The thing is that the whole signal path from the mix bus to the C/R output is on the same PCB, so should not be affected by the rear con. Looking at the schemo, I tend to suspect one of the links J4/5 and J104/105 may be misplaced.Gareth Connor said:
SoundCraft 600 problem with master module
- Thread starter dario84
- Start date
Help Support GroupDIY Audio Forum:
Gareth Connor
Well-known member
I have looked into this a bit deeper and have edited my original post to reflect the results - this eliminates the confusion of surmising on what-if or maybe.
Quoting Abbey: The thing is that the whole signal path from the mix bus to the C/R output is on the same PCB, so should not be affected by the rear con.
Yes, that is true under the circumstances of the master PCB and the rear conn PCB having traces that are correctly matching.
The PCB mis-match happened when someone undertook incorrect alterations to the rear conn PCB and introduced errors. The whole thing is now 33+ years ago, so defining or stating exactly which circuit nets were messed up and for what reasons will be a tad difficult. All we have to go on is what Dario's desk has and what the designer recalls. A photo of the rear conn PCBs traces would tell me whether it was my original (correct) layout or the modified layout, however, only one side of the board is realistically possible to photograph due to the tight spacing of the jack sockets.
One thing is for definite, I was mightily miffed to return from a holiday to find that a PCB from an electrically perfectly matching pair of PCBs that I had signed-off had been worked on and fouled-up by someone who had not bothered to check what he was doing, or even follow-through his changes with a set of detailed checks that would have revealed the errors. The standard detailed checks involved circuit schematic, PCB and mechanical; ALL had to match to each other 100% to be able to sign-off a job as complete (still my practice 33+ years later). The master PCB was done on CAD and the rear conn PCB was a manual design done with Chartpack adhesive tape and pads on drafting film (oh the good old bad old days during the transition from manual design to CAD
)..... Checking of manual designs was always a time consuming job, well worth the effort, and inexcusable not to do.
I have now cross-referenced with the published schematics and checked the cross-connecting of C36 & its neighbour (C136) shown in the reply #6 pic. This is a L-R reversal of the Studio Monitor outputs. It's probable that crossing the caps was an attempt to get the "Left" and "Right" printing on the rear panel metalwork to match with the L & R signals in the console. Refer to my schem/PCB/mech check comment above. Panel printing does ring a bell as there were all sorts of quick-fire changes taking place in the console's development phase....... daft stuff such as "should the connector on the left be the left signal in relation to the guy sat at the faders or the guy at the back of the desk plugging it up?" :
My typical response: "I don't give a ****, just make a decision, document it, tell me about it in writing and stick with it! I'll make the electronics match your decision."
The crossed-over resistors are in the balanced external monitor input paths. The pairs of resistors are the fixed-level input (2-track B) and the quads are on 2-track C, the jumper linked +4dBu or -10dBV input. Again, this looks to be getting the electrical paths to match the rear panel printing in relation to L&R.
Specific recommendation for Dario:
With the console as it is, if the phase (polarity) and left and right signals through the whole of the master module and rear connectors are correct, do not remove the modifications to make it "look better". Accept the module as it is and for what it is.
It is my belief that the mods are there to make the signals appear in the correct places, i.e. to make it work as intended.
Quoting Abbey: The thing is that the whole signal path from the mix bus to the C/R output is on the same PCB, so should not be affected by the rear con.
Yes, that is true under the circumstances of the master PCB and the rear conn PCB having traces that are correctly matching.
The PCB mis-match happened when someone undertook incorrect alterations to the rear conn PCB and introduced errors. The whole thing is now 33+ years ago, so defining or stating exactly which circuit nets were messed up and for what reasons will be a tad difficult. All we have to go on is what Dario's desk has and what the designer recalls. A photo of the rear conn PCBs traces would tell me whether it was my original (correct) layout or the modified layout, however, only one side of the board is realistically possible to photograph due to the tight spacing of the jack sockets.
One thing is for definite, I was mightily miffed to return from a holiday to find that a PCB from an electrically perfectly matching pair of PCBs that I had signed-off had been worked on and fouled-up by someone who had not bothered to check what he was doing, or even follow-through his changes with a set of detailed checks that would have revealed the errors. The standard detailed checks involved circuit schematic, PCB and mechanical; ALL had to match to each other 100% to be able to sign-off a job as complete (still my practice 33+ years later). The master PCB was done on CAD and the rear conn PCB was a manual design done with Chartpack adhesive tape and pads on drafting film (oh the good old bad old days during the transition from manual design to CAD
)..... Checking of manual designs was always a time consuming job, well worth the effort, and inexcusable not to do. I have now cross-referenced with the published schematics and checked the cross-connecting of C36 & its neighbour (C136) shown in the reply #6 pic. This is a L-R reversal of the Studio Monitor outputs. It's probable that crossing the caps was an attempt to get the "Left" and "Right" printing on the rear panel metalwork to match with the L & R signals in the console. Refer to my schem/PCB/mech check comment above. Panel printing does ring a bell as there were all sorts of quick-fire changes taking place in the console's development phase....... daft stuff such as "should the connector on the left be the left signal in relation to the guy sat at the faders or the guy at the back of the desk plugging it up?" :
My typical response: "I don't give a ****, just make a decision, document it, tell me about it in writing and stick with it! I'll make the electronics match your decision."
