This is my dad’s old JX3P, which has been well-kept since purchasing in 1983. It is the model without MIDI thru on the DIN board, and the main PCB is the later model 052H440C. I have found three faults with it. These are:
The main issue: I cannot write new patches either from the synth or the PG-200, regardless of the position of the write protect switch on the DIN board (ie. not even when in the ‘programmer’ position). Nor can I write to the sequencer, which still contains a sequence I wrote in 1989 (!!). When I try to write all 16 patch buttons flash.
Two other more minor issues:
- the bender range switch only switches between the WIDE and MID ranges. The MID and NARROW ranges are the same.
- The brilliance slider has no effect on the sound.
I’ve had some generous help in a mail thread (actually with a member of this forum) but I thought it would be worth opening the question here too.
So far, under guidance, I have established that the write protect signal reaches both the panel and main board correctly when activated and deactivated - I have established that the voltage changes as expected at TR14/R40 on the panel board. It also changes as expected on the main board at CN7 or R90.
Since then, I have taken out the panel board for inspection. I have dusted all the boards (none were incredibly dusty though). I have also cleaned the sliders and the volume pot (with Caig D5). A visual check revealed nothing in particular, I have tested connections on the board where I can, but I am at a bit of a loss as to how to proceed really.
Finally, there is a round green sticker on IC52 saying “JX-3P V10”. Maybe that helps…? Also, I have checked the battery voltage and it looks fine (C and D patches are recalled as expected - the ones edited before this issue occurred).
Hello,
in our mail thread you did not mention that the bender width switch does not work.
With this information it is quite clear that the problem is around switch matrix. The mid/narrow switch and the RX-Mode switch share the same Bus line (SW SC4). Assumingly the LFO button and the hold pedal do not work too. Also the Bender will work only in one direction.
Check the panel pcb for hair cracks in the traces of the lines of the switch matrix.
Regards Florian
1 Like
Hi again Florian. Yeah I discovered the bender issue after more methodical testing. Sounds like a promising analysis. The bender does work in both directions though, just not at all three ranges. Anyway I’ll take a look. Thanks!
Hmm looks like there is a small error on the panel schematic - CN6 in the schematic seems to be CN8 on the actual board. Good to know…
Ok, the M position on the bender range switch was a bit janky. Possibly oxidized? Certainly seems better after a bit of D5. Or maybe it needs replacing. Anyway, I’ll put things back together and see if that’s fixed it (I can see that the EDIT and SEQ write buttons are connected to D33 on the schematic which is connected to the switch in the medium position, so it seems likely…?).
No joy… The pitch bender is still not using the Medium range. It’s still not possible to write data, the bender pitch bends in both directions, LFO and HOLD work fine. I did check for damaged traces. I will have to investigate further. Perhaps there are some components that I can check too?
Today I checked all the diodes and all the traces in and around the switch matrix and found nothing wrong. I suppose I will have to look at the parts of the circuit beyond the matrix.That’s a lot of guesswork with my limited knowledge, but I’ll give it a go. I welcome suggestions:)
If anyone wants to take a look at this, I am referring to the service notes here: https://www.florian-anwander.de/roland_jx3p/schemo/JX-3P_SERVICE_NOTES.pdf
I’ve checked continuity from the bender switch to pin 11 of mIC34 on the main board now. I see that there is a logic table for that chip so maybe it’s worth checking the outputs of that.
In the file I linked to, there is a test point TP19 for reading the panel. Does anybody know what values you would expect to see at the test point?
I had a look at a JX-3P repair video by Mindburner and then checked the data lines on the CPU and all 1 to 8 are pulsing away, so that looks fine. He mentioned the DAC test point (TP-10) in the middle of the main board, and that it should be pulsing. I am not seeing any visible pulsing there, just a slight lift in voltage.