sweetbeats
Reel deep thoughts...
Yes, there are artifacts in the audio band, which is why the meter shows -20VU with the fader all the way down...very noisy noise floor when the module is oscillating which is why I care about it.
Tascam M-___ Story...
- Thread starter sweetbeats
- Start date
Yes, there are artifacts in the audio band, which is why the meter shows -20VU with the fader all the way down...very noisy noise floor when the module is oscillating which is why I care about it.
sweetbeats
Reel deep thoughts...
Success!
This is a pic of a flatlined oscilloscope hooked up to one of the I/O modules.
Why does it matter that the scope trace is flat? Because I've been trying and trying to get the thing to oscillate and I can't. Why won't it oscillate? Because I think I've finally tackled this issue. I executed the steps as intended detailed in an earlier post, and achieved success at step 1...replacing the first amp stage on the EQ board with a new TI-branded TL072P. This one...
This is a pic of a flatlined oscilloscope hooked up to one of the I/O modules.
Why does it matter that the scope trace is flat? Because I've been trying and trying to get the thing to oscillate and I can't. Why won't it oscillate? Because I think I've finally tackled this issue. I executed the steps as intended detailed in an earlier post, and achieved success at step 1...replacing the first amp stage on the EQ board with a new TI-branded TL072P. This one...
The Ghost of FM
Banned
Good stuff indeed! 
Cheers!
Cheers!
sweetbeats
Reel deep thoughts...
Thanks fellas! Pretty jazzed about it. Now to do the rest of the modules (except #1 and #2...those already have "upgraded" opamps).
I'm also trying to decide whether or not I should socket and replace the other JRC parts with TI parts, and I'm doing a little experimenting with changing the opamp that drives the direct out and handles input buffering from the stock 4556 to a 5532. This is something I've been thinking about for a long time. The output DC offset increases with the 5532 in place so I need to test and see if that effects the noise floor. If it doesn't, I will likely install 5532s there on each module. I'm not a fan of wholesale "upgrades" like that, but I have an affinity for the 5532 (I've consistently liked the sound of circuits utilizing the 5532), it's considered practically vintage a this point but still holds it's own spec-wise in my opinion...so using it isn't the "gimme the fastest bestest chip in the universe" syndrome, but could be a nice change since anything going through the input to the module goes through that opamp...it's a critical component, and I consider the 5532 to be "better" than the 4556...and I just like it, and that location calls for an opamp that has some guts which the 5532 has.
I figure I'll get these decisions made and implemented since I need to pull those modules apart again to replace that one opamp on each EQ board anyway.
I'm also trying to decide whether or not I should socket and replace the other JRC parts with TI parts, and I'm doing a little experimenting with changing the opamp that drives the direct out and handles input buffering from the stock 4556 to a 5532. This is something I've been thinking about for a long time. The output DC offset increases with the 5532 in place so I need to test and see if that effects the noise floor. If it doesn't, I will likely install 5532s there on each module. I'm not a fan of wholesale "upgrades" like that, but I have an affinity for the 5532 (I've consistently liked the sound of circuits utilizing the 5532), it's considered practically vintage a this point but still holds it's own spec-wise in my opinion...so using it isn't the "gimme the fastest bestest chip in the universe" syndrome, but could be a nice change since anything going through the input to the module goes through that opamp...it's a critical component, and I consider the 5532 to be "better" than the 4556...and I just like it, and that location calls for an opamp that has some guts which the 5532 has.
I figure I'll get these decisions made and implemented since I need to pull those modules apart again to replace that one opamp on each EQ board anyway.
sweetbeats
Reel deep thoughts...
I feel like I've come a long way. This mess is the master section pulled apart again:
I pulled it apart because I wanted to see if there might be some critical opamps in there to replace, or at least have ready replacements on-hand and enough sockets on-hand in case anything does need replacing in the future. Plus there has been some funkiness with the headphone amp.
Anyway, the last time I pulled apart the master section it was just really intimidating...lots of mystery (to me) circuitry and prototype jankiness to boot. This time? Totally comfortable, and it took me well under a couple hours to ID the function of the 24 dual opamps and 2 single opamps. This will also help for future troubleshooting. It's to the point where I can just look at the trace side and I know what the components are on the other side without looking...and any looking is just to confirm what I suspect is on the other side of those solder lands. Takes a lot of the mystery out of the M-__, which gives some strength to what it lacks in practicality (because it's one-of-a kind, and there exists no technical documentation).