The crossed-over resistors are in the balanced external monitor input paths. The pairs of resistors are the fixed-level input (2-track B) and the quads are on 2-track C, the jumper linked +4dBu or -10dBV input. Again, this looks to be getting the electrical paths to match the rear panel printing in relation to L&R.
Specific recommendation for Dario:
With the console as it is, if the phase (polarity) and left and right signals through the whole of the master module and rear connectors are correct, do not remove the modifications to make it "look better". Accept the module as it is and for what it is.
It is my belief that the mods are there to make the signals appear in the correct places, i.e. to make it work as intended.
dario84
Active member
I have checked the phazing and the L/R positioning of the backpannel and all is right so I guess the guy knew what he was doing..
but that leads me back to my original problem.. that the R output on the C/R speakers is lower than the left.
still confused about those 1K trimmers on the PCB.. so from what I uderstand is that those adjust the amount of voltage going to the ICs and therefor the signals can be adjusted to the same level.. is that about what is happening there or am I blowing in the wind?
I bought the original NE5532 ICs and I will try them out if you suggest that this could be the problem. one of the techs here had the opinion it might be one of the ICs or a bad Cap around them.
so from what I have researched is that the NE and the TL are the same pinout and belive that those were swaped just becouse the old ones didnt work and there was no NE-s at hand to replace them so they put the TLs in there. but TL-s have a bit of a different ratings and specs... or what is your opinion on that?
I also see alot of different TL-s from different making and different labeling like TL072-CP and ones without the CP... what is the original?
ps. there is photo of the backpannel pcb as much as I could get in there.. I'm in the middle of a project so can t really tare down the setup at the moment. but I hope it can shead some light toward new discoveries
but that leads me back to my original problem.. that the R output on the C/R speakers is lower than the left.
still confused about those 1K trimmers on the PCB.. so from what I uderstand is that those adjust the amount of voltage going to the ICs and therefor the signals can be adjusted to the same level.. is that about what is happening there or am I blowing in the wind?
I bought the original NE5532 ICs and I will try them out if you suggest that this could be the problem. one of the techs here had the opinion it might be one of the ICs or a bad Cap around them.
so from what I have researched is that the NE and the TL are the same pinout and belive that those were swaped just becouse the old ones didnt work and there was no NE-s at hand to replace them so they put the TLs in there. but TL-s have a bit of a different ratings and specs... or what is your opinion on that?
I also see alot of different TL-s from different making and different labeling like TL072-CP and ones without the CP... what is the original?
ps. there is photo of the backpannel pcb as much as I could get in there.. I'm in the middle of a project so can t really tare down the setup at the moment. but I hope it can shead some light toward new discoveries
dario84
Active member
backpannel output pcb 2
dario84
Active member
backpannel output PCB 3
These trimmers adjust the Common Mode Rejection Ratio (CMRR) of the Electronically Balanced Output Stages (EBOS). They don't change significantly the output level (maybe a tiny fraction of dB), unless teh wiring is wrong and one leg is left floating.dario84 said:still confused about those 1K trimmers on the PCB.. so from what I uderstand is that those adjust the amount of voltage going to the ICs and therefor the signals can be adjusted to the same level.. is that about what is happening there or am I blowing in the wind?
A TL072 can be substituted to an NE5532 temporarily; the main parctical difference is it is not capable of driving 600 ohm loads.
CP indicates that it is Commercial, for operation between 0° and 70°C and P that the package is plastic DIP. You may also see an additional figure; that is generally the date of manufacture or batch number.
Gareth Connor
Well-known member
dario84 said:
You kinda missed the point......
The crossed over resistors and caps are because of un-necessary changes.
If the guy who modded my signed-off designs knew what he was doing he would not have made changes: there would be no cross-over of components.
As already said, it is my belief that the mods are there to make the signals appear in the correct places, i.e. to make it work as intended.
This is not the same as the board layouts being correct.
The mods hide the guy's errors.
And yes, the two styles of track layout on the master rear conn PCB photo shows me that you have a rear conn PCB that was wrongly changed.
In answer to your reply #16:
Hard left or hard right pan will result in a 3dB (or thereabouts) level rise. It is normal for this and other consoles.
In answer to your reply #22
The 1k trimmers are associated with the balanced output circuits. Augmenting what Abbey has said, they are connected across the "+" and "-" output lines. The nature of the EBOS driver circuit, also known as a "cross-coupled output", is that the signal on each output line will be slightly different in level when measured with respect to ground. This is due to component tolerances. The trimmer sets the "equality" of the level so that the same level is present on both the "+" and "-" lines. Usually the measurement and setting is done at 1kHz.
