MS Mini - Mid Side Matrix Construction Information

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.
User avatar
mediatechnology
Posts: 5805
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

MS Mini - Mid Side Matrix Construction Information

Post by mediatechnology »

MS Mini Mid Side Matrix Construction Information

This is the construction thread for the Mid Side Mini MS Matrix.

Image
MS Mini Mid Side Encoder-Decoder PCB Assembled

Buy Assembled and Tested, Boards with ICs or Bare Boards here: https://ka-electronics.com/shop/index.p ... ide%20Mini
Mid Side Mini Assembly and Test Documents: https://ka-electronics.com/images/pdf/M ... 1023_1.pdf (Updated 7/10/23)
Mid Side Mini Mouser Project Manager Bill Of Materials: https://www.mouser.com/ProjectManager/P ... 238566cd15

Image
MS Mini Mid Side Encoder-Decoder Schematic
The essence of leftism is denying the obvious.
https://ka-electronics.com
Fragletrollet
Posts: 16
Joined: Sat Oct 31, 2020 9:55 am

Re: MS Mini - Mid Side Matrix Construction Information

Post by Fragletrollet »

Hello!

I see most circuits using That chips normally use +/-15V, although the datasheets state they will run happily up to +/-18V. How does different rail voltages affect the headroom of the chips? The stated "Input Voltage Range" in the datasheet, does that refer to a +/-18V setup?

This will be used in a valve compressor with very clean amps with high headroom, so I believe the That chips onboard here will define the headroom of the unit, as they have lower headroom than the compressor. What (dis)advantages are there to running this at +/-18V compared to +/-15V?

Also, what would be the easiest way to bypass the M/S processing board, to use the unit as L/R? Something like bypass relay wiring of the M/S board + relaying the I/O of the compressor from the M/S I/O to the general I/O? ...so if I'm correct, 4 dpdt relays should do the job?

Edit: oh and 35V for the caps when using +/-15V and 50V when using higher rails, I presume?
User avatar
mediatechnology
Posts: 5805
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: MS Mini - Mid Side Matrix Construction Information

Post by mediatechnology »

Hi -
I see most circuits using That chips normally use +/-15V, although the datasheets state they will run happily up to +/-18V. How does different rail voltages affect the headroom of the chips? The stated "Input Voltage Range" in the datasheet, does that refer to a +/-18V setup?
Using +/-18V is no problem - you'll pickup just under 1.6 dB of additional headroom.

For the 1246 the +27.5 dBu maximum input level is specified with +/-15V supplies.
On the 1646 the +27.5 dBu maximum is specified with +/-18V supplies but that's with a 600Ω load so in practice if the load is bridging the 1646 can usually provide +27.5 dBu with +/-15V and a tad more with +/-18V rails.

I don't see a downside to using +/-18V.
Also, what would be the easiest way to bypass the M/S processing board, to use the unit as L/R? Something like bypass relay wiring of the M/S board + relaying the I/O of the compressor from the M/S I/O to the general I/O? ...so if I'm correct, 4 dpdt relays should do the job?
Maybe an insert switcher board with no Flip and only one insert's worth of relays would be the simplest?

I recall the BOM specifying 35V caps.
The essence of leftism is denying the obvious.
https://ka-electronics.com
Fragletrollet
Posts: 16
Joined: Sat Oct 31, 2020 9:55 am

Re: MS Mini - Mid Side Matrix Construction Information

Post by Fragletrollet »

No need to up voltage rating on caps when using +/-18V?
User avatar
mediatechnology
Posts: 5805
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: MS Mini - Mid Side Matrix Construction Information

Post by mediatechnology »

Since the bypass caps reference to ground they'll only see 18V of their 35V rating.

In a lifted ground fault scenario the caps are in series so they would only see 36V of their combined 70V rating.
They probably wouldn't split equally but in the unlikely situation where one cap was fully-shorted the other would only see 36V.
The essence of leftism is denying the obvious.
https://ka-electronics.com
Fragletrollet
Posts: 16
Joined: Sat Oct 31, 2020 9:55 am

Re: MS Mini - Mid Side Matrix Construction Information

Post by Fragletrollet »

Thanks for the explanation, makes sense :idea:
yannluaudio
Posts: 4
Joined: Fri Nov 22, 2024 5:24 am

Re: MS Mini - Mid Side Matrix Construction Information

Post by yannluaudio »

Hi there,

What is the specified maximum output level of this design ?
I measured only ca. 16dBu at the output of IC9 (ca.22dBu at the output of IC7) with +/-16V supply.
FYI I put some equivalent INA134 instead of the THAT1240.
User avatar
mediatechnology
Posts: 5805
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: MS Mini - Mid Side Matrix Construction Information

Post by mediatechnology »

yannluaudio wrote: Fri Nov 22, 2024 5:42 am Hi there,

What is the specified maximum output level of this design ?
I measured only ca. 16dBu at the output of IC9 (ca.22dBu at the output of IC7) with +/-16V supply.
FYI I put some equivalent INA134 instead of the THAT1240.
Thanks for joining us and your question.