Anyway, I'm excited in a geeky way.
I pulled it apart because I wanted to see if there might be some critical opamps in there to replace, or at least have ready replacements on-hand and enough sockets on-hand in case anything does need replacing in the future. Plus there has been some funkiness with the headphone amp.
Anyway, the last time I pulled apart the master section it was just really intimidating...lots of mystery (to me) circuitry and prototype jankiness to boot. This time? Totally comfortable, and it took me well under a couple hours to ID the function of the 24 dual opamps and 2 single opamps. This will also help for future troubleshooting. It's to the point where I can just look at the trace side and I know what the components are on the other side without looking...and any looking is just to confirm what I suspect is on the other side of those solder lands. Takes a lot of the mystery out of the M-__, which gives some strength to what it lacks in practicality (because it's one-of-a kind, and there exists no technical documentation).
Anyway, I'm excited in a geeky way.
sweetbeats
Reel deep thoughts...
So I got all of the opamp functions identified in the master module, and identified a few that I want to replace with some different parts.
The headphone amp section periodically still makes some random sputtering noises for awhile until the console has been powered for awhile. The amp circuit is really simple and is the same as found in a number of other Teac units and is based on the 386 chip. The opamp is the only active device in the circuit and I suspect the source of the noise might be there so I'm replacing those with new TI branded 386 parts.
The overload LED comparators for the stereo buss are mismatched...one is a 339 and the other is a 2903. These *are* equivalent parts, but I think they should match. Hair-splitting? You betcha. But I've got it all pulled apart so making it "right" now makes sense. I'm putting new TI 2903 parts in there.
The output drivers on the balance amp boards for the balanced stereo output are 072s. I think 072s are a fine part. They've been *widely* used *everywhere* and *continue* to be used even today. There are even examples of "holy grail" equipment out there based around the 072. But the 2134 is an easy and popular and reliable drop-in substitute for good reason, I have them on-hand, and this is a critical amp stage, so I'm putting those in there.
The interstage stereo buss drivers are 5532s. All good there as far as I'm concerned, but what's on the boards I'm not even sure of the brand. They look like very early parts and they would be at this vintage. So I'm swapping in new TI branded 5532 parts. Hey...I'm in there already right?
The control room output driver is a 4556. I'm doing the same thing there as I'm doing on the input/output modules with the input buffer/direct out driver and putting in a 5532 there. The control room buss is a critical buss for mixing. Along those lines I'm also replacing the 082 control room interstage driver with a 2134 for the same reason.
There is quad logic switching chip on one of the boards that has corrosion on it's legs from the historical rodent pee. Putting a new TI part there too.
Back to the control room buss for a moment...here's where things get interesting...there is a super-duper prototypey lookin kludge on the backside of the control room interstage driver...it's always freaked me out. So I now know all that is is the driver for the MONO circuit on the control room buss...it's a simple summing amp circuit. But it's crazy looking, and clearly a bizarre afterthought. The control room PCB is the same as the studio PCB in the master module...they should have made a different PCB just for the control room board but this is prototype-land and they didn't, and as a result kludged in this mono summing amp. There is a vacant DIP-8 site on the control room PCB that is populated on the studio PCB. I'm going to modify traces and locate the summing amp there rather than it hanging in space like it was.
Here is the cluster mess that is the kludged mono summing amp:
At the tip of my thumb there is the vacant DIP-8 site to where I want to relocate the summing amp. This location on the studio PCB is where the summing amp for the two ECHO RETURN inputs lives...dual opamp, one channel per return channel, so that's why the site is not populated on the control room PCB:
Naturally the work will require cutting traces and jumpering connections, adding filter caps etc., but I reverse-engineered a schematic for how the mono summing amp is connected to the control room buss interstage driver mentioned above, and though in the end it won't look as nice as factory, it'll be a far-sight better than the krazy kluster pictured above and ultimately a more reliable installation. And there is already a fair amount of cut and jumpered traces throughout this console; part of it's character.
The mono summing amp schematic sketch...I may re-draw at some point as this was the first draft "sketch-as-you-go" version as I was working through the circuit:
Here's the collection of parts I've pulled from the master module:
I've done all I can at this point until parts arrive, but with a couple of exceptions I'm using good quality machine-pin type sockets for any opamps that are being changed out, so I've got those installed ready for the replacement opamps to be plugged in when they arrive.
