Help Needed for Allen & Heath Quasi mixer

Help Support GroupDIY:

musika

Well-known member
Joined
Feb 11, 2009
Messages
294
Location
Ohio USA
I'm recapping a 22X2 Allen & Heath Quasi mixer from 1973 and want to implement a bypass switch so that the signal will pass from the mic preamp to the output busses.  Can someone help with design (where the switch goes) and type of switch needed?  I am including the schematics that I think are relevant and a pencil drawing of where I think the switch should be placed.  I would also like to be able to set up a test ( I do not have lots of test gear) to insure each channel strip works prior to reinstalling due to the fact that the busses are hard wired.  So any advice on this would be great too. 
 

Attachments

  • UNADJUSTEDNONRAW_thumb_c9a3.jpg
    UNADJUSTEDNONRAW_thumb_c9a3.jpg
    284.5 KB · Views: 49

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
21,409
Location
Hickory, MS
That sketch looks you are connecting the bypass to a power supply pin?

Note:The Baxandall tone control stage is inverting so bypassing around that tone control will result in a polarity inversion... This could be audible in some cases...

JR

PS: They spelled bus wrong, but lots of people did back then..
 

musika

Well-known member
Joined
Feb 11, 2009
Messages
294
Location
Ohio USA
They made a two versions of this thing back then, a 2 and 4 output.  The 4 out had a toggle added to permit switching.  Since this particular mixer was a 2 bus version, the holes were already drilled through the frame and face plate but no switch included, and so I was looking at bypassing some circuitry that could help overall noise from 22 channels - also considered a channel mute (maybe this would be a better solution). 
 

cyrano

Well-known member
Joined
Sep 6, 2010
Messages
1,250
Location
Europe
JohnRoberts said:
That sketch looks you are connecting the bypass to a power supply pin?

Had me fooled for a second too...

It's an audio path, directly connected to the bus, with a somewhat confusing name: fader app? supply. No DC power there.
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
21,409
Location
Hickory, MS
cyrano said:
Had me fooled for a second too...

It's an audio path, directly connected to the bus, with a somewhat confusing name: fader app? supply. No DC power there.
OK you will still get a polarity inversion bypassing the inverting tone control.

JR
 

Bo Deadly

Well-known member
Joined
Dec 22, 2015
Messages
2,726
Location
New Jersey, USA
cyrano said:
Had me fooled for a second too...

It's an audio path, directly connected to the bus, with a somewhat confusing name: fader app? supply. No DC power there.
Uh, that's "fader amp supply" and it's definitely DC power alright.

MA1 output has DC bias. So whatever you bypass needs to be into a coupling cap. Maybe break the conn between the 47u and the collector of the transistor and then source from MA1. Note that you can just leave the feed from MA1 into the bax. Or you can break before the 10u of the second stage transistor. That would bypass the fader too but you would loose the feed for the F/B bus.

But I'm not sure how much that's going to improve noise. If you have a noise issue, you should diagnose that separately. There are probably other things that you could do that would make a bigger difference. In particular, the 220K pot makes for a pretty high source impedance. I wonder if there's a high gain NPN that has good noise performance with a high source impedance.

Or, a slightly crazier idea would be to put a JFET follower with a series resistor to match the OSI of the BC169. That might seem like a drastic measure but the key to low noise is low impedance so it might just be a win in this particular case. And it's like 3 parts so you could probably finagle it onto the board without making too much of a mess.

Another thing that would probably make a measurable difference is changing the supply for an SMPS with RLC filter.
 

musika

Well-known member
Joined
Feb 11, 2009
Messages
294
Location
Ohio USA
the 220K pot makes for a pretty high source impedance. I wonder if there's a high gain NPN that has good noise performance with a high source impedance.

Thanks for pointing this impedance issue out.  When I looked further into the preamp components compared to my Mod2 ( which was 3 years later), they used  a 47K log pot for preamp gain.  All other amp components are pretty much identical.  Unfortunately I do not have any good way to compare - just my ears - and I have the Quasi torn apart at the moment.  Anyways thanks for pointing this out!

I also noticed a braided shield added to the Mod2 version on the transformer secondary.  Photo included.  Additionally ground from tranny soldered to the case of gain pot.
 

Attachments

  • E52F4DA7-22DA-4694-9438-6CEFFEFC422B.jpeg
    E52F4DA7-22DA-4694-9438-6CEFFEFC422B.jpeg
    101 KB · Views: 35

Bo Deadly

Well-known member
Joined
Dec 22, 2015
Messages
2,726
Location
New Jersey, USA
musika said:
Thanks for pointing this impedance issue out.  When I looked further into the preamp components compared to my Mod2 ( which was 3 years later), they used  a 47K log pot for preamp gain.
That load has to match the transformer secondary. So the transformer would have to be different or maybe wired in parallel instead of series. So be careful or your mic Z will drop 4x.
 

musika

Well-known member
Joined
Feb 11, 2009
Messages
294
Location
Ohio USA
So the transformer would have to be different or maybe wired in parallel instead of series.

I believe you are correct.  Closer look at the two documents (one for the MOD2 and Quasi) shows two different transformers.  Sowter 3379 used in the Quasi. I have written to them asking for documentation for this transformer and will post to the TD when I receive an answer.  Thanks for the info on the preamp!
 

musika

Well-known member
Joined
Feb 11, 2009
Messages
294
Location
Ohio USA
Block Diagram showing 3379 transformer.
 

Attachments

  • Allen Heath Quasi block diagram.pdf
    105.7 KB · Views: 15

musika

Well-known member
Joined
Feb 11, 2009
Messages
294
Location
Ohio USA
I received a reply from Brian Sowter regarding the 3379 transformer used in the Quasi mixer.  He mentioned that this was all that he had for this particular transformer since it is no longer manufactured.  Very hard to read... I tried to clean it up a bit, but it does read 3379 in the upper right  corner.  Also I can see 1:6.
 

Attachments

  • Screen Shot 2020-08-20 at 11.21.14 AM.png
    Screen Shot 2020-08-20 at 11.21.14 AM.png
    408.9 KB · Views: 27
Top