The Control Room output is after the headphones amp. Check the headphones output. Are the levels the same here?
If yes, then the problem is one of the following EBOS circuits.
If no, then the problem is somewhere before the headphone socket.
The only thing more painful than performing customer service on a design 3 decades later, is having to explain other people's mistakes.
You should buy Gareth a beer.
JR
You should buy Gareth a beer.
JR
dario84
Active member
Gareth Connor said:
I understand that gareth! and I know what you were talking about. but I ment that the guy who made the mod knew what he was doing to fix the error. anyway i completly agree with you - it was never supposed to happen in the first place.
I will inform you all about news and solutions as soon as I find some reall time to not rush and make some progress.
I am beyond thankfull for all the help and if ever I have the privilege to meet you the beer or a dinner is on me!
dario84
Active member
Here I am again guys!
I have completly recaped the master module with Panasonic FC Low ESR series Caps 105C with original values. I hope the FC series is the right one as I have read a lot and I just decided to put those in for everything. is there any other suggestion that I should have take in consideration? will be glad to try other configurations as well!
there is no noise at all with all channels swiched off. adding them gives me more and more noise and different from each channel.. so I guess I need to recap those as well.. wich position caps should I try first and what values would be appropriete for the channels?
I changed all the opamps aswell to the original:
first I tryed the NE5532 on the original spots and the sound cleared a bit.
then I changed the other TL072cp with the new ones that I ordered and the sound lost the low end and hi freq had a certain cliping going on and I've put back the old ones and I belive they sound is more natural and no clipping.. (am I just imaganing thing or is there something to it here?)
by the way the C/R outs still the same problem... the right channel significantly lower
- I have measured the voltages with the multimeter from the outputs with an open cable it is showing me this:
L channel:
Tip - 1 half voltage
Ring - 1 half voltage
R channel:
Tip - 1 full voltage
Ring - 0 voltage
I hope you understand what I ment with 1/2 voltage and 1 full voltage ( it is relative to what I am sending from the master - but it was a 1k osciloscope wave)
also what I noticed is that the headphone volume has the same problem but only when the C/R pot is turned very low. it becomes same in the half position.
any ideas?
I have completly recaped the master module with Panasonic FC Low ESR series Caps 105C with original values. I hope the FC series is the right one as I have read a lot and I just decided to put those in for everything. is there any other suggestion that I should have take in consideration? will be glad to try other configurations as well!
there is no noise at all with all channels swiched off. adding them gives me more and more noise and different from each channel.. so I guess I need to recap those as well.. wich position caps should I try first and what values would be appropriete for the channels?
I changed all the opamps aswell to the original:
first I tryed the NE5532 on the original spots and the sound cleared a bit.
then I changed the other TL072cp with the new ones that I ordered and the sound lost the low end and hi freq had a certain cliping going on and I've put back the old ones and I belive they sound is more natural and no clipping.. (am I just imaganing thing or is there something to it here?)
by the way the C/R outs still the same problem... the right channel significantly lower
- I have measured the voltages with the multimeter from the outputs with an open cable it is showing me this:
L channel:
Tip - 1 half voltage
Ring - 1 half voltage
R channel:
Tip - 1 full voltage
Ring - 0 voltage
I hope you understand what I ment with 1/2 voltage and 1 full voltage ( it is relative to what I am sending from the master - but it was a 1k osciloscope wave)
also what I noticed is that the headphone volume has the same problem but only when the C/R pot is turned very low. it becomes same in the half position.
any ideas?
That's is 100% normal; increasing the number of stems connected to the summing amp increases its noise gain.dario84 said:
It looks like one of the legs is grounded (the one with no signal), so that may be anywhere on the PCB; check for solder bridges.
I would think it's a different issue; potentiometers do not track well at the beginning of their rotation. There's not much you can do about it, except try to find one that tracks better...
dario84
Active member
hey guys!
so after loosing my mind and going over all of it again i ended up changing the diodes and reflowing the headphone connector.. the problem went away..
now I want to balance the output amps with the 1k trimmers caouse I'm getting a slight difference on volume on both sides. where would be the best point for a probe to measure the voltage? and what should I be reading to have them ballanced? or how do I aproach this?
thanks again!
Dario
so after loosing my mind and going over all of it again i ended up changing the diodes and reflowing the headphone connector.. the problem went away..
now I want to balance the output amps with the 1k trimmers caouse I'm getting a slight difference on volume on both sides. where would be the best point for a probe to measure the voltage? and what should I be reading to have them ballanced? or how do I aproach this?
thanks again!
Dario
The trimmers are there to adjust output symmetry, not balance between channels. They are adjusted with a test jig consisting in two 10k (IIRC) resistors coming from each leg and joined; an AC voltmetter is connected at the junction, and the trimmer is adjusted for minimum signal.dario84 said:
