Soundcraft 1600/800B

[Riley Casey]

Hey all!

I appreciate all the comments here- right after I bought the board, I relocated from Ohio to the DC area. It took me months to arrange for pickup of the Soundcraft boards, and I finally have them here. I've been swamped with work--- does anyone have a recommendation on someone fairly local to the DC area who can help me get this console in working order? I'd love to finally get this thing up-and-running, and ideally I could see someone work on it so I can tackle some of the work myself...

Thanks in advance

If the 800B was local to the DC area I have to wonder if that was my board from the 1980s. If it has the remnants of a burned 17V build out resistor on the aux master board ( that failed half an hour before doors at a major show ) it may well be. In any case if you can't find anyone who works on consoles on a regular basis I can take a look at them. In addition to the 800B I also built a 1600 clone console for a remote truck (Allen Bradley pots, P&G faders, LED bar graphs for all inputs plus 24 channels of monitoring ). Biggest improvement to the performance of that console was installing large gauge copper buss bars in the chassis pan and wiring every XLR pin one and every channel ground to that bar with 20 ga wire. Made a substantial difference in the noise floor.

 

[abbey road d enfer]

wiring every XLR pin one ... to that bar with 20 ga wire.

Which was gospel in those years, before Muncy et al demonstrated they should go to the chassis as directly as possible. Actually, there were no GSM, no WiFi in those times.

 

[Riley Casey]

Which was gospel in those years, before Muncy et al demonstrated they should go to the chassis as directly as possible. Actually, there were no GSM, no WiFi in those times.

Taking pin one buss bar was actually in response to Neil Muncy's paper. The aluminum back panels were not the best ground compared to six inches of wire to the buss bar as it turned out. The original design had pin one pass to the ground trace on the PCB at the mic input pads which was SOP up til then.

 

[ruffrecords]

Not disagreeing with John but to get completely 'nerdy' if you measure old removed caps you can spot trends like modules that are near bulkheads tend to have fewer 'failures' and even certain areas of the circuit boards. When shifting through 3000 or more caps you have to find something interesting to do!

There is probably a PhD in there for someone.

Cheers

Ian

 

[Newmarket]

Taking pin one buss bar was actually in response to Neil Muncy's paper.

Oh dear - you've said "buss" - get your flack jacket on

 

[Newmarket]

As a summing amp, a 2520 is not better than the 5534's that are in this position. A 990 would be better.

I'm not familiar with the '990' . So I'm curious - What makes it a better option cf 5534 for summing ?
Thanks.

 

[Matt Syson]

What makes it a better option cf 5534 for summing

First you have to define what 'better' is. and consider all the rest of the audio chain and the resistor values in the summing bus system. What are your definitions of 'best'?
I think it was the 1600 type modules I 'experimented' with and discovered a significant part of the distortion of the mic amp has no bearing on the op amp used. The original grounding system was well intentioned but a bit lacking in rigorous implementation to which adding lots of thick copper wires is a bit of a 'sledgehammer to crack a nut' approach.

 

[JohnRoberts]

I'm not familiar with the '990' . So I'm curious - What makes it a better option cf 5534 for summing ?
Thanks.

it's more expensive?

JR

 

[abbey road d enfer]

I'm not familiar with the '990' . So I'm curious - What makes it a better option cf 5534 for summing ?

1.1nV/sqrtHz noise voltage (10dB better than 5534) Slightly higher noise current, but still the OSI is 10x lower than 5534. Typical bus impedance on a 32 channel is about 200-250 ohms.

 

[Newmarket]

What makes it a better option cf 5534 for summing

First you have to define what 'better' is. and consider all the rest of the audio chain and the resistor values in the summing bus system. What are your definitions of 'best'?

Understood. I was interested in why Abbey considered it better rather than setting my own criteria.

 

[Newmarket]

1.1nV/sqrtHz noise voltage (10dB better than 5534) Slightly higher noise current, but still the OSI is 10x lower than 5534. Typical bus impedance on a 32 channel is about 200-250 ohms.

I see. Thanks.

 

[Matt Syson]

So while it might appear to be slightly 'quieter' certainly on paper, you have to evaluate whether this 'improvement' is significant when you take into account the noise of the channels and the other amplifiers following the summing itself. If lower overall NOISE is your goal then there are many other areas that should receive attention, possibly requiring wholesale resistor and capacitor exchanges and 'upgraded' chips elsewhere. Intelligent use of summing groups and ensuring all channels not needed in a particular 'mix' are unrouted (taken off the bus) may yield 'better' results for nothing.

