PSU 0v connection help

[sonolink]

Hello everybody

I'd like to wire a VCA reference voltage and a summing amp to a Midas XL309 PSU. On the PSU there are 4 terminals labelled +18, 0, -18, 0.

The circuits are labelled accordingly -18, +18, but to me it's unclear which 0v should be returned to which 0v terminal on the psu.

So my question is: does it matter? If it doesn't can I wire all 0v together and connect them to only one of the psu 0v? Or should I bridge the psu 0v?

If it does, how can I know which belongs to which?

Thanks for your time and help
Cheers
Sono

 

[sonolink]

This is the VCA voltage reference circuit

 

[JohnRoberts]

Yes it matters.

That simple virtual earth summer will be very sensitive to tiny voltage differences between 0v (master and channel 0V).

Likewise that VCA circuit is also sensitive to mVs of DC voltage difference. You will notice that the VCA conditioning op amp is dividing down the control voltage to reference it to the local VCA 0V.

JR

 

[sonolink]

John
Thanks a lot for your reply.  Unfortunately I'm not sure I follow you on what I should do. Sorry for the language barrier

 

[ln76d]

John
Thanks a lot for your reply.  Unfortunately I'm not sure I follow you on what I should do. Sorry for the language barrier

Try to find manual for psu or console (or similar model from the same manufacturer).
There should be description which ground would be proper.
Look at attachment, it may be helpful, it's  from XL2 service manual:
http://bee.mif.pg.gda.pl/ciasteczkowypotwor/SM%20scena/Midas/xl2-service-manual%202.pdf

 

[sonolink]

Thanks for your reply In76d,

Actually, I have the service manual.  That's not the problem.  My problem is that I have many 0v connections but there are 2 0v terminals on the psu. How do I determine what 0v connection should be wired to which 0v terminal for the circuits I attached?

For example,  on the first schem, the summing amp, R35 has one end connected to 0v. On the PSU there are TWO 0v tsrminals (one is the dc return of +18v and the other the dc return for -18v. To which 0v terminal should R35 be connected. Same goes for the other 0v connections.

How can I determine that?

Thanks for your help
Sono

 

[ln76d]

Currently i'm going sleep
Put service manual over the web and paste link here.
Tomorrow we will look for some solution

Cheers

 

[sonolink]

Here's the XL3 service manual:

https://www.dropbox.com/s/fqzpp1tn6ug3jfm/midas_xl3.pdf?dl=0

Cheers
Sono

 

[ln76d]

I'm little confused
I've checked the service manual which is really good, but i don't know what you want to do?
Do you want to add another VCA and summing amp circuit to the console or you are making new routing or what?
In service manual all connections are labeled and most of the proper grounds goes through the multi pin connectors to the midas grounding scheme, so why you want to connect it directly to the psu? Is there any issue with original grounding scheme?

 

[sonolink]

Ok. I'll try to explain myself again

I want to rack 8 stereo modules of this console (XL304) and a VCA faders module (XL302). I am building now a VCA control reference voltage circuit to be able to use the faders (schem on my 2nd post) and 18 summing amp circuits for each of the 16 aux sends and a stereo master out (schem on my first post).

I have an original Midas xl3 Psu. This unit is a bipolar PSU and has 4 terminals: -18v, 0v, +18v, 0v. I understand each 0v terminal is the dc return for either-18v or +18v. So, there are TWO 0v terminals.

My question is, how do I know which 0v connection on the circuits, goes to which 0v terminal on the PSU? How do I know they should belong to -18v or +18v dc return?

My initial idea was to bridge both 0v terminals and connect all 0v together to one wire that connects to one of the psu 0v terminals, since I thought both terminals were the same, but according to John Roberts that's not a good idea. So my question is, to which 0v terminal should I connect the circuits' grounds and how can I determine that?

Thanks for your time and help

 

[JohnRoberts]

My speculation is that the two 0V wires get connected together at the destination (inside console) and that becomes the single common 0V node.

Of course it i will not stay 0V if there is current flowing  in circuits.

JR

 

[ln76d]

Now it's more clear

If you have XL390 PSU (which seems to be original unit ), all 0V are tied together inside.
Much more important would be wiring "0V" inside the rack.
You have a little more easier work, because there will be less modules and less "0V"/Ground paths to wiring
If you have acces to the original "VCA control reference voltage" look for "0V" paths.
Try to make similar grounding scheme as in midas console, so if on the each module is more than one "0V" path, wire them separately. You can try with star ground and connect it to the one "0V" in PSU.
That's my idea
If i made mistake in some point, please someone to corrected me (John?  ).

 

[JohnRoberts]

The poster will get a quick education about console layouts (sorry).

Too much to put into a single post, it takes console engineers years to figure this stuff out.

One obvious scenario used for VCA control where mV of voltage change results in dBs of gain change, is to distribute the control voltage scaled up to several volts, then pad it down again near the VCA so errors between the local and distant 0V gets divided down by the pad too.

JR

 

[sonolink]

Guys
Thanks a lot for all the input. I think I have the answer I needed and indeed the 0v can be bridged. On the original console that happened inside. I really needed to confirm this before risking making any wok party

My idea is to bring all 0v together an then take them to the PSU in a starground configuration indeed

The poster will get a quick education about console layouts (sorry).

You're right and it's my pleasure to keep learning thing

One obvious scenario used for VCA control where mV of voltage change results in dBs of gain change, is to distribute the control voltage scaled up to several volts, then pad it down again near the VCA so errors between the local and distant 0V gets divided down by the pad too.

John could you elaborate a bit on that? I'm not sure I understand

Thanks again to both for your time and help
Cheers
Sono

 

[JohnRoberts]

One obvious scenario used for VCA control where mV of voltage change results in dBs of gain change, is to distribute the control voltage scaled up to several volts, then pad it down again near the VCA so errors between the local and distant 0V gets divided down by the pad too.

John could you elaborate a bit on that? I'm not sure I understand

Thanks again to both for your time and help
Cheers
Sono

If the VCA gain changes 1 dB for every 6 mV of control voltage change,  a mV of noise can make a big difference. Instead  say we divide down the actual control voltage at the VCA by a factor of 10x, that way we send 60mV/dB around between modules and divide down any errors by the same 1/10.

JR

 

[sonolink]

Yes of course! It's funny how some things aren't obvious until you get to look at them from a different perspective

Cheers
Sono