A Variable Width Circuit for The Precision MS Matrix

Construction information for DIY projects, including the MS Mid Side Matrix, Elliptic Equalizer, Mastering Console, Phono Transfer System, Insert Switcher and the Dual Class-A Amplifier. You can post your baby pictures here.
Post Reply
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

A Variable Width Circuit for The Precision MS Matrix

Post by mediatechnology »

I was contacted today by someone wanting to build the MS matrix. They had the simple question: "How do I add a width control?"

The obvious answer of course is to vary the ratio of mid to side between the encoder and decoder. If there's already a processor in that insert just use it's gain trim.

But what if I just want to hook up a pot to the board for a Width Control? Where?

Turns out it's pretty easy. I'll draw a schematic and post it. For now let's use the original block diagram.

Image

To make the board into a variable width circuit do the following:

Use the decoder as your encoder. The Left and Right Inputs are at "E." The PC board point marked "GL" (G Left) becomes "GM" or G Mid. "GR" becomes "GS" or G Side.

Add a two section 10K linear pot at insert point "G." Wire the pot so that at CW rotation mid is at full, side attenuated. Full CW rotation is zero width (Mono), full CCW is full Side.

Take the output of the "Encoder," which is Mid and Side and feed that into the "Decoder" inputs at point "A." Why the encoder/decoder role change? Because U9 is there to provide a buffer for the pot.

The Width-controlled Left and Right outputs are now taken from point "D."

Image
MS with variable width added.

Here's a short demo of the variable width control:

http://www.ka-electronics.com/content/M ... a_Live.mp3 (3.5 MB)
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Variable Width Circuit for The Precision MS Matrix

Post by mediatechnology »

This picture shows how easy it is to provide variable width for the MS Matrix.

Image
MS Matrix With Variable Width Control

(I don't normally build finished projects with solid telephone wire. But, this is science.)

Some notes:

Insert "GL" is Mid Level
Insert "GR" is Side Level

The pot for Mid/Side ratio is a 10K dual linear cross wired so that as one increases level the other decreases.
The pot should have a linear taper to provide, at mid-point, -6dB attenuation.
(As shown, full Side is at full clockwise "CW" rotation.)

Because the roles of the encoder and decoder are reversed, the PCB silkscreen requires redefinition.

This is how the encode inputs and outputs map:

Left In > PCB Mid In
Right In > PCB Side In

Mid Out > PCB Left Out
Side Out > PCB Right Out

The two jumpers looping across the top of the PCB are the Mid/Side Process Insert.
Variable width with additional processing in either Mid, Side or both can be inserted here.

The decode inputs and outputs map:

Mid In > PCB Left In
Side In > PCB Right In

Left Out > PCB Mid Out
Right Out > PCB Side Out

The builder may want to change the gain in the 5532 stage from -6dB to 0dB.
There will be gain changes as width is changed: It's the nature of MS.
There's always going to be more energy in mono.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Variable Width Circuit for The Precision MS Matrix

Post by mediatechnology »

The builder may want to change the gain in the 5532 stage from -6dB to 0dB.
I did and it works much better now having an essentially unity gain "encode-decode" gain structure rather than ~ -6-8 dB loss.

I also strongly recommend raising the input resistors of U9, and also the feedback, to reduce loading on the pot. (Cfb is also lowered to 10 pF.)
The pot loading (which is in the MS domain) results in an approximately -0.8dB encode-decode insertion loss.
To provide bypass without level shift, the value of the bypass resistor is raised to 56K2.
An alternative is to buffer the pot with an outboard unity gain stage.
In reality the 0.8dB dip in the middle of rotation (normal stereo) is quite small compared to the level changes resulting from changes in correlation as the width is varied from mono to full side.

Image
Mid Side MS Matrix with Variable Width and Bypass
Nescafe
Posts: 22
Joined: Sat Aug 03, 2013 10:39 am
Location: Jakarta-INDONESIA

Re: A Variable Width Circuit for The Precision MS Matrix

Post by Nescafe »

Hi,
I just think about using MS in Tube comp and also Pico Compressor (original mono and not finished yet :mrgreen: ), using this also as balancing debalancing card, since the compressor itself have gain, could You tell me what is the difference between using xfader and using compressor gain output in individual MS channel?

Thank You

OOT: any information or document to guiding me how to put a feedback/feed forward mod switch in the Original Pico?

Thank You again
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Variable Width Circuit for The Precision MS Matrix

Post by mediatechnology »

could You tell me what is the difference between using xfader and using compressor gain output in individual MS channel?
None really. Using the comp gain does the same thing.

In terms of switching the Pico into feed-back mode that would be a question for Roger Foote.
Nescafe
Posts: 22
Joined: Sat Aug 03, 2013 10:39 am
Location: Jakarta-INDONESIA

Re: A Variable Width Circuit for The Precision MS Matrix

Post by Nescafe »

Hi Wayne,

Thank You, so I don't need to mess with another pot :D

OK maybe I should make another thread about Pico, I think that I will put FB mode for the MS mode using Pico.
Post Reply