There are some other repairs I'm doing that I think I saw in the past but ignored not feeling confident to be messing with things...a couple wire junctions with electrical tape as insulation (going to replace with heat-shrink)...
And a number of components upon which I see corrosion from historical rodent pee. I want this thing to be reliable, so I'm replacing anything that looks like these parts with the blue-green death on them:
So that's the update. Stay tuned. I'll complete these modifications/repairs on the master module and then get back to those select modifications on the input/output modules and then finally complete the comprehensive testing that got interrupted when the oscillation issue was discovered.
The headphone amp section periodically still makes some random sputtering noises for awhile until the console has been powered for awhile. The amp circuit is really simple and is the same as found in a number of other Teac units and is based on the 386 chip. The opamp is the only active device in the circuit and I suspect the source of the noise might be there so I'm replacing those with new TI branded 386 parts.
The overload LED comparators for the stereo buss are mismatched...one is a 339 and the other is a 2903. These *are* equivalent parts, but I think they should match. Hair-splitting? You betcha. But I've got it all pulled apart so making it "right" now makes sense. I'm putting new TI 2903 parts in there.
The output drivers on the balance amp boards for the balanced stereo output are 072s. I think 072s are a fine part. They've been *widely* used *everywhere* and *continue* to be used even today. There are even examples of "holy grail" equipment out there based around the 072. But the 2134 is an easy and popular and reliable drop-in substitute for good reason, I have them on-hand, and this is a critical amp stage, so I'm putting those in there.
The interstage stereo buss drivers are 5532s. All good there as far as I'm concerned, but what's on the boards I'm not even sure of the brand. They look like very early parts and they would be at this vintage. So I'm swapping in new TI branded 5532 parts. Hey...I'm in there already right?
The control room output driver is a 4556. I'm doing the same thing there as I'm doing on the input/output modules with the input buffer/direct out driver and putting in a 5532 there. The control room buss is a critical buss for mixing. Along those lines I'm also replacing the 082 control room interstage driver with a 2134 for the same reason.
There is quad logic switching chip on one of the boards that has corrosion on it's legs from the historical rodent pee. Putting a new TI part there too.
Back to the control room buss for a moment...here's where things get interesting...there is a super-duper prototypey lookin kludge on the backside of the control room interstage driver...it's always freaked me out. So I now know all that is is the driver for the MONO circuit on the control room buss...it's a simple summing amp circuit. But it's crazy looking, and clearly a bizarre afterthought. The control room PCB is the same as the studio PCB in the master module...they should have made a different PCB just for the control room board but this is prototype-land and they didn't, and as a result kludged in this mono summing amp. There is a vacant DIP-8 site on the control room PCB that is populated on the studio PCB. I'm going to modify traces and locate the summing amp there rather than it hanging in space like it was.
Here is the cluster mess that is the kludged mono summing amp:
At the tip of my thumb there is the vacant DIP-8 site to where I want to relocate the summing amp. This location on the studio PCB is where the summing amp for the two ECHO RETURN inputs lives...dual opamp, one channel per return channel, so that's why the site is not populated on the control room PCB:
Naturally the work will require cutting traces and jumpering connections, adding filter caps etc., but I reverse-engineered a schematic for how the mono summing amp is connected to the control room buss interstage driver mentioned above, and though in the end it won't look as nice as factory, it'll be a far-sight better than the krazy kluster pictured above and ultimately a more reliable installation. And there is already a fair amount of cut and jumpered traces throughout this console; part of it's character.
The mono summing amp schematic sketch...I may re-draw at some point as this was the first draft "sketch-as-you-go" version as I was working through the circuit:
Here's the collection of parts I've pulled from the master module:
I've done all I can at this point until parts arrive, but with a couple of exceptions I'm using good quality machine-pin type sockets for any opamps that are being changed out, so I've got those installed ready for the replacement opamps to be plugged in when they arrive.
There are some other repairs I'm doing that I think I saw in the past but ignored not feeling confident to be messing with things...a couple wire junctions with electrical tape as insulation (going to replace with heat-shrink)...
