Hello all,
Notice: Lengthy note follows for the tech nerds among us, but important to convey technical observations in order to resolve an inoperative Combo Organ.
I’m working on repairing a Korg CX-3 Digital Combo Organ (and thusly, we commence to enter into the depths of nerdiness. 
) It’s owned by a friend.
I’m looking to contact anyone who is/was either a service tech or has been very frequently involved with electronic troubleshooting and repair of the digital CX-3 instruments. I’m an EE who has been digging into troubleshooting this unit.
I specialize in repairing and restoring nearly any electronics dating from before WWI (yes, from before 1913 or so just after radio first came on the scene) up to around 1970s and 80s. I’m nearly 67 now, so have seen the dawn of microprocessors and personal computing. And I know of the wonderful mechanical tone generators Hammond had originally designed back in the late 1930s (B3, BC, AB, etc.) and have all Hammond Service Manual documents up till about 1975 in my archives. So this is in my wheelhouse for troubleshooting and repairing, if possible. Enough of the ‘where I come from’ stuff.
I’ve got the CX-3 Schematic Manual, Rev 1 from Nov 2000. It’s a “Provisional Version”, so there may be accuracy risks. But the hardware seems to match very closely. The unit’s owner found a pdf of that document from a fellow in Europe (in PT, I think). It has the digital CX-3 schematic plus very minimal troubleshooting info (strong emphasis on ‘minimal’! 
).
I also have the Owners Manual the unit’s owner had downloaded from Manualslib.com. Those of you who’ve been deep into these units or were service personnel may have come across these same documents by now. I’ll reference identifiers in the Service document in my observations/questions which follow.
The hardware makes a lot of sense from a designer’s point of view as I had done digital logic design back in the 80s and 90s. Rather classic - there’s no PALs, ASICs or FPGAs, etc. used, so one can pretty much follow the hardware logic right through the whole system. Nice.
But the Majic of this device is found much, much more so in the software (‘firmware’) in the Flash ROMs and, I suspect, the two Effects DSP chips (the two TMS57070s, IC19 & IC20 on Page 6).
And not knowing the firmware programming & processes is where troubleshooting gets really challenging without a technically comprehensive service document.
Using the Self Test Mode found in that Schematic manual (simultaneously depressing both “Program 4” and “Program 6” keys then turn power ON), I do get the following two end result states. Either one or the other randomly occur, which is concerning:
1.) Mostly I get the error display “MIDI Loop Error” and the test stops. I do not have any experience working with MIDI equipment, so what could the Loop Error mean in this context? There is no specification in the service documents on anything about using a loopback connector of which I’m familiar with in other types of communications interface testing. (Keep in mind I had reported above there is very minimal troubleshooting detail given.)
2.) However, repeating the self test, I will sometimes get “DSP2 Test Failed”. The schematic does not point out which of the four DSP chips on the Main board is DSP2.
I think best tactic going forward is to learn from whomever is very familiar with troubleshooting these electronics just what to expect as a baseline upon initial power “ON” of the unit. Here’s what I find currently after Power On…
Power supply voltages (all 4) are correct. Digital display shows the default Program #1 info: “A1: Organized” which appears at power up. The setup parameters can be changed and stored in flash User ROM, IC22 on Page 3. LEDs which should likely be relevant to initial power up are also illuminated. The Main MPU “X1” clock crystal is right on frequency.
The Rotary function (simulated “Leslie” effect) does appear to work as the digital signal viewed at the DAC (Digital to Analog Converter) input on Page 7 rhythmically changes until I shut it off. There is evidence of correct keyboard scanning through IC23 (a uPD71055 PIO chip) found on Page 3. This is because the Keyboard Self Test function will work once I’m able to move into the manual test functions during Self Test. It calls out keys on the display beginning with note A7 at farthest right and continuing leftward verifying keyboard stroke action.
All of that indicates the Main MPU, I/O interface circuits and Main Program ROMs all found on Page 3 of the Schematic appear to be okay.
However, moving onward through the circuit to Page 5, there is no serial data exiting the Main MPU which should be expected to be sent to the second H8S2352 microprocessor of which is the SUB-MPU, IC15. This MPU chip controls the the Tone Generator DSP chips (the two Motorola DSP56362s, IC7 & IC8).
Looking further into that SUB-MPU circuit, I see no activity on the data and addressing lines to and from the Flash ROM, IC6. Only the “A1” and “RD/” inputs to the ROM chip show the SUB-MPU trying to read the firmware program. This may be the problem…the MPU appears to not get the first address in the Flash ROM accessed and its 16-bit data word read into the MPU in order to increment to the next address location and so forth. And the Main MPU may be waiting for the SUB-MPU to finish this boot process and report back to the Main, hence the lack of notes/tones generated.
With that said, I had earlier found a discussion elsewhere online where the CX-3 firmware can be updated through the MIDI ports. Since there is apparently no activity from that SUB-MPU, IC15, I suspect someone could have “bricked” the Mainboard before the current owner bought the thing. Warnings were given to be absolutely careful in doing this update process.
If so, does anybody believe it could be re-flashed? And exactly what MIDI hardware equipment and software would I use? I’ve never, ever had contact with MIDI stuff before.
The MIDI interface into the unit is only through the Main MPU, IC11. However the update program could very likely access the second SUB-MPU Flash ROM, IC6, via the Main MPU. Slow process as the interconnection is from MIDI port through the Main MPU and via the serial I/O (the TXD1, RXD1, and SCK1 pins on IC15) to the SUB-MPU, but it’s the only manner how the SUB-MPU firmware can be updated.
If anybody, including any non-technical folks have questions or wants clarification on the terminology, just let me know. I tried to keep that level of detail lowered.
Alternatively, instead of spending tons more of time on troubleshooting this thing, which the unit’s owner cannot really afford, does anybody know where I can obtain a functioning digital Main board for this unit??
Thanks for reading this far. And thanks in advance for any help or information you may provide!! If you feel somebody off this Forum may help, I presume it would be okay if you copy and paste this text into a message to them. Somebody somewhere has got to know a lot about these Digital CX-3s still out there. Thanks! -Chris Fandt, Western NY State, 29 May, 2020
May 29, 2020