original 670 poor version DIY

[SUPERMAGOO]

and decided to make the original sidechain-amp but somewhat simplified.
 with 6L6gc.
and .... works !!!!!!!!!!!!!!!!!!!

 

[SUPERMAGOO]

http://www.youtube.com/watch?v=p6L7TOTs0MA

 

[12afael]

at least is the bigger 670 clone I´ve seen. 

 

[SUPERMAGOO]

one of the first tests with the compressor.
converter is an awful sound blaster.
the two tracks are normalized.

http://atvmagoo.googlepages.com/cortitosincompmp3-01.mp3
http://atvmagoo.googlepages.com/cortito3.mp3

 

[SUPERMAGOO]

well I went to my studio with a converters of better quality to test this
wrong builded compresor.
I used to compress the bass track.
bass/tube-preamp/poorchild/rme-fireface 800.
to something incredible happened to me.
I can not understand something so poorly constructed that it sounds so
good!
The bass guitar was very poor quality.
the only error is in the mixdown is a bit strong master compressor out.
There are currently no plugin. except for the reverb drums

http://atvmagoo.googlepages.com/TOBOGAN.mp3

 

[SUPERMAGOO]

I need help.
I have a problem. only when the compression is very strong. pulse is derived from a high Frequency.
is Paracytic oscillation.
when this happens the compressor compresses loud and wrong.
I can not know where it occurs.
may be because the heaters are not independent?.
could be because the pin 5 (shield) it never connects?
very missmatch in transformers?
any idea? ??? ??? ??? ??? ??? ??? ??? ???

 

[SUPERMAGOO]

http://www.youtube.com/watch?v=KWnfBd7O6v0

 

[tomelectro]

I just had a similar problem with a 670 clone that someone brought me to fix.  When in moderate to heavy gain reduction, the left side-chain amp would launch into oscillation around 60kHz, which would cause the channel to over-compress, and also cause the right channel to go into gain reduction, even with no signal present on the right.  This clone used a printed circuit board for the side-chain amp, and I think it was a layout issue, especially since it happened predominately on the Left channel, and the component layout was completely different between left and right channels.  After spending much time playing around with the layout, grounding etc, I finally just threw in the towel and changed the 20pF feedback caps (from plate to grid of the last stage of the side-chain amp) to 47pF, which cleaned it right up. (If my calculations are correct, the 20pF caps were providing a rolloff around 120kHz, the 47pF bringing it down to under 60k).

So if you have a scope, try taking a look at your sidechain signal and see if there's any extra junk floating around on it.

 

[Winston OBoogie]

...I think it was a layout issue...

Me too.  You have to treat the side chain amp as a Hi-Fi power amplifier in terms of frequency response and stability.  

Not too much going on with the signal amp that can mess up so, in a way, even more care needs to taken with the power amp than the signal.


SUPERMAGOO, glad you got the 'MonsterChild' happening, for a while anyway.  I'm sure a tidy layout and a good checking of stability with appropriate compensation added will get you up and running.  I don't know what your output transformer is but, I'd check around the 6L6 - transformer nodes for weird stuff going on.  It's the only place you have any feedback happening so I'd start there.  Just as Tom suggested I guess...

 

[SUPERMAGOO]

I just had a similar problem with a 670 clone that someone brought me to fix.  When in moderate to heavy gain reduction, the left side-chain amp would launch into oscillation around 60kHz, which would cause the channel to over-compress, and also cause the right channel to go into gain reduction, even with no signal present on the right.  This clone used a printed circuit board for the side-chain amp, and I think it was a layout issue, especially since it happened predominately on the Left channel, and the component layout was completely different between left and right channels.  After spending much time playing around with the layout, grounding etc, I finally just threw in the towel and changed the 20pF feedback caps (from plate to grid of the last stage of the side-chain amp) to 47pF, which cleaned it right up. (If my calculations are correct, the 20pF caps were providing a rolloff around 120kHz, the 47pF bringing it down to under 60k).

So if you have a scope, try taking a look at your sidechain signal and see if there's any extra junk floating around on it.

Good idea.
I could improve it by another way.
I used a capacitor 330p where the original says "wiring-cap."
It would cut off something that high frequency audio.
I will try what you say.

 

[SUPERMAGOO]

...I think it was a layout issue...

Me too.  You have to treat the side chain amp as a Hi-Fi power amplifier in terms of frequency response and stability.  

Not too much going on with the signal amp that can mess up so, in a way, even more care needs to taken with the power amp than the signal.