And a number of components upon which I see corrosion from historical rodent pee. I want this thing to be reliable, so I'm replacing anything that looks like these parts with the blue-green death on them:
So that's the update. Stay tuned. I'll complete these modifications/repairs on the master module and then get back to those select modifications on the input/output modules and then finally complete the comprehensive testing that got interrupted when the oscillation issue was discovered.
Last edited:
sweetbeats
Reel deep thoughts...
Well, all the work is done. Still need to triple check my work and then put the master module back together and then test. But the work is done. The work I did to relocate the control room buss mono summing amp...I honestly can't say whether or not it was worth it...too close to the work at the moment to be objective. I do think what I've done is better than the kludge that existed before, it's just that I'm not sure if there is a more elegant solution than what I ended up doing...like using proto board and mounting it somewhere on the control room PCB. I did consider that option but mounting was going to be a challenge and then that assembly would cover a good number of components on the board...anyway, here is the before:
And here is the after:
Some closeups...where the little orange blob is there in the middle is where the kluster kludge was located:
And here is the formerly vacant DIP-8 site now populated...
And here is the component side...the cluster of components there to the right of the "CONTROL ROOM" script is what has been relocated, and includes some necessary additional components as well:
I feel pretty good about replacing the components that were corroded from rodent pee...there was still dried rodent pee crust in the holes in the PCB for all the leads of all those components...got that allllll cleaned up...and I found a polystyrene cap in the feedback loop of the monitor buss output driver that was about 400% out of spec, and I suspect it was from bein' peed on.
And here is the after:
Some closeups...where the little orange blob is there in the middle is where the kluster kludge was located:
And here is the formerly vacant DIP-8 site now populated...
And here is the component side...the cluster of components there to the right of the "CONTROL ROOM" script is what has been relocated, and includes some necessary additional components as well:
I feel pretty good about replacing the components that were corroded from rodent pee...there was still dried rodent pee crust in the holes in the PCB for all the leads of all those components...got that allllll cleaned up...and I found a polystyrene cap in the feedback loop of the monitor buss output driver that was about 400% out of spec, and I suspect it was from bein' peed on.
The Ghost of FM
Banned
Holy forensic crime scene evidence removal! 
I admire your acquired skill set to pull all that off! This is way advanced DIY stuff here!
Now.
Will it all work when you plug it back in??!!
Fingers crossed for ya!
Cheers!
I admire your acquired skill set to pull all that off! This is way advanced DIY stuff here!
Now.
Will it all work when you plug it back in??!!
Fingers crossed for ya!
Cheers!
sweetbeats
Reel deep thoughts...
You bet I'm hoping so.
I've triple checked the work against the schematic I drew, against the guts I removed, and against common sense/logic. It looks good. Hopefully have it all back together in the next day or two and take it for a spin. I expect something not to work right which is healthy preparation.
I forgot to mention earlier I found an error on the studio PCB; the orientation for the filter cap for the -15V rail at the echo return input buffer amp was screened on the PCB backwards. I don't know if the original cap was installed backwards or not, but when I recapped it I followed the screening on the board so that new cap has been on there backwards. I'm surprised it's lasted this long, though I suppose the current flow is pretty low there. At any rate, it's correct now, and I've checked all the others to make sure they are installed correctly according to the circuit, whether or not the screening matches. I'm going to do that on the input/output modules as well. That's one thing I've noticed about this console...often there will be local filter caps where power enters a PCB, sometimes not. It's a good idea though and most often you will find local filter caps installed. Teac put 100uF caps at almost EVERY *opamp*...like not just global local filter caps, but a set of filter caps for EACH OPAMP. Never seen that before, and 100uF is pretty hefty ...on the M-__sometimes a pair of opamps will share a set of filter caps, but thats the exception...so anyway, just one more piece of evidence that the engineers were allowed to have their way on this console.
I've triple checked the work against the schematic I drew, against the guts I removed, and against common sense/logic. It looks good. Hopefully have it all back together in the next day or two and take it for a spin. I expect something not to work right which is healthy preparation.
