Soundcraft 600 output issues diagnosis help

[sam humans]

For some time I have had an issue with the left side master output on my SC600. It has been slightly distorted and lacking in bass. (The module, entire board actually, has been completely recapped, fyi) I traced the issue back to IC5A or 5B. Upon replacing 5B the problem was solved, though now the outputs are out of phase. What!!?

Listening to the output it was startlingly obvious, because I had just the left side turned up, and upon the initial listen it sounded better that it ever had. However, when I brought up the other speaker all the lows and mids disappeared unless hard panned. Sounded like phase so to double check I swapped the banana plugs in my old KRKs and sure enough, full bodied sound.

Additionally, the headphone out, which previously had the same issue as the master out, only more pronounced with the left side almost non existant, is working %100 for the first time, no phase issues. And further, the Studio out now works only out the Right side, no left at all, and the Control Room out is completely gone. Both worked, but with left side distortion, prior to the replacement of IC5B.

The schematic can be found here, 9 pages in:
http://www.soundcraft.com/downloads/fet ... des&id=666

I have a decent amount of electronics knowledge but the only tools as my disposal are a multi-meter and a soldering iron. If anyone out there is savvy enough to take a look at the schematic and see if the conditions mentioned above might point to a specific component or juncture, it would be immensely helpful.

Complicating things is that this board did, at one point, have some sort of proprietary automation installed in it. It is my beliefe that the problem has it's origins in this, and is what led me to tracing the signal from the output back to IC5B, removing mods and checking for changes after each removal. I was thrilled when the replaced IC yielded positive results! The strange phase issue, however, has me scratching my big dumb head.

I know it's along shot, but any help would be appreciated. Thanks!
-S

[Nick Sevilla]

Hi Sam,

The only way to know where the phase got reversed in your Left output, is to check each Left output versus the Right, by recording a sine wave through the board. Include the Headphone output in you test, since it's tap if close to IC8. IC15 is your RIGHT C Room output, not the left side at all.

IC5 is your Mix Left output, and has a lot to do with your Control Room outputs.

So you need to make a recording in your DAW to determine what is going on, and eliminate the outputs which are correct.

Is the Mix Output flipped? Then the issue is at the IC5 wiring or the IC itself.
Are any of the Tape Outputs flipped? Then the issue is in the wiring there.
Is the output flipped at the Studio Output? check that part.
Is the output flipped only at the C Room? check that part.

Once you isolate the specific Left output(s) which are phase reversed, you can then determine exactly where your signal is flipped.

My guess is that IF only the C Room output is flipped, and nothing else, the. whatever IC you put into the socket for IC8 is reverse polarity from the original. Did you use an EXACT replacement, or one that is "close enough"?

[Jim Williams]

There is no phase inversion in that design. No opamp would reverse it either, that's why they are called "operational amplifiers", the design sets polarity, not the part.

I would look to what ever someone did in there, sounds like they messed it up pretty good.

[sam humans]

Is the Mix Output flipped? Then the issue is at the IC5 wiring or the IC itself.
Are any of the Tape Outputs flipped? Then the issue is in the wiring there.
Is the output flipped at the Studio Output? check that part.
Is the output flipped only at the C Room? check that part.

Once you isolate the specific Left output(s) which are phase reversed, you can then determine exactly where your signal is flipped.

My guess is that IF only the C Room output is flipped, and nothing else, the. whatever IC you put into the socket for IC8 is reverse polarity from the original. Did you use an EXACT replacement, or one that is "close enough"?

Hi Nick, thanks for the helpful replies.

First, yes I used the exact same part. It's a Texas Instruments 5532P in there, that's what I replaced it with, albeit a much newer one. Made by TI though, not JRC or anyone else.

2nd, the output that flipped upon replacing IC5B was the Mix out. I see that the tap for the main out comes first in the chain, which is what led me to swap the IC's in the first place, but doing so had so many unexpected ripple effects that I'm having a hard time understanding what's going on.

Here's what's up, with it's condition pre-replacement of 5B in parentheses.

