My Calrec Sidecar console build and questions.

[bradholland]

Hey Again!

So after some good advice from Gareth Connor (who I'm hoping will chime in here)
I have gone ahead and tried to graft sections of the schematic of the master section from the calrec S2

as I'm really only going to use my sidecar console as a DAW mixer I don't feel I need all the extra functions like talkback, PFL, AFL etc etc. I really only want a stereo master output and stereo insert, I do have the MU3694 Twin VU meter and  MU 4353 LED Bargraph which I would like to also hook up.
I intend feeding the main mix buss back into the last channel in the bucket which is a stereo Line channel. I Have a  DL3678 Dynamics module which I will patch into the channels insert (to use as a buss compressor)

I have some questions about the schematic I have drawn...

Firstly, does it look right? lol
I dont see power connections on the op-amps passed the AD797, I dont see them connected in the main schematics either, they should be connected, right? Should I go ahead and feed them power the same way that the AD797 is being fed?

To take a feed off for my meters am I right in thinking I need to branch off after the OP275 op-amps and create another buffer driven feed for them there?

Something else i don't understand, is why the signal starts off from the channels antiphase and is then converted by the OP275 back into phase?
any other comments and general advice will be greatly appreciated..

 

[bradholland]

Here she is so far...

 

[Harpo]

Firstly, does it look right? lol

Pin numbers at your opamps don't have much in common with a AD797,
distortion cancellation cap connected between offset nulling pins doesn't look that healthy, ...

 

[bradholland]

Hi Harpo,
Thanks, yeah I need to make that cancelling cap look a bit prettier, it is connected though.

as for the pins not having much in common, do you mean to say I have connected them wrong? I've tightly followed the schematic and followed the same pinouts?
see this attached image...

or are you saying something else?

 

[RuudNL]

What is the idea behind the double feedback loop?
(R1-C2 and R3-C1)

 

[bradholland]

from the schematic notes....
"The level on the virtual earth input is extremely low due to the high loop gain and performance is improved at LF by the double feedback arrangement"

 

[Gareth Connor]

Hi Brad,

Looking good so far......

Power:
The general power feeds and grounds to chips, etc, are often not shown on complex diagrams as there is already a lot going on and additional lines going to pins 4 & 8 (or 4 & 7 if single op-amps) will only clutter up the schematic: keep the signal path as clearly readable as possible. There is a table on the right hand side of the Calrec schematic sheets that states what the power pins are for each chip type that is commonly used on the schematic.

Buffering:
As a general rule-of-thumb, as long as the feed to an external meter is from a low-impedance source, additional buffering is not necessary. The high-end pros tend to have a habit of buffering everything so as to enure that the main signal path(s) cannot come under the influence of external factors (loading, short-circuits on long cables, interference crosstalk on long cables, etc) that could get back into the console. In a small and controlled environment such as your sidecar, the additional buffering for a local meter in the sidecar's upstand is of no direct benefit to the sidecar project.

Phase flip:
The important thing is to keep phase integrity on all signals entering or leaving the console / sidecar.
Where signals pass through processing stages (EQs, filters, dynamics, pan circuits, etc), and especially if the signals can be switched so as to bypass the processing stages, phase integrity must be maintained whether the processing stage is bypassed or in circuit. The inclusion of seemingly irrelevant phase inversion buffers will be to do with maintaining phase through the overall signal path. When creating a shorter than original path in an adopted (adapted?) design, make a note of the desired phase at points where signal enters/leaves the new circuit and add/delete phase inverters as required whilst keeping the overall path as short as possible.

Not really trying to push a friend's literary creation but I can highly recommend a good book on audio circuitry......
Small Signal Audio Design by Douglas Self. It's recently been released as a second edition print with a whole lot more in it than the first edition.
It's a good bedtime read for analogue enthusiasts, but is more at home as a reference book on the tech-bench!

Trusting this helps.

 

[bradholland]

It helps a lot, thank you Gareth.
I did notice the power connections on the schematics after I posted this.
I will update the schematic this evening and update the post.

I trust that I will need to 'clean' the power lines to the chips in the same way that is displayed going to the 1st AD794 chip or can I do this once and then feed to all chips?
I did include that phase inversion buffer as I saw in teh schematic how phase changes as the signal is passed through the circuit.

One thing I am having a little trouble with is grounding. unfortunatley, on all the elco connectors, shield is not connected, only hot and cold. I currently have it set up so that pigtails are coming out the back of the console, but this results in me having to make a chassis connection between the console, my motu 24io and all my outboard equipment that is connected.

I think I can remedy this by wiring it all to a patchbay though...

I'll update teh schematic later and hopefully you can give it a once over before I get it printed up!

Really, really appreciate the advice from an expert here, thanks for taking the time..

Brad

 

[ricardo]

I trust that I will need to 'clean' the power lines to the chips in the same way that is displayed going to the 1st AD794 chip or can I do this once and then feed to all chips?

No.  You need to have at least 2x10u AT EACH OPA .  Certainly for the AD797 and preferably within 1cm for 5532/4.

These caps go to 'dirty earth' in 1980's Calrecspeak.

It's NOT enough to decouple rail to rail or use just 100ns.

 

[bradholland]

ok will bear that in mind, thanks ricardo.