I recently acquired a Roland JV-90 as a 'fixer-upper' project. After replacing multiple front panel switches and other minor faults, I'm left with one that stumps me: Dead notes. I've tried the go-to fix of cleaning the circuit board and contact strips with isopropyl alcohol but to no avail. As a test I tried triggering the notes with a different contact and it didn't work, still no sound. I checked the synth engine over MIDI and it responds over the full 76-note range including the dead notes. Physically, the keys for the dead notes look OK. Seeing how the notes didn't respond when I tried using different contacts, I don't think it's the keys. I tried a full system reset in the hope that it might fix anything software related but it also made no difference.
Has anyone experienced anything similar and, if so, do you know any fixes that work?
JV-90 dead notes
Re: JV-90 dead notes
Hi
Many of these failures are solder joints going bad or PCB tracks cracking with the flexing of the board and all that can be checked with a continuity tester usually found in the resistor section of a multi-meter.
So although I haven't worked on a JV-90 I have worked on keyboards before and after these were made.
Most keyboards are scanned in blocks of chromatic keys. Usually from the low notes and 8 at a time.
Common trick. If there are ICs in sockets associated with the key scanning on the CPU board, push down on them and they may go click and break the oxide that has built up and reconnect to the IC socket.
Removing the IC and putting it back may be enough to scrub the contact on the pins to make connection.
But the best way is to remove them (one at a time and don't forget which way they go ), clean the IC pins with De-Oxit and put a 'little' into the socket.
Common places to look..,
1) for example, if note 4 from the bottom doesn't play, along with note 12 and 20 and 28 etc then something is wrong with connection at the CPU or buffer before it on the 4th of 8 line.
2) If the dead notes are all grouped in a block of 8 then that group is not being selected and there will be a a couple of chips like the 40H138 3-to-8 line decoder/demultiplexer. One of those has a dry joint or connection that is broken or the IC is an ex-chip.
3) if the dead notes are 'random' then there will be multiple dry joints, broken tracks etc.
If you find a problem at a solder joint, you can just try and re-solder it
Best of luck. Hope this helps
Royce
Many of these failures are solder joints going bad or PCB tracks cracking with the flexing of the board and all that can be checked with a continuity tester usually found in the resistor section of a multi-meter.
So although I haven't worked on a JV-90 I have worked on keyboards before and after these were made.
Most keyboards are scanned in blocks of chromatic keys. Usually from the low notes and 8 at a time.
Common trick. If there are ICs in sockets associated with the key scanning on the CPU board, push down on them and they may go click and break the oxide that has built up and reconnect to the IC socket.
Removing the IC and putting it back may be enough to scrub the contact on the pins to make connection.
But the best way is to remove them (one at a time and don't forget which way they go ), clean the IC pins with De-Oxit and put a 'little' into the socket.
Common places to look..,
1) for example, if note 4 from the bottom doesn't play, along with note 12 and 20 and 28 etc then something is wrong with connection at the CPU or buffer before it on the 4th of 8 line.
2) If the dead notes are all grouped in a block of 8 then that group is not being selected and there will be a a couple of chips like the 40H138 3-to-8 line decoder/demultiplexer. One of those has a dry joint or connection that is broken or the IC is an ex-chip.
3) if the dead notes are 'random' then there will be multiple dry joints, broken tracks etc.
If you find a problem at a solder joint, you can just try and re-solder it
Best of luck. Hope this helps
Royce
Re: JV-90 dead notes
Hello Royce!
Thank you for your suggestions. I think it might be a variation of option 1 in my case as the dead notes are: D#3, B3, G4, D#5, B5, G6. which are all at 8 semitone intervals. Oddly, all the notes below C3 work just fine. Do I need to look for a broken track on the PCB? FWIW, the ribbon from the upper contact boards to the main PCB looks a bit mangled along one edge.
All [except one] of the IC's are surface mounted. The one that isn't surface mounted has it's legs going through holes in the PCB and is soldered down on the back. So, nothing socketed :-\
Thank you for your suggestions. I think it might be a variation of option 1 in my case as the dead notes are: D#3, B3, G4, D#5, B5, G6. which are all at 8 semitone intervals. Oddly, all the notes below C3 work just fine. Do I need to look for a broken track on the PCB? FWIW, the ribbon from the upper contact boards to the main PCB looks a bit mangled along one edge.
All [except one] of the IC's are surface mounted. The one that isn't surface mounted has it's legs going through holes in the PCB and is soldered down on the back. So, nothing socketed :-\
Re: JV-90 dead notes
So every 8th note and some below C3 work = it is most likely a break in the track (above C3) of one of the 8 key data lines (not one of the group of 8 selectors).
WIth a quick glance of Service Note, it does not seem include the keyboard PCB (or PCBs) so I am guessing if there are multiple PCBs on the keyboard they are either soldered jumper wires or connectors and this is most likely where your problem is.
So counting each key from the bottom note of the keyboard 1 2 3 4 5 6 7 8 then repeat 1 2 3 4 etc determine what D#3 B3 etc are.
You should be able to work out what the problem is.
If this doesn't fix it it will be a damaged track on the D#3 board. Use the multi-meter's continuity 'beep' to find the break.
If it is in a connector you can always just use some wire to make the join with out having to get a new PCB jumper.
All the best
Royce
WIth a quick glance of Service Note, it does not seem include the keyboard PCB (or PCBs) so I am guessing if there are multiple PCBs on the keyboard they are either soldered jumper wires or connectors and this is most likely where your problem is.
So counting each key from the bottom note of the keyboard 1 2 3 4 5 6 7 8 then repeat 1 2 3 4 etc determine what D#3 B3 etc are.
You should be able to work out what the problem is.
If this doesn't fix it it will be a damaged track on the D#3 board. Use the multi-meter's continuity 'beep' to find the break.
If it is in a connector you can always just use some wire to make the join with out having to get a new PCB jumper.
All the best
Royce