Keyboards are not recognized on Yamaha Electone EL-90

[vgb14]

So, I recently fixed a problem with my EL-90 hanging that I posted in a recent thread.

Unfortunately, a few weeks later, the hanging problem returned.

I went ahead and changed out all of the remaining elec. capacitors on the PU board.

I also studied the DM board and changed out any capacitor that might be related to 5V logic. If anyone is thinking of working on the DM board, I will caution you that it is much more difficult than the PU board, which is single-sided. The DM board is double-sided plus it has ground planes which make it extremely difficult to desolder capacitors from.

When I get a chance, I will provide a capacitor reference document for the DM board much like the one that pieanoman created for the PU board.

The problem with the EL-90 now is that none of the 3 keyboards function. It won't even accept the 3-key input on startup to get into the diagnostic program.

Interestingly, the Rhythm function works fine with full sound from the speakers. I also loaded some MIDI files that were created with the organ, and they play perfectly. So most of the organ works. It just doesn't see any of the keyboards.

I'm wondering if anyone has seen this issue. From the spec, I see that the upper keyboard, lower keyboard, and pedal keyboard feed the DM board via connectors CN02, CN05, and CN03, respectively. From there, they share the same logic, which involves IC chip IC72, IC73, IC09, and IC1. I'm going to attempt some testing of these components using a scope to see if I can isolate the issue further.

Also, does anyone know if it's possible to get a replacement DM board for this machine?

[vgb14]

So, it appears that Yamaha is handling the keyboard using SPI logic.

Since I also have an EL-30 which uses the same Toshiba CMOS Gate array chip (TC17G005AN-0023, XB809A0) and keyboard chips (KBS1), I scoped out the keyboard connector pins on the EL-30 and found the following

pin 1 +5V supply
pin 2 /DIC +5V
pin 3 /DRU Initially +5V, press key and hold -> 0V -> +5V; release key -> 0V -> +5V
pin 4 /KDU Initially +5, press key and hold -> Note specific waveform -> +5V; release key -> Note specific waveform -> +5V
pin 5 CLK continuous
pin 6 and 7 are Ground
pin 8 /EVRU Same behavior as pin 3

Compare this to what is happening with the EL-90
Fresh boot, pin 3 /DRU Initially +5V, press key and hold -> 0V and stays there. It should return to 5V, but it stays at 0 and keys have no effect from this point on.

The clock on pin 5 is always running regardless. Machine continues running and I can start and stop rhythms, adjust volume, etc., but keys do nothing.

Fresh reboot, pin 4 /KDU Initially +5V, press a key and hold -> Note specific waveform -> +5V then that's it. No effect from keys from this point on again. Aside, from keys, machine keeps functioning.

Fresh reboot, pin 8 /EVRU, same reaction as pin 3.

I scoped both the upper and lower keyboard connectors and received the same results.

I have verified that 5V to the board is consistent and doesn't drop. I have also tried pressing on the board, pushing the plug-in style ROM chips, and wiggling the wires with no effect. Continuity in the wires from the MK boards to the DM board is good.

There don't appear to be any bad solder joints, but I am going to put it under a microscope to get a closer look.

[vgb14]

El-90 Organ is fixed

Through a lot of hard work and persistence, the organ is functioning 100% again.
I am posting this for the benefit of any Electone owners who may experience similar issues.

Following from my last post, I traced the serial data between the keyboard and gate array chip using an oscilloscope. The logic for the keyboard data is based on the SPI model (Serial Peripheral Interface).



Basically, the Gate array chip (TC17G005AN-0023, XB809A00) provides the Serial Clock and makes a data request from each of the keyboards. When a keyboard sees its data request input signal go low, it gets the okay to send note data via the KD pin to the Gate array chip via the IC73 (HC08) chip.

It was apparent that the Note Data in the /KD pin coming from each of the keyboards was successfully getting beyond the HC08 (IC73) chip. HC08 is an IC with some "AND" gates in it.

Starting from the lower keyboard and using the schematic as a guide, I followed the note waveform from CN05 pin 4 to IC73 pin 2, to IC73 pin 3, to IC73 pin 4, to IC73 pin 6. From here, it should go to /KD on the Gate array chip (TC17G005AN-0023) pin 37.

However, the waveform wasn't getting to pin 37. An ohm reading on the trace from IC73 pin 6 to the Gate array chip pin 37 showed some significant resistance. The fix was to run a bypass wire from IC73 pin 6 to the Gate array pin 37.

So what caused the problem? In my prior thread, I mentioned that I fixed the issue by replacing leaking capacitor C161 on the DM board. Although the leak didn't appear to create any obvious damage, under a microscope it was apparent that it had ever so slightly ate into the trace carrying the KD data between IC73 and the Gate array. Unfortunately several traces run directly beneath C161.

It's probably a good idea to replace at least C161 and C145 on this board if they are original as they appear to take the most abuse, at least in my case. Ideally, you'd want to replace all of the caps on the DM board, however, this board is a double-sided board with Ground planes which makes it extremely difficult to replace components on.

The PU board, on the other hand, is single-sided and far easier to replace caps on.