SUPERMAGOO, glad you got the 'MonsterChild' happening, for a while anyway.  I'm sure a tidy layout and a good checking of stability with appropriate compensation added will get you up and running.  I don't know what your output transformer is but, I'd check around the 6L6 - transformer nodes for weird stuff going on.  It's the only place you have any feedback happening so I'd start there.  Just as Tom suggested I guess...

I know.
I'm using things that should not be operated and serious "transformers-missmatch".
2c51 and was not good.
but it's really amazing what it produces on snares, drums Buss, over heads, voices, basses.
controlamp out in the first test I used a guitar amp transformer.
later.
I have calculated a transformer 10k/600 / 18 ohms using data transformer sowter replacement.
I used to test ordinary silicon steel. think about the next test for a m6 steel.
and the circuit and it changed quite

 

[tomelectro]

Also, I'd be careful about working from that re-drawn version of the 670 schematic - I've found a few mistakes on it (if that's the same one that's been floating around for awhile. It looks like it).

 

[12afael]

huge knob cool!

 

[SUPERMAGOO]

Also, I'd be careful about working from that re-drawn version of the 670 schematic - I've found a few mistakes on it (if that's the same one that's been floating around for awhile. It looks like it).

this link is correct schem?

http://atvmagoo.googlepages.com/670_Schematic.pdf

 

[SUPERMAGOO]

I just had a similar problem with a 670 clone that someone brought me to fix.  When in moderate to heavy gain reduction, the left side-chain amp would launch into oscillation around 60kHz, which would cause the channel to over-compress, and also cause the right channel to go into gain reduction, even with no signal present on the right.  This clone used a printed circuit board for the side-chain amp, and I think it was a layout issue, especially since it happened predominately on the Left channel, and the component layout was completely different between left and right channels.  After spending much time playing around with the layout, grounding etc, I finally just threw in the towel and changed the 20pF feedback caps (from plate to grid of the last stage of the side-chain amp) to 47pF, which cleaned it right up. (If my calculations are correct, the 20pF caps were providing a rolloff around 120kHz, the 47pF bringing it down to under 60k).

So if you have a scope, try taking a look at your sidechain signal and see if there's any extra junk floating around on it.

I replaced 20p to 120p mica-plate. excellent!
I remove the "wiring cap" without problems.
compresses better and extends the range

 

[Winston OBoogie]

I replaced 20p to 120p mica-plate. excellent!
I remove the "wiring cap" without problems.
compresses better and

extends the range

Do you mean that you now have more high end?  I'm guessing you have less compression happening at the top end because you're possibly rolling too much off the side amp. 
I would do frequency response, square wave tests etc of JUST the power amp.  What you want is good response with stability and a bit of headroom but you don't want to roll off too much top.  Or maybe you do? 
You might only need to go up a bit in value from the 20pF you had in there.  Or you might need some other network on the transformer.  Could be just a secondary termination or something?

Or, if you're happy, leave it as is. 

Cool unit anyway 

 

[tomelectro]

this link is correct schem?

http://atvmagoo.googlepages.com/670_Schematic.pdf

Yeah, that looks like a copy of an original.  There was a nice clean high-res jpg that used to be around (posted by CJ?) . I don't have a place to post it, but if you want to PM me with your email, I'll pass it along.

Glad you got the sidechain to behave better!

 

[SUPERMAGOO]

I replaced 20p to 120p mica-plate. excellent!
I remove the "wiring cap" without problems.
compresses better and

extends the range

Do you mean that you now have more high end?  I'm guessing you have less compression happening at the top end because you're possibly rolling too much off the side amp.  
I would do frequency response, square wave tests etc of JUST the power amp.  What you want is good response with stability and a bit of headroom but you don't want to roll off too much top.  Or maybe you do?  
You might only need to go up a bit in value from the 20pF you had in there.  Or you might need some other network on the transformer.  Could be just a secondary termination or something?

Or, if you're happy, leave it as is.  

Cool unit anyway  

yes. have more hi-end.
I would have to pay more attention if it has too much roll-off (control-amp).
just 20p is unstable (motorboating distortion) and is necessary to place the "wiring cap".
I will try to 47p and 68p.
this roll-off could produce an interesting effect!

 

[SUPERMAGOO]

this link is correct schem?

http://atvmagoo.googlepages.com/670_Schematic.pdf

Yeah, that looks like a copy of an original.  There was a nice clean high-res jpg that used to be around (posted by CJ?) . I don't have a place to post it, but if you want to PM me with your email, I'll pass it along.

Glad you got the sidechain to behave better!

send this schematic!
thanks
and sorry for my terrible English.

[email protected]

 

[analag]

at least is the bigger 670 clone I´ve seen.  

No it's not...I've done it bigger than that.

How's that lookin...this one is a few years old and is working very well. Had to break out the transformer winder and get busy.