How critical is the choice of buffering opamps...

[radiance]

The project is a 20 channel (10 stereo channels) active summing box with for each channel a stepped gain rotary switch, a stereo aux send rotary switch and maybe a very simple passive EQ (one cut/boost rotary and one frequency selection rotary for 9 different frequecies). I guess this can be called a mixer right?
Well, the question is: for buffering purposes, should I go for discrete opamps or is that way over done? I have ordered some 990 pcb's and planned to use them as make up gain stages for the passive EQ's and maybe as summing opamps.

My goal is to make a good sounding 20 channel mixer with 2 stereo groups and a master buss, so I can connect some outboard (1176, SSL, Pultec..) and some FX units. I will use it to sum the outputs of my DAW and some synths...

 

[Svart]

if you are going through the trouble of building the box I would put the best you can afford in there. you are already going the discrete route, which when building something like this the cost is really not going to be that outrageous.

I say keep going as you are.

:thumb:

 

[radiance]

[quote author="Svart"]if you are going through the trouble of building the box I would put the best you can afford in there. you are already going the discrete route, which when building something like this the cost is really not going to be that outrageous.

I say keep going as you are.

:thumb:[/quote]

True..., but most of my DIY outboard gear has a 5532/5534 input stage (1176/ssl). Will I hear an improvement in sound quality when I use discrete buffering opamps on the channels, when after that the signal is going through my (5532/5534 input stage) outboard gear inserted in the groups?

Also, it's going to be quite expensive if I go for 24 position elma switches.
I don't know how much the parts of those 990 are going to cost me, but it ads up, it's 10 stereo channels u know...

 

[Svart]

Very true on all points. I only meant that in the long run, spending a larger sum of money in the beginning can save you more time and money later on. I speak my opinion on this because more often than not I end up saying " i wonder what it would sound like this way..." and change something. I am currently doing this with my Forssell compressors, although they sounded great with the IC I/O i put some FetBloaks on one and immediately decided to change them all (4 of them).

If I were building the hub of all the outboard units as you are now I would try to make it as good as possible since you are now creating the foundation for good or bad sound.

You make a good point though, there are plenty of IC solutions that would work very well and probably just as good as any other out there.

I bought enough PCBs for 12 Bloaks and enough discrete parts for 20 and the total was less than 100$. To me that is worth it.

How many opamps are you estimating to use?

 

[radiance]

[quote author="Svart"]
How many opamps are you estimating to use?[/quote]

Well, it depends. I'm not completely sure about the design (it's all pretty new to me..). It can be between 30 to 50 I guess.
I think I'm going to design a pcb for this which will accept both discrete and normal opamps, then fiddle around and have a listen to various opamps.
This is all in a very early stage and I feel I'll have to learn a bit more before I start building.
Right now it's just brainstorming and collecting stuf I might need in semi large quantities...(when I find it on evilbay or for cheap :grin: )

 

[radiance]

I also considered using gain bloaks but I hear they get quite hot right?
I think that will be a problem when stuffing 30 or so of them inside a 5 unit case....