Mix Out: Phase reversed (used to have week left signal)
Studio Out: Right side only, no left output present. (used to be OK)
Control Room Out: no signal at Right or Left. (used to have very distorted, almost non existent Left output, but Right side was fine)
Headphones: working normally for the first time. (used to have very distorted, almost non existent Left output, right side was fine).

If this clears things up and/or shines a light on where the issue might be, let me know.

Otherwise, I will keep at it today, moving backwards from IC5 A/B and see what other mods I run into. I think my main issue, really, is deciphering the schematic. It's much more complex than what I'm used to reading, and the simple task of following the signal path from schematic to PCB is where I sometimes get very, very lost. Doesn't help that it's a dense, partially modded, 2 sided PCB with a million vias. Ugg.

thanks again,
-S

[sam humans]

A few things of note I've discovered today:

R39 and 139 were both 1/2 removed. That is, the resistor was conveniently left on the board, but one side was removed from contact. Upon reconnecting R39, the Right side of the headphone out became distorted, where-as before connection it was clean on both sides. Also, the CR out now works on the Left side only, and it is very distorted... where-as before connection of these 2 resistors it was not working at all.

Previously the Studio out was working Right side only, but clean. Still right side only, but after re-connection of R39/139, it is now distorted.

Also, before connecting R39/139 I noticed that turning the Studio volume up all the way resulted in a reduction in volume and increased distortion from 3 o'clock on. Now it works without that issue, but as described above.

[Nick Sevilla]

Hi Sam,

I would first get your Left and Right Mix outputs working.
Ignore the rest until you get those working properly.

If the PCB is correct on against the schematic, then you HAVE to trace wires from the edge connector throught to the output connector on the back.

Slow painful boring work, but hey, when you're done, you'll know your board better.

As to messing with anything that seems "extra", DON'T! at least not until you have the console working right again. There might be a reason for those mods, like broken pcb leads or who knows what crap.

As an aside, I once had to get an old Neve 8048 console back into service.
We took out over 35 lbs of extra wiring from a "Mod" gone horrifically wrong.
6 months later, we had the only 48 inline Neve on the US West Coast which worked 100%.

Perserverance is key. That, or you can send it to Jim Williams and have him curse at it for a while...

[sam humans]

Slow and steady, I'll be putting in the old sweat equity (that soldering iron's hot you know!) over the next few days, and will check-in with nay further head-scratchers.

I just noticed that the PCB is missing 6 47uF caps. I had the master section recapped by a local shop, so I'm not sure if these caps have been taken out for a reason by Soundcraft or someone else... or if they were simply missed during the recap. They are all physically located in a row, and electrically located in the same spot connected to the outputs of an NE5532: C22, C122, C43, C143, C35, C135.

Any reason why these would have been removed and/or was this a mistake? What purpose do these caps serve, electrically speaking?

thx!
-S

[dfuruta]

They are all physically located in a row, and electrically located in the same spot connected to the outputs of an NE5532: C22, C122, C43, C143, C35, C135.

Any reason why these would have been removed and/or was this a mistake? What purpose do these caps serve, electrically speaking?

These caps block DC, to prevent an offset on the output and to prevent frying the opamp should the user do something stupid (like putting phantom power into the output connector). Seems like a mistake that they were removed.

[The Scum]

Are the missing cap locations just totally empty? Or have they been replaced with wire jumpers?

If they're simply empty, they were probably forgotten. If they were jumpered, some people consider that an optimization...see the previous post for a caveat.

I'm on the road right now, and the schems are unreadable on my iPod. I'll give it a closer look when I'm home. Those cross-coupled output drivers can do screwy things if they're not terminated correctly.

How do you have things connected? Balanced or unbalanced? If unbalanced, you're grounding the cold conductor, right?

[sam humans]

Are the missing cap locations just totally empty? Or have they been replaced with wire jumpers?

...

How do you have things connected? Balanced or unbalanced? If unbalanced, you're grounding the cold conductor, right?

First question: they were just empty, no jumpers. I put caps in this afternoon, but it made no difference as far as the outputs were concerned.

Second question: All balanced, at least as far as the Mix and Control Room outs are concerned. Interesting you should ask though, as when jacking the TRS CR outs into my Duet to check output/phase I noticed that if I went in to just the first click I got a strong signal, though when pushed in all the way the signal became weak and distorted. Which, if I'm wrapping my head around this correctly, means I might be looking for a ground short somewhere in the sleeve of the CR outs?

