MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Roland CSQ-600 - Repair/Restoration Adventure
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Roland CSQ-600 - Repair/Restoration Adventure
Sunshine Jones


Has anyone here ever gotten under the panel of a Roland CSQ-600?
I swear I had it working for a moment, and then I lifted up the circuit board to put it together for testing and it was gone again...

Upon inspection, the sequencer was stuck in load mode: metronome always playing.

I opened and inspected to find some serious corrosion from a bad, and now missing battery. Lovingly restored all the traces, and they DMM out as great contacts now, no blurred lines.

After doing this, the sequencer was then stuck in play mode.
Beeps when pressing load, but the loading doesn't happen, and the metronome never sounds.
So I replaced all of the latch related IC's - load latch, play latch, start/stop latch (mostly 4013's)

The change now is that with the corresponding levers for the 4 tracks engaged, the sequencer will switch between 1,2,3 and 4, but the clock never actually engages and it won't get out of play mode to load anything.

I'm reading the service notes and find them befuddling. Lots of explanation, but somehow it doesn't make things clear. It's as if they really struck gold here and invented something the hard way and so it doesn't make a lot of sense to my novice eyes.

I'm going to proceed and replace IC 124 (counter) IC 130 (the differential) and IC126 (the Schmitt NAND logic) but it's weird not being able to see what the thing ought to look like on a scope, or really fully understand what it actually happening here.

If anyone has revived a CSQ I'd love to talk with you about it.
Sunshine Jones
Just for fun, here is a before and after of the trace repair



At discovery




After cleaning and scratching



After repair and testing
dksynth


it looks a little bit like these weren't attached before and now are? I'm not familiar with the machine, maybe it's supposed to be connected.
Sunshine Jones
dksynth wrote:

it looks a little bit like these weren't attached before and now are? I'm not familiar with the machine, maybe it's supposed to be connected.


These are all traces which have been replaced with decent grade bus wire. The traces were eaten apart by battery acid. I cleaned it, and replaced the broken copper with good wire.

I see the points in question you've circled, and they are connected on the top of the board by a jumper. So I bus wired into the far side, and fed it back through. So it's connected as it was before, only the bus wire IS the jumper now. It's electrically the same.

This part of the board is solid. Meters out. All good connections, and correct voltages with no crossed wires.
Sunshine Jones
Well, this project is not looking good.

I've replaced the CPU, all of the IC's are now replaced and socketed (including the ram,) and the traces are corrected from the corrosion, as well as a new battery is installed (and the diode is removed to prevent the new battery from being charged.)

The new CPU brought the load latch to life, and that was encouraging.
What's been discouraging is the play latch.

I can go through the motions of loading up a sequence in any of the 4 slots, pressing reset will move the led indicator to the correct location, but the play light is now always on, and the CSQ doesn't play back.

Looking at the schematic, it looks like the CPU goes to the logic IC's, to the LED's and then into the loop for function.

So if it isn't the CPU (replaced) or the new logic IC's (replaced) then where the error is has me completely baffled.
Sunshine Jones


Today I installed a second CPU assuming that perhaps something was bad with the initial reflashed one. The results are identical.

Upon careful inspection, it seems that what's actually wrong is that the PLAY clock does not start. On the scope there is a scanning signal everywhere there should be, and the loop is complete. Yet no play clock begins when play is pressed.

Background:
All traces repaired, battery replaced, all IC's replaced (socketed) and everything tests as expected.

Today's Adventure

Today I tested the PLAY, STOP/CP, B STEP, F STEP, LOAD, and RESET switches. They are sound, and send the correct signal. The voltage is high, and it low when pressed.

All of the circuitry is working along the way, all transistors are good, all diodes are good, and the IC's are sound and their sockets are sound.

I inspected the memory position switches in the same manner. They all appear to be as expected.

The PLAY button is pressed - this sends a pulse being scanned for
And in theory this will complete a loop around the CPU. This loop appears to be intact.

I can not see at the moment what might be obstructing the play clock.

Load functions appear to be normal
Metronome functions also appear to be normal

What I have not done yet is test the sync in or out functions.

Baffling. Definitely upside down in this project at the point, but really wanting to solve the mystery and see this sequencer working again.
Driving myself a little mad.
Sunshine Jones
Some developments.

It seems there was a trace which was partially lifted which was preventing correct power from reaching about half of the IC's on the PCB.

A careless reading of the IC sockets scanned as if the situation was normal, but comparing the readings to the Service Notes and schematic revealed that not enough power was reaching the IC's and so there is the explanation for erratic LED's and IC behaviors.

The trace corrected and the sequencer sprang to life.

* partial celebration *

After spending some delighted time sequencing and allowing the internal clock to step through the 4 sequences and enjoying the CSQ-600 it was time for further testing.

1. The Din Sync clock is completely wrong.
It's sending nearly 1/2 time, and not clocking well at all. At stop, the slaved TR-606 (set to receive sync) stutters at the 1st beat until play is pressed again.

Completely wonky.

2. The CV tune is strange on several counts.

A. The actual play through it off - Though the calibration knob allows this to be pretty much corrected.

B. The CV Key play, in order to play the sequences via an external keyboard, is completely wonky. Not even really 1v/Ocy but more like 3 Octaves crammed in to one.