Where are your jumpers set for Mid encode and Mid decode gain/attenuation? (And also side?)
Where are you inputting signal? (Encode with loop-through or directly into decode?)
What is the actual input level?
Are both channels being driven?
Are you using INA137 for the 1246 or did you use 1246 where they are required? (You mention the substitution of INA134 for 1240s which should be fine.)
With those details I can give you a better answer.

Here is a summary that may help:

Clipping points:

The single-ended outputs and internal nodes of the 1240/1246 clip at about +21 dBu with +/-15V supplies. The 1246 inputs accept up to +27 dBu before their output clips at +21. The 1646 balanced outputs clip at +27 dBu or +21 dBu single-ended.

If the Mid decode output at the 1646 is connected single-ended it will clip at +21 and the maximum level you'll be able to produce at IC9 will be +16 dBu before the 1646 clips because there is +6 dB gain in the 1646. If however the balanced output of the 1646 is used, IC9 will be able to produce about +21 dBu before the 1646 clips.

Gain structure:

The Mid encode level can be set to -6dB to prevent a +6dB mid build-up when both channels are driven.
The Side encode level can be either 0 dB or 6 dB depending on whether a 1240 or 1246 is installed. (In this configuration the 1246 provides 6dB gain rather than attenuation.)

The Mid decode gain can be either 0 dB or +6 dB. The +6 dB gain is to compensate for when, and only when, the Mid encode is set to -6 dB.
The Side decode gain can be either 0 dB or -6 dB. The -6 dB attenuation is to compensate for when, and only when, the Side encode is set to +6 dB.

If you let me know where you are inputting signal, on how many channels and where your jumpers are set I should be able to give you a better answer.
The essence of leftism is denying the obvious.
https://ka-electronics.com
yannluaudio
Posts: 4
Joined: Fri Nov 22, 2024 5:24 am

Re: MS Mini - Mid Side Matrix Construction Information

Post by yannluaudio »

mediatechnology wrote: Fri Nov 22, 2024 6:38 am
yannluaudio wrote: Fri Nov 22, 2024 5:42 am Hi there,

What is the specified maximum output level of this design ?
I measured only ca. 16dBu at the output of IC9 (ca.22dBu at the output of IC7) with +/-16V supply.
FYI I put some equivalent INA134 instead of the THAT1240.
Thanks for joining us and your question.

Where are your jumpers set for Mid encode and Mid decode gain/attenuation? (And also side?)
Where are you inputting signal? (Encode with loop-through or directly into decode?)
What is the actual input level?
Are both channels being driven?
Are you using INA137 for the 1246 or did you use 1246 where they are required? (You mention the substitution of INA134 for 1240s which should be fine.)
With those details I can give you a better answer.

Here is a summary that may help:

Clipping points:

The single-ended outputs and internal nodes of the 1240/1246 clip at about +21 dBu with +/-15V supplies. The 1246 inputs accept up to +27 dBu before their output clips at +21. The 1646 balanced outputs clip at +27 dBu or +21 dBu single-ended.

If the Mid decode output at the 1646 is connected single-ended it will clip at +21 and the maximum level you'll be able to produce at IC9 will be +16 dBu before the 1646 clips because there is +6 dB gain in the 1646. If however the balanced output of the 1646 is used, IC9 will be able to produce about +21 dBu before the 1646 clips.

Gain structure:

The Mid encode level can be set to -6dB to prevent a +6dB mid build-up when both channels are driven.
The Side encode level can be either 0 dB or 6 dB depending on whether a 1240 or 1246 is installed. (In this configuration the 1246 provides 6dB gain rather than attenuation.)

The Mid decode gain can be either 0 dB or +6 dB. The +6 dB gain is to compensate for when, and only when, the Mid encode is set to -6 dB.
The Side decode gain can be either 0 dB or -6 dB. The -6 dB attenuation is to compensate for when, and only when, the Side encode is set to +6 dB.

If you let me know where you are inputting signal, on how many channels and where your jumpers are set I should be able to give you a better answer.
Thank you for your answer.

Here is the setup : It is actually a loop-through configuration : no IC11/12/3/4 (pin6 of IC5 to pin3 of IC7 and pin6 of IC6 to pin3 of IC8)
Instead of the 1246 I use the 1256 (low cost)
Jumpers : Mid encode : 0dB - Mid decode : 0dB - Side decode : 0dB
Input Level on both channel : 22dBu
Output clipping at 16dBu (single-ended output)
But indeed I measured 22dBu at the balanced outputs : (between + and - output).
So I think I have the response to my answer. I forgot to measure the balanced output :)
User avatar
mediatechnology
Posts: 5805
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: MS Mini - Mid Side Matrix Construction Information

Post by mediatechnology »

Glad I could answer your question.
no IC11/12/3/4
I think you'll want to install an INA134 or 1240 in IC3 and IC4 even though you might not be using it balanced.

Even though it might be small, the source impedance of whatever you put in the insert will affect decode accuracy because it becomes part of the internal resistance network of the decoders.
IC3 and IC4 act as buffers.
If it is a "0Ω" op amp output in the insert, with no build-out resistor, it will be OK.
The essence of leftism is denying the obvious.
https://ka-electronics.com
Post Reply