Taming the SSL's threshold sensitivity.

I came across a couple of ways for doing this:

1) resistor in series with one of the legs of the threshold pot.
2) a different value pot.
3) a combination of both.

What's the best way of doing this?

I also thought it would be nice to have a dedicated topic for this subject since I found the above answers all in different posts.

Thanks.

The threshold is defined by the amount of sidechain VCA gain, set by the voltage at TL072 pin1 (yes, I've switched the two half opamps at the PCB)

This voltage is proportional to the threshold pot wiper voltage, amplified by the 220k series resistor and the 56k feedback resistor.

So there is a lot of places around there you could alter the threshold.

To me it seems easiest to change the 220K resistor in series with the threshold pot wiper (from point "F"). Lower values gives bigger threshold ranges, higher values gives smaller threshold ranges.

Jakob E.

btw: added to SSL Meta..

I have a similar problem...
the thing is that if I set the threshold pot to 50% (0db) it is much more lower in real (at all ratio settings)


when I tested the sidechain in mod, I used two 4k7 (instead of the 47k)resistors to sum the two sidechain paths from each channel...
then the threshold range was really good and useful

before the pot-range from "none till toooo much compression" was very small...

I just would say "if it sounds good, it is good"

so Jakob, is this a way I can go with it ??

mat

Thanks Jakob, I will go for a smaller threshold range.

I didn't get usable results when changing the 220k

but I've got another idea ... you don't need to change the witdth of the threshold range, you can "move" the range... after testing I finally came to a quite useable result...

solder a 22k resistor in series with the red marked connector on the small ssl pcb:




when the threshold pot ist turned up to 50% of its range you have
0db when ratio is set to 10:1

@ radiance: please try it and tell me what you think about it...

[quote author="matthias"]I didn't get usable results when changing the 220k

[/quote]

I had the same problem.
For me, a 47K resistor in series with the + leg of the potentiometer did the trick.
I don't know exactly what is going on but to me it seems that the range is moved but also broadened.

For what it's worth...
after trying the above options out (and some others), i found with my clone that a 27k in series on the + leg made the threshold pot behave closest to the threshold pot of my alan smart c1.
Am using 2180 vcas.

Roger,

I don't see any problems in that..? I often used 2155's for sidechain vca's..

Jakob E.

Shouldn't alter threshold - maybe ratio a bit, but nothing really annoying..

Jakob E.

is there any way to measure what the actual range of response is in the threshold control?

I too need to tame the sensitivity.. with the threshold fully clokwise i still compress a good noticable amout, but the range seems Huge. would like 0 to truely be 0 though...

will thry the 22k, 27k, 47k resistors are the various places and report what happens. any other suggestions?

i put 2 1M8 resistors in place of the 47Ks.. that is the only thing that seemed to get it close to where it should be.

Now correct me if i am wrong, but.. if i am feeding a 0VU(-18dBfs) signal which comes out as -18dBfs in bypass, shouldn't i also see -18dBfs on the output when the comp is in, and i have it set to "0" threshold and makeup is at "0".
this is partially what i have been trying to achive... aside from getting this beast working.

I get very close to this (-18dBfs) in Ratio 10 and 2dBless (-20dBfs) in Ratio 4 and 6 dB less (-24dBfs) in Ratio 2..... :?


when i start changing the Ratio things get stupid. when i set the ref level to -10dBfs, and have the threshold at "0".. if i change ratios from 10 to4 to2 nothing happens..... i see only -18dbfs.... shouldnt it be -18 for 10, -16 for 4, and -14 for 2..... WTF?

to inform you, i have a 22K on the threshold leg as well

ok been typing this post for a hour.. (been up for a day and a half too.. so going to sleep before i get any more white hairs......)


will test more tomorrow (today... its 415am)

2181LBs are what i am using..

anyone haing issues with ratios and sesnitivity??

SORTED!! WOOHOOO


SO in the end..... the only deviations from the original Gyraf specified pars ar as follows
3 2181LB VCAs

the 2 47K summing resistors are now 2 1M8 (1.8M) resistors ....

and an 8K2 (8.2K) on the leg og the threshold pot.....

this is still not 100% accurate, but it so damn close it really doesn't matter.


Cant wait to build another!!!

so what VCAs dont have this problem......?

1M8 instead of 47K?
That's a rather BIG difference. But hey, if it works.....

Rather Big Difference is what i thought too. doesn't make any sense at all.

but it works well now.... so i won't complain.

I also have to tame the outpus of this thing (just remembered..) with 27K instead of the 15K. so maybe input are running hot (not temperature, but in level)... and also because the resistance is higher there that would make the need for a higher resistance at the summing location.....

the odd things is that although i have the 27Ks in place of the 15Ks i am running balanced........

Hows this for no sense.

On myu second SSL, adding resistors with either top or bottom legs of the threshold pot makes no difference at all to the range.

Feeding a sine wave to input, I hit 20db of compression at not far over halfway on the threshold.

I would have thought adding the resistor of the values suggest here, would make at least SOME difference wouldnt it ?? thanks.

THANKS RF, will do...

One question...

Will messing round with the resistors to change the threshold have any effect on the ratio's ??

I just got them sorted, dont want to mess em up ....?

Guess Ill give it a go and find out.

Sorry guyd. Another inept question.

Seeing the threshold range seems to vary with input level, what level should you feed it when you are making changes to try and broaden it ? thanks again. :grin: