Allen and heath system 8 mods?

Okay.. I've got a vintage allen and heath system 8 mkII 24 channel... and I recently aquired the schematics for it. I've been considering getting it serviced for awhile and wondered if any of you would be able to come up with any cheapish mods to it that would significatly improve the quality...

http://www.prismmusic.net/Mk2.zip

The schematics for it are up there.. Thanks to anyone who helps out!

What about a simple method of biassing the PNP differential on the mic input (Q1-Q4)? Substituting a single current source for R12 and R13 would help improve your CMRR without changing gain (someone correct me if I'm wrong here). Gain is set by the values of R16, R18, VR1 and the 22 ohm resistor next to it .

Maybe a PNP emitter follower on the output of that same differential--it would lower the differential's output impedance as it heads into IC1. Those would be my first two guesses, simple and cheap.

Re-cap the power supply with low-impedance caps. I personally like the Nichicon PW series, others will point you in the direction of the Nichicon HE's. I don't think you can go wrong with either type.

I'd recommend fast-rectifier diodes in the power supply as well. It looks like this uses a bridge rectifier.

You could go for some fancier opamps, but that might not be the best choice here, and it certainly doesn't fit into the "cheap" category.

Serviced...is it getting noisy on you like so many of the older A&H boards do? If so, a re-cap of the PS should help you with that problem.

The noise isn't that bad, but any reduction in noise is generaly a positive.

I kinda lost you through some of that. unfourtionately my electronics knowledge is a little shaky (though I'm trying to improve it)

I kinda lost you through some of that. unfourtionately my electronics knowledge is a little shaky.

Click to expand...

Well, you're in a great place to start learning, I can assure you that.

What parts are confusing to you? I'm happy to explain things as best as I can...

Okay, buffering the outputs of the differential pairs formed by Q1&Q3, and Q2 and Q4 is not the most crucial thing here, since output impedance is already low.

To provide your differential with a current source, remove those 8.2K resistors, R12 and R13, and substitute a current source instead--try using a pair of diodes in series--that runs from the V+ to the emitters of Q1/Q3, and Q2/Q4.

Since we're using diodes, we want to connect the positive end (the anode) to the most positive voltage, and the cathode (the band) will point to the less positive voltage.

Connect them like this: anode of diode 1 to V+, cathode of diode 1 to anode of diode 2, cathode of diode 2 to emitters of Q1/Q3. Do the same thing for Q2/Q4.

Theoretically speaking, the two diodes will appear as a very large resistor to the emitters of those particular transistors, and should increase CMRR to around 100dB IIRC.

Also, what type of PNP transistors are Q1-Q4? Maybe there's a lower noise sub available...

If you look at the bottom of the schematic it says that Q1-4 are ztx214, I was thinking I might swap the op amps on the first two channels to give me two nicer channel strips, if that would make much of a difference?..

Thanks for explaining more, I get what you mean to do now, but I don't quite understand the logic of using the two diodes in series, from my somwhat faulty electrical knowledge I would have thought the second would be redundant?

also, are any of those caps worth changing, most peopel seem to have said this is the thing holding that desk back, also would it be worth changing out some of the carbon resistors for metal film ones?, if so which ones in particular?

Thanks very much for answering again.. it's the only way us newbs learn

If you look at the bottom of the schematic it says that Q1-4 are ztx214

Click to expand...

Those are pretty good, I don't think you'll see any improvement from subbing transistors.

I was thinking I might swap the op amps on the first two channels to give me two nicer channel strips, if that would make much of a difference?

Click to expand...

Yes, I think it would. The existing TL072's are good, many people here like them. I personally find them to be a little on the bland side. Their current draw is low, about 2.5ma per channel. You won't want to double or triple your existing opamp current draw in a console. Try the OP275 as a sub for IC1. It's BJT, not FET, but it has lower noise and faster slew, while current draw remains the same. I think the price on these from Analog Devices is around $2 each.

I don't quite understand the logic of using the two diodes in series

Click to expand...

IIRC, it simulates the behavior of a transistor-based bias, but it's a little easier to implement. See, the overall Common Mode Rejection of the circuit is affected by the value of the existing resistor here. Using a current source instead of a resistor increases the Common Mode Rejection to around 100dB. I'll have to look at my copy of the schematic to tell you what CMRR is currently.

also, are any of those caps worth changing

Click to expand...

Absolutely. C3, C4, C7, C8, C9, C11 in the input stage. Like I said in my previous post--many of us here like Nichicon low impedance electrolytics like the UW, UZ, and HE series because of their tonal qualities.

would it be worth changing out some of the carbon resistors for metal film ones

Click to expand...

I like Vishay Dales--they are widely regarded as one of the nicer sounding metal films, and they're pretty cheap compared to Holco/Roderstein, etc. I would suggest changing out the resistors in the signal path.

*edit*--clarification

are those OP275's pin for pin compatible with the tlo72's?..

also.. with those diodes.. is there any particular type I should use there... or would just a run of the mill low voltage zenner do?

Thanks very much for all your help

The OP275's are pin compatible with the TL072.

You can use a 1N4148 or 1N914 diode.

You're welcome. Let us know how it goes!

> a vintage allen and heath system 8 mkII

That's vintage???

It is boring vintage. If all is well, these systems don't have "a sound" until you smack them into overload.

> biassing the PNP differential on the mic input (Q1-Q4)? Substituting a single current source for R12 and R13 would help improve your CMRR without changing gain (someone correct me if I'm wrong here).

Where do you connect one current source to two emitters with a DC blocking cap between them?

You often see two current sources, one per emitter. Yet you very often DON'T see this, even when the budget could cover the cost. The reason seems to be that the current sources are actually high-gain amplifiers, and inject surprising amounts of noise into the audio stages, especially at lower gain settings.

CMRR is rarely a problem in small studio work.

> emitter follower on the output of that same differential--it would lower the differential's output impedance as it heads into IC1.

This odd topology assumes that the input pair act as current sources. They don't, because of the collector resistors, but a radical change of impedance will screw-up the gain.

I've assumed simple emitter followers. If you do the compound PNP-NPN with the second collectors tied to the first emitters, so second stage signal current flows in the 22+20K gain-set resistor, you do get a more-linear (on paper) stage.

It's an elderly low-price mixer. Fix the aging parts. Electrolytic caps are most likely to go sour. Don't use exotic virgin silver and acorn-tanned deerskin capacitors, stay with the original ten-cent spirit. Great value for money. Generic Panasonics will be good-as the originals and give it another 20 years life. By then it really will be vintage, and the next owner/restorer can decide which way to go on caps.

I vote for leaving the first three channel opamps as they are, TL072 or similar BiFTs. The one that drives the assign switches might want to be something beefier. But you should go around all the electrolytics first.

Where do you connect one current source to two emitters with a DC blocking cap between them?

Click to expand...

You don't. You set one current source up for one side of the differential, and the other source up for the other side, as you said.

The reason seems to be that the current sources are actually high-gain amplifiers, and inject surprising amounts of noise into the audio stages, especially at lower gain settings

Click to expand...

I'm confused. The emitter tail (source) doesn't set gain in the differential circuit, but it does have a part in setting CMRR. I thought the point would be to make the source appear as a very large value resistor...is the trade off here an increase in current noise?

pulling this thread back from the grave.. it's looking like the desk might get serviced soon.. and I wouldn't mind getting some stuff done while I'm there.. so if anyone else has and ideas (or can clear up the last idea) that would be great .. also is the route of beefing up the power supply to allow chunkier op amps worth thinking about?

Hey, I've been looking at doing the power supply on this desk.. but I've been having trouble tracking down the nichichion a pw and he series. what other caps would you guys sugest? Or any other basic mods I could do?