 

[abbey road d enfer]

So while it might appear to be slightly 'quieter' certainly on paper, you have to evaluate whether this 'improvement' is significant when you take into account the noise of the channels and the other amplifiers following the summing itself.

Indeed improving one factor whilst ignoring others is disappointing.
However, improving bus noise is an element of confidence, just like the sound of doors in a car; it does not impact performance, but has been found to be a decisive element in purchase motivation.

Intelligent use of summing groups and ensuring all channels not needed in a particular 'mix' are unrouted (taken off the bus) may yield 'better' results for nothing.

The designer of the 1600/800B was very much aware of it; that's why the resistors of unused channels are disconnected from the group bus. However, the L/R mix bus are connected permanently to all channels, groups and FX returns, so it's clearly an area that could be relatively easily improved.

 

[FIX]

If you want to stick to an IC, use a TL072. They sound good. Otherwise for clarity, an OPA2134 is nice, but all of it is a waste of time, if you want to make a difference, you should use an old API 325 card, the op-amp and transformer will give you some flavor. You really only need the summing amp, the booster is fine the way it is. IF you really want to be cool and attract chiks, use a mic pre of your favorite flavor. Put a 200 ohm resistor across the input and hook the transformer to the bus. I did that so a console at Sony Publishing in Nashville once with a couple of V72 tube amps and the mastering guys could never figure out what console the mixes were from...

 

[FIX]

Never been a real fan of the 990, though I use two in a mic pre I make. It has way too much gain, the Tonelux op-amp is really sweet, my design of it limited the gain to 75 dB so it is really smooth, and is so stable that you don't even need compensation... The first open loop pole is at 10KHz, where the 990 is at something like 200Hz. don't get all hung up on "nano-turds per root canal". None of that shit makes any difference. Only the tone does. Old 2520 opamps have a slew rate of only 2.5V/usec. And they can be noisy, but they made half the records you hear. No one ever said "This song would have been a hit if that guy hadn't used 5532's on the mix bus"

One way of minimizing bus noise is to hit the bus harder from the channel, or do active balanced busses. Thats what I use on the FIX console. I hit the bus anywhere from 0 to +4, where -10 to -2 is the norm. That translates to easily 6 dB less noise. IF you are doing 8-14 channels, it's not really a problem. Like I said, don't get wrapped up with all the crap.

 

[FIX]

But wait there's more...

Like I said earlier, don't pick opamps based on noise or bandwidth, as the internal design of the opamp is more important, like the original 5524 was internally compensated to look good with a 10KHz square wave so they could advertise better slew rates. most of the OPA amps are nice inside, as is a TLO72, where a TLO82 is a completely different animal. Older 741s are also nice because they don't have a lot of gain and don't get away from you easily.

Train your ears, not your eyes.

 

[abbey road d enfer]

If you want to stick to an IC, use a TL072.

In the summing amp? 18nV/sqrtHz!
Why not an uncompensated uA748?

They sound good.

Many opamps sound good, as long as they are in the proper circuit.

Otherwise for clarity, an OPA2134 is nice, but all of it is a waste of time, if you want to make a difference, you should use an old API 325 card, the op-amp and transformer will give you some flavor.

Flavour? You mean distortion?

 

[abbey road d enfer]

don't get all hung up on "nano-turds per root canal". None of that shit makes any difference. Only the tone does.

I agree, science is useless.

Like I said, don't get wrapped up with all the crap.

Thank you for putting things in the right perspective.

 

[abbey road d enfer]

No one ever said "This song would have been a hit if that guy hadn't used 5532's on the mix bus"

That's the most threadbare argument used by ones who advocate tinkering vs. science/technology. Like the Godwin point of technical discussion.

One way of minimizing bus noise is to hit the bus harder from the channel, or do active balanced busses. Thats what I use on the FIX console. I hit the bus anywhere from 0 to +4, where -10 to -2 is the norm. That translates to easily 6 dB less noise.

But why do you care? You just said it's wasted time.

IF you are doing 8-14 channels, it's not really a problem.

The subject is 1600/800B, which came in minimum 24 channel.

 

[RogerAF]

Typical bus impedance on a 32 channel is about 200-250 ohms.

Just curious, how are you calculating that? I would not think the output resistors on the bus drivers are parallel. They each connect to a separate output stage and the bus.