 said:

Thanks for your answer. I have found a distributor for 4301 --> mouser.com.
Are you sure - Behringer is using the same IC´s like THAT? I have opened the XENYX Q502USB. On the platine there I can find a 14-PIN-SMD-IC "074" for the compressor-section. Is this not a 4x-op-amp 071?

Maxi22

Click to expand...

As I wrote earlier, opamps are a non-subject, what matters are the RMS and VCA.
I don't know what THAT circuit you're referring to, since your OP is empty, but on the THAT app note, the opamps are either 5532, 5534 or LF351, but they could be also TL0 or even 741, there would be performance differences, but the circuit would still work. Now if you use a 4301, three opamps are already in there.
Fig. 2 in http://www.thatcorp.com/datashts/dn107.pdf shows that you don't need any external opamp.
Fig. 11 in http://www.thatcorp.com/datashts/dn125.pdf shows a one-knob compressor that just requires a few transistors in addition to the 4301.
Behringer use TL074 because it's a very cheap opamp that offers good performance.

Here's the up to date and complete Yamaha MGP mixer one knob compressor again.

Transistors are Toshiba. Either get them from eBay, or use BC/2N transistors with the same specs.
NJM4580 and NJM4565 = NE5532
1SS355 = 1N914

Easy to build, and does the job very well.

metalb00b00 said:

Here's the up to date and complete Yamaha MGP mixer one knob compressor again.

Transistors are Toshiba. Either get them from eBay, or use BC/2N transistors with the same specs.
NJM4580 and NJM4565 = NE5532
1SS355 = 1N914

Easy to build, and does the job very well.

Click to expand...

This bears an uncanny resemblance to my degree course final year dissertation in 1973! I built a compressor using 741 op amps which rectified both the input and output signals. The input  was then passed through a piece-wise linear stage to set the compression slope and then this was compared with the output in a comparator. The comparator drove  FET attenuator. The idea was the compressor slope would be independent of the FET characteristics. I will see if I can dig out the schematic.

Cheers

Ian

Thanks all for your answers

@ abbey road d enter:  yes, I know - a RMS is necessary... But on the Behringer-platine I couldn't found a special chip for RMS. There are only the 4580 an even the 074 (and 339 for LED-meter). Thats the reason for me to get/understand the schematic how it works on Behringer.

Thanks for the schematic of YAMAHA


Maxi22

Maxi22 said:

@ abbey road d enter:  yes, I know - a RMS is necessary... But on the Behringer-platine

Click to expand...

What Behringer model? Why has the schemo disappeared from your OP?

I couldn't found a special chip for RMS. There are only the 4580 an even the 074

Click to expand...

These are both opamps. So, no RMS, no VCA...? It may be an opto or FET compressor.

(sorry for my bad English)

Behringer XENYX Q502USB or XEENYX QX1002USB etc.

There are all SMD on the platine. Most of them are not marked/named. It can be, one of the transistors is an FET... There is no VCA-chip. I think the named-IC "074" means a 4x-op-amp...


Thanks
Maxi22

Maxi22 said:

Behringer XENYX Q502USB or XEENYX QX1002USB etc.

There are all SMD on the platine. Most of them are not marked/named. It can be, one of the transistors is an FET... There is no VCA-chip.

Click to expand...

OK, so it's an FET compressor. I don't know why the subject drifted to VCA's. The precedentlyposted Yamaha schematic should not be too different tha the Behringer's.

  I think the named-IC "074" means a 4x-op-amp...

Click to expand...

Yes, 074 means almost undoubtedly TL074. By today's standards, it's a general-purpose opamp, not too good in terms of noise performance.


Thanks
Maxi22
[/quote]

Thanks for your answers. I think too, the Behringer- and Yamaha-Schematic are similarly...


Thanks
Maxi22

ruffrecords said:

This bears an uncanny resemblance to my degree course final year dissertation in 1973! I built a compressor using 741 op amps which rectified both the input and output signals. The input  was then passed through a piece-wise linear stage to set the compression slope and then this was compared with the output in a comparator. The comparator drove  FET attenuator. The idea was the compressor slope would be independent of the FET characteristics. I will see if I can dig out the schematic.

Cheers

Ian

Click to expand...

I had another at this schemo, and noticed a nice detail. The big issue with FET compressors is they have to operate the FET at quite a low level (typically about -30dBu) in order to keep distortion at a reasonable level; as a result, teh make-up gain stage becomes quite noisy. Here, they have put the FET in the inverting input branch, which operates the opamp at idle with a moderate noise gain, varying from 2 to 4, depending on the compression pot. When compression happens the noise gain increases, up to about 30-100, but noise is masked by signal.
As you noticed the combination of feedforward and feedback detection deserves some attention.
On the negative side, I see that the detector is half-wave and the FET has no distortion compensation on its gate; both tend to indicate that low distortion was not a major concern in the design.