I forgot to mention earlier I found an error on the studio PCB; the orientation for the filter cap for the -15V rail at the echo return input buffer amp was screened on the PCB backwards. I don't know if the original cap was installed backwards or not, but when I recapped it I followed the screening on the board so that new cap has been on there backwards. I'm surprised it's lasted this long, though I suppose the current flow is pretty low there. At any rate, it's correct now, and I've checked all the others to make sure they are installed correctly according to the circuit, whether or not the screening matches. I'm going to do that on the input/output modules as well. That's one thing I've noticed about this console...often there will be local filter caps where power enters a PCB, sometimes not. It's a good idea though and most often you will find local filter caps installed. Teac put 100uF caps at almost EVERY *opamp*...like not just global local filter caps, but a set of filter caps for EACH OPAMP. Never seen that before, and 100uF is pretty hefty ...on the M-__sometimes a pair of opamps will share a set of filter caps, but thats the exception...so anyway, just one more piece of evidence that the engineers were allowed to have their way on this console.
sweetbeats
Reel deep thoughts...
Well, good news is there was no smoke.
The not as good and somewhat expected news is that the master section isn't 100%. What I've tested so far is working just fine except the control room buss is, like, FULL throttle...no way to attenuate it and it's just kuh-RANKED coming out of the headphones and the control room out jacks. So I'll have a look. Not too worried about figuring it out.
And the headphones still sputter for awhile upon powering the console.
Oh well. I'll take another look at that and maybe do things proper and scope the signal path and find out where it starts.
Anyway, just thought I'd give a quick report that nothing caught fire.
The not as good and somewhat expected news is that the master section isn't 100%. What I've tested so far is working just fine except the control room buss is, like, FULL throttle...no way to attenuate it and it's just kuh-RANKED coming out of the headphones and the control room out jacks. So I'll have a look. Not too worried about figuring it out.
And the headphones still sputter for awhile upon powering the console.
Oh well. I'll take another look at that and maybe do things proper and scope the signal path and find out where it starts.
Anyway, just thought I'd give a quick report that nothing caught fire.
The Ghost of FM
Banned
sweetbeats
Reel deep thoughts...
Hmmm...
I'm suspecting I made a rash assumption that because I was able to sub in 5532s for 4556s on the input/output modules, that I could do the same on the master module (which I did with the control room buss output driver). The 5532 is NOT always a drop-in replacement. I didn't do due-diligence and inspect the circuit first. I haven't tested it yet, but I put a new 4556 back in that spot and I have a hunch it may take care of some or all of the issues I'm experiencing.
I think I need to fabricate an extender cable so I can power the master module outside if the mixer frame...
I'm suspecting I made a rash assumption that because I was able to sub in 5532s for 4556s on the input/output modules, that I could do the same on the master module (which I did with the control room buss output driver). The 5532 is NOT always a drop-in replacement. I didn't do due-diligence and inspect the circuit first. I haven't tested it yet, but I put a new 4556 back in that spot and I have a hunch it may take care of some or all of the issues I'm experiencing.
I think I need to fabricate an extender cable so I can power the master module outside if the mixer frame...
altruistica
Member
Hey Cory,
I don't think I've ever read this thread before......
You deserve a medal man.......2008-2014, I really do salute your perseverance and staying power to see something through. I think I know why you do it though, because just on removing the heads from my MSR16 yesterday, I'm completely infatuated with the beautiful engineering of the machine. Every part of it is a wonder and I think we develop relationships with inanimate objects because of the love that went into its design (no I'm not smoking anything funny....it's just how I see it these days).
Al
I don't think I've ever read this thread before......
You deserve a medal man.......2008-2014, I really do salute your perseverance and staying power to see something through. I think I know why you do it though, because just on removing the heads from my MSR16 yesterday, I'm completely infatuated with the beautiful engineering of the machine. Every part of it is a wonder and I think we develop relationships with inanimate objects because of the love that went into its design (no I'm not smoking anything funny....it's just how I see it these days).
Al
sweetbeats
Reel deep thoughts...
On the rrrrroad again... (Yes you should sing it like Willie Nelson)
The Ghost of FM
Banned
Was it disassembled to make the weight more manageable to lug back and forth from the van to the house?
Cheers!
Cheers!
sweetbeats
Reel deep thoughts...
Yyyyyep! I'm moving it solo. If I recall correctly the empty frame alone is about 80lbs.
The Ghost of FM
Banned
I figured so!
I can still remember when I picked up my MS-16 from the seller's house. Took it all apart and moved it piece by piece... Even with my brother there to help me!
Any other goodies bound for the back of your van?
Cheers!