Again, reading the service notes there are some arcane calibration procedures to follow for times when components have been replaced - I have replaced them all - so it seems that deciphering this page of the service notes is the next step here



Calibrating the voltages for VR104, VR106 and VR107 makes sense to me.
Calibrating and verifying the KCV Function also makes sense.

The table to the top right: turning VR106 CW and CCW as compared to a DM reading is totally confusing...

I will do all that I can, looking closely at the general calibration of the CV, the Frequency Counter CP2, The Tempo Clock CP3, and the prescribed adjustments to that beautiful D/A converter.

But I don't see how this will effect the DinSync output clock...

I'll keep at this, and I'll report back with any further developments.
Sunshine Jones
Well, I opened it back up after some hours of fun sequencing and performed all of the calibration procedures.
If you look closely at the last page of the service notes (posted above) there are contact points provided for the DMM so that these calibrations may be performed.

I set the voltages and tuned the width and pitch of the sequencer as perfectly as I am able, and the performance has absolutely improved. However this has no effect on the din sync clock output.
I put a scope on the clock output, and the din sync output and discovered that they are both the same.



Reading the service notes it seems that the IC128 divides down the clock to 24 ppqn and sends this out the din jack, and then divides it again another 50% for the CR-78 clock interface.

So it seems that the division for sync 24 is not happening correctly, and something is afoot in there between the jack board and IC128 (all the way at the bottom of the circuit board.)

Looks like another loop to follow, and components to test, and traces to inspect.
Sunshine Jones
Well, I have calibrated the CSQ-600 and it's great!
But the clock output of the din sync port remains 12 ppqn.



I made the circuit:
From the 555 clock source, through the latches, into the division, to the split, to the terminal, to the jack board, to the jack...

DMM confirms continuity. Happy beeps all around.
The components are all good.
But the din sync clock output is 12 ppqn. Not 24 as it should be.

There are a few mystery adjustment potentiometers on the board, but I'm not about to start turning things to see what happens... Or am I?

I'm going to look closely now at the signature switch circuit on the sync/load section where one can switch between 4/4 and 3/4 time and see if something is weird in there.

Super frustrating that it all checks out.
Something is wrong, and I can't see it yet...
Jaytee
Is the clock divider IC one of the ones you swapped out earlier?
Sunshine Jones
Jaytee wrote:
Is the clock divider IC one of the ones you swapped out earlier?


Yes.

I've replaced, and socketed all of the IC's.
Everything is fresh, tested and new.

But since I can't see anything electrically malfunctioning (yet) it wouldn't be a terrible idea to swap these IC's out again and see what happens...
Good idea.
Jaytee
Okay, so maybe this is a long shot, but I vaguely recall a similar issue happening with Fuzzbass’ suboscillator circuit on his TTSH expander boards. He was using an older circuit design with a new logic chip and the result was two octave below the source instead of one.....or something along those lines. It turned out that the original circuit had been designed to account for flaws in the logic chip that had been fixed in modern revisions.

Reading over your issue, it sounded like kinda the same symptom, and it occurred to me that if the logic chip in question had been one that you swapped out already, it might be the same issue (new chip in old circuit working in an unexpected way).

I wish I could point you to the thread I read this in, but I have no idea which TTSH thread it was. Maybe you can contact Fuzzbass directly. In the meantime though, perhaps just try putting the original chip back in to see if anything changes?
Jaytee
Jaytee wrote:
Okay, so maybe this is a long shot, but I vaguely recall a similar issue happening with Fuzzbass’ suboscillator circuit on his TTSH expander boards. He was using an older circuit design with a new logic chip and the result was two octave below the source instead of one.....or something along those lines. It turned out that the original circuit had been designed to account for flaws in the logic chip that had been fixed in modern revisions.

Reading over your issue, it sounded like kinda the same symptom, and it occurred to me that if the logic chip in question had been one that you swapped out already, it might be the same issue (new chip in old circuit working in an unexpected way).

I wish I could point you to the thread I read this in, but I have no idea which TTSH thread it was. Maybe you can contact Fuzzbass directly. In the meantime though, perhaps just try putting the original chip back in to see if anything changes?


Edit: easier to find than I expected:https://www.muffwiggler.com/forum/viewtopic.php?p=2498371&hig hlight=#2498371

He may have actually had the inverse symptom you had, so I guess an even longer shot now, but can’t hurt to check. Just a shot in the dark.
Sunshine Jones
Well, I have suffered a serious setback today.

I went back in and re traced my din sync loop, from the OSC to the CPU, through to the division, the jumper and the jack and again everything was electrically sound. Good continuity, and all components in order.

So again I had to regard the IC's which were socketed and replaced.

First checked the sockets for continuity, and they were good.
Then replaced all of the IC's associated with the clock output for the Sync/Load function.

Well with the new IC's in place the CSQ was back where I started at the beginning of all this: beeping...

So I quickly swapped the IC's back and tried it again.
Now it's dark. Beeps at power up, and that's all it does.

Damn.
I was so close.
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group