Hi all - first time poster.
I have built the Q&D comp 2 (thanks to Small Bear for the SSM2166Ps - the DIP form is incredibly hard to find now) and am looking for a way to vary the attack and decay. Has anyone else attempted this?
The averaging capacitor between Pin 8 and ground sets the attack and decay, but is way above the levels of variable capacitor available.
What I am thinking of doing is setting up a load of different capacitors of the values I need, and selecting between them with a single pole 12 position rotary switch, the break-before-make kind. The downside is I will need a lot of capacitors (maybe 30), and most of them will be doing nothing at any one time.
I am not asking for a wiring diagram or schematic, but just to know if I am on the right track or if there is another solution I am missing. Eg. is there such a thing as a capacitor network that would do what I need (I am a bit of a newbie at this)?
Many thanks,
Rexel
This isn't the answer to your question, but welcome to the forum. I built this effect too, sounds really cool.
Hi Rexel, Welcome!
I've used a on/off/on (not sure how its called) switch.
At the "OFF"(center) position it has ONLY a 2.2uF (or 4.7uf not sure)
And then at one side of the ON state I have 10uF that when engaged will parallel with the other 2.2uf, resulting 12.2uf
And at the other way around position it's a 22uF that also will parallel with the 2.2uf when engaged resulting a 24.2uF
It works really nice for me, easy to adapt to bass, guitars, vocals etc...
BTW: I THINK that the way you described you will have many redundant settings, I mean you better have a lot less cap options...
I'm sure Mark Hammer can explain all of it a lot better :)
Hope it can helps!
Kleber AG
Hi,
The SSM2166 is intended to make it very easy to implement a number of dynamic processing functions with only a few external components. The SSM2166 datasheet is intended to show you how to do that without also showing you the internal schematic of the chip. What they permit us to know is that the Avg Cap between pin 8 and ground sets the overall time constants, and that it is averaging out the envelope derived from a patented full-wave rectifier circuit. Just HOW it does that remains a mystery, so there is not a lot of advice to offer. If it were a garden variety half-wave rectifier, I'd say you could probably stick a pot in parallel with the Cavg to continuously vary decay time, but sadly, we don't know if that is possible.
On the other hand, since they intended this chip to be incorporated into mid-to-high-end designs, we can assume for the moment that the single caps are intended to provide attack parameters that complement whatever the decay parameters are that the same cap produces. That's a guess, but I can't see how they could have marketed the chip AND withheld crucial design/application info unless the "patented process" did something like that, so I think it's a pretty reasonable guess.
The appnotes indicate reasonable cap values between 2uf and 47uf. Usually, the larger cap values that accompany slower attack and decay settings are for low-frequency instruments or anything that has slow rise and fall times itself...like voice. If your plan is to use it only for guitar, then there probably isn't a whole lot of utility in opting for more than a couple of cap values. For example, a 2.2uf cap with a 4.7 or 10uf cap switched in parallel for 3 different settings.
You need to remember that longer decay times have the effect of slowing the recovery time from the last note. If your intention is to hear at least a little of the pick attack before the note "retreats", then shorter decay times are preferred, especially if you tend to pick fast. In the case of voice, for example, there ARE not rapid fire transients in the signal, so slow decay/recovery times are quite reasonable and have no blurring effect on the signal. Guitar is a different beast.
Certainly, feel free to mess around with the values that produce musically useful differences, but I have strong doubts as to whether any 2166-based compressor dedicated to a single instrument would need more than 3 or 4 different time-constant settings, or even whether they would be audibly different much of the time. Of course, if your intention is to produce a device that you could stick just about any signal source into, from snare to children's choir, that's another matter. For guitar, though, a handful of settings that increase the average time constant by a factor of 3-4 in each position is probably sufficient.
I might point out that the dynamics are also affected by the downward expension and the threshold at which it takes place as well. In service of living up to its "Q & D" name, Jack's version does not focus as much on this aspect as the chip permits, but it is worth exploring in my experience.
Thanks everyone for the comments - very helpful.
I will try a 1k pot in parallel with Cavg and see if that makes any difference to the sound.
Kleber AG and Mark Hammer - from what you say it sounds like I should be able to get away with far fewer caps and switch settings than I had thought - something of a relief to say the least. The compressor is to be used mainly for vocals, so I will pick some values and group them together as you suggest.
Thanks again!
Rexel
A 1k pot is probably a LOT smaller value than what is needed (assuming the circuit responds to such "invasions"). I'd recommend something larger, like a 500k pot, otherwise you'll end up with absolutely none of the lag you put a larger value cap in there for in the first place.
If vocals is the primary use, then perhaps the useful cap values would be a default of 10uf with paralleled caps of 10uf and 33uf to yield 10uf, 20, uf and 43uf, or approximately 2x and 4x the attack/decay time.
Cheers Mark - I appreciate your help :D
I have a nice 2 pole 6 way rotary switch in mind, and I should have all the caps I need to make it work within the value ranges you suggest.
Next is making it work with balanced inputs, balanced outputs and phantom power, but I think I might have that sorted. If not, I'll be back :wink:
Cheers,
Rexel
What none of us here appear to have played with is the opportunity for tone shaping provided by the chip. You will note that the beedback path and the leg to ground of the first op-amp are open to invitation, as is the link between the output of the first stage and input to the second.