Thanks everyone for the help! It's great to just get a little feedback and talk some of this out. The more I trace back from the main outs, the more I discover. Kinda bummed that someone so thoroughly trashed this PCB, though I'm sure that once I'm done (fingers crossed!!) I'll be happy to have a working board, and getting my trouble-shooting skills up to par is something that has long been on my list.

-S

[The Scum]

Since things are wacky, lets back up a stage or two.

How healthy is the signal at the insert points?

[The Scum]

OK, I've got a screen big enough that I can see the schems.

The missing caps are the AC coupling caps on the negative outputs of the output balancing amps. Removing those caps breaks the feedback path that is essential for those amps to work properly.

Are their conjugate pair caps (C21, C121, etc) on the positive outputs in place? Are all of the other components around those opamps in as well? I suppose it's possible that it has been modded to be purely unbalanced.

[sam humans]

Far as I can tell the signal at the inserts is healthy. That is, the send is the send, the return is the return, and it the phase issue remains in the Mix outs.

Also, all other caps are present and accounted for. There are some other missing parts, however. Specifically, what on the PCB is labeled F1 and F101, which (annoyingly) are called TR11 and TR13 on the schematic. These are FET transistors in the PFL/AFL switching network. In this case they have obviously been removed on purpose, as 2 of the legs have been tied together on both. Also, these connect directly to the resistors I found half removed from the circuit, R39 and R139, which in turn tie directly in to the A and B posts of the Studio Speakers switch.

Why would TR12 and TR14 be left in, while TR 11 and TR13 were removed? The schematic calls for transistor type J112 in those positions, though what's actually in there are 2N5459's. (I do have some 2N5458's left over from a previous project, though can find little info as to whether these might work in that application.)

At this point my intuition tells me that the problem lies in the PFL/AFL switching network. It would account for why the assorted problems I have encountered so far are slightly different at each output, as of course the PFL/AFL is setup to leave the Mix out uninterrupted while routing the solo'd signal to the CR/Phones and Studio outs.

SO!

Should I try and replace those 2 missing FETS?
Will a 2N5458 work in place of what should be a J112 (but is actually a 2N5459)?
Last, is it possible that those missing caps damaged the ICs around which they are placed? If so, would that account for why IC5B was dead?

[The Scum]

I have 3 thoughts.

1> This thing seems to be a bit of a basket case. Without knowing how and why it's been modded, it's very hard to know how to maintain it. Do you know how it was being used perviously?

It sounds like they subverted the solo somehow...maybe making a preview feature for live broadcast (so they can cue the level for a source in the context of the mix, then enable it by assigning it to the 2-bus.). But that's a flying SWAG based on the status of a few JEFTs...When you solo a channel, does it put that source on top of the mix, rather than muting the mix?

How do you feel about taking it back to stock? It's not like J112's are hard to find.

2> Work your way through it in the order of signal flow. Get the 2-mix outputs working 100%, then move on to the solo logic. If that output is wonky, then everything downstream will feel it's effects: The CR switching selects the line amp output.

3> If you're serious about maintaining this beast, a $50 Craigslist oscilloscope will prove very worthwhile.

[sam humans]

I have 3 thoughts.

1> This thing seems to be a bit of a basket case. Without knowing how and why it's been modded, it's very hard to know how to maintain it. Do you know how it was being used perviously?

Thanks Scum for all your help... basket case indeed! My understanding of how it was previously used was with some sort of mute automation. And no, the PFL/AFL system was not working as designed, nothing would happen. That is until today, when the J112s arrived in the mail...

- Now PFL/AFL works, but Left side only.
- Control Room out now works as designed, but is susceptible to noise from the Studio Speakers switch. That is, when turning up the Studio Speakers to near full volume, it causes the CR outs to distort.
- Studio Output works Left side only
- Mix Out is still out of phase

As far as I can tell, I have returned it to stock condition. Obviously there are still a few issues to work out, principally finding where the Phase is swapped and why the Studio Outs not work. Any input on that is greatly appreciated!

[/list]