I must have been trying to hard and to late yesterday because today I started again going through the interrupt sequence and yes $4576 is there and happening frequently.I will check the timing of it and add it on after.
Interrupt starts at $2038 then $3d8C and finally $4576 A9 and ending at $45AA 60 RTS Strangely the following instruction is $45AB A5 which I don't understand unless it's reading the ROM before executing the code. Next is $3d95 45, where it was before the JSR followed by $3d96 68 up to $3d9B 40 RTI. Again the following instruction is $3d9C 48. Then the following:
01F9 48
01FA 48
01FB 48
2dCE AD
2dCF 88
2dd0 0A
0A88 0A
So it looks the timer interrupt in software is ok and is exactly 327.2uS, just as dclxvi said it should be. Again I rechecked pin1 IC1 and there is no change to the strobe output. So do you agree the CPU port is faulty?

Pin 1 isn't the output of Timer 1. Pin 1 is strobe output of timer S, which is a 128 ms pulse. The routine at $6CCF enables the strobe output (by setting bit 4 of memory location $F5) and sets the pulse width by storing $0F into memory location $FA.

$6CCF isn't called from the ISRs, it's called from the main loop. The two paths that lead to $6CCF both begin at the JSR $6618 at $5943 in the main loop, but I haven't analyzed the routines that lead there, so I don't know under what conditions it is supposed to reach $6CCF. If you don't see address $6CCF in the logic analyzer trace, I wouldn't expect to see anything on pin 1.

Sorry about my misunderstanding. I will check for JSR $6618 at $5943 later tonight and report the results.

$5943 and $6618 are appearing in the LA trace as shown. Now looking for $6CCF

Today, I am happy to bring you some good news. I spent time today looking around IC4 MI-1 interface and noticed pin 34 was always lo. whereas pins 36,35,33,32,31 were all hi. When I disconnected the pedal keyboard at CN3 the upper and lower keyboards have started to work. I think that goes some way to testing the operation of IC4. Later tonight I will remove the PK20L Pedal pcb and see what's going on. I still don't have any lights or any of the buttons going, so I still have problems to solve. Still feeling good about today's progress.

Today I have been checking signals from IC4 MI-1 interface to the DRV pcb. I have as follows
/TXREQ LO no activity
RRDY LO no activity
SD LO no activity
SCLK HI no activity
/MIC HI no activity
Does anyone have ideas about how DRV board might start up? Does it start with something sent to DRV and then waiting for a reply or does MI-1 wait for a ready signal from the DRV pcb. Any suggestions or checks much appreciated.

I also took out the Pedal keyboard PK20L today and it appears that IC1 M50926 has failed. It is not scanning keys. The internal clock is running 2MHz. I have left it all plugged in but disconnected /EVRP to keep the other keyboards running.

Am I correct in thinking that IC2 deals with keystrokes and how they will sound and IC1,among other jobs, deals with enabling all the rhythm sounds? I appreciate there is much more but I still have not grasped how everything works in relation to each other with this.

DRV board:
To me, it looks like RRDY at CN1 pin 2 could be a high_active "ready" signal which tells the MI-1 that data is ready to be read.
RRDY only can go high, if pin 31 (/RRDY) at _all_ of the microcontrollers IC1 AND IC2 AND IC3 is low.
Basically the transistors Tr1, Tr2, Tr3 (fed by IC1, IC2, IC3 /RRDY) form an open collector NOR gate switching RRDY to low.
//RRDY has a 10k pullup resistor (RN4) to +5V on the DM board.
Would suggest to check with an ohmmeter if there is a short circuit between RRDY and GND while the HS-5 is powered off, just to see if Tr1..Tr3 might be dead.

PK20L pedal keyboard: a (rotated) picture from the organ forum about the SPI like communication between the MI-1 and the keyboards:

Thanks for your ideas. I had already checked TR1,TR2,TR2 on the DRV pcb. I did in fact pull all three of them from the board and some of the desk lights came on but not as they should have. Now I have the pedal board disconnected and the U and L keyboards I will try again looking at the RRDY line.

It's unlikely that IC50 on the DM board (TC40H032) (74HCT32 ?) in the SCLK line to the DRV board is defective,
nevertheless check if the signals on IC50 pin 5 and pin 6 look identical.


Hmm... this implicates that Tr1, Tr2, Tr3 on the DRV board are not defective, and that RRDY probably also means "data ready for write".

//The difference between "I" and "1" doesn't come out clear in the schematics.

The IC1..IC3 microcontrollers on the DRV board.
IC1 controls the LEDs on the PNI board through 2L1..8 and 2D1..9 for: appregio chord, keyboard percussion, rythm, balance, ensemble, manual balance, tremolo/symphonic, flanger/delay.
IC2 controls the LEDs on the PNI board through 1L1..8 and 1D1..9 for: reverb, sustain, orchestral voices, bass voices, percussive voices, lead voices.
IC3 controls the LEDs on the CP, MM, PN2 boards through D1..10, G0..8 for: c.s.p, f.m.p., tempo, tempo/data, bar/beat, registration memory.

Are all of the LEDs related to one of the chips IC1, IC2, IC3 always dark (or always lit) ?
Is the /RRDY output (pin 31) of one of the chips IC1, IC2, IC3 always high (while there is activity on /RRDY of the other chips) ?

I will check for those signals this morning. All leds are off for IC1, IC3 and IC3 and the switches are being scanned but no data is transfered.

Today I disconnected the RRDY signal at CN1-2 and checked that RRDY was hi at IC4 and it was Hi. Reconnected and removed TR1 and test....no change. Refit and remove TR2 and magic IC1 and IC3 are no running using RRDY and switches and leds are on.

When I scope on IC2 pin 31 I can see the pin is active and some leds are lit even though there is no communication. REVERB level 5, Arpegio 2, ORCHESTRAL VOICES COMBI 2, PIANO. Tomorrow I will check for driver transistor and switch failures and the scan lines from IC2.

Thanks for sticking with me and these problems. I can't understand at present the failures to the pedal CPU and DRV occurring at the same time. It would be unusual for more than one fault at a time.

Any suggestions where to get a pedal cpu, M50926 XC299001?

You are welcome, and I don't give up that fast too.

Try searching at Digchip, but I don't have any experience with this.
M50926-402SP or M50926-505SP ?

Had no luck with trying to find a (Mitsubishi ?) M50926 datasheet.
So I can't tell whether a M50926 has to be programmed by the end user or already contains the software for scanning the pedals in mask ROM by the manufacturer.


One possible explanation would be a long term failure due to an ESD event.
Like when somebody already had tried to repair that HS-5 before you while not taking enough precautions against electrostatic damage.
Does the HS-5 show any signs of previous repair attempts ? //"unskilled/unqualified third party repair" that is.

If a thunderbolt during a storm would have hit the ground 100 meters away from the HS-5, shifting the ground potential of AC mains
while the HS-5 had been up and running (the previous owner of the HS-5 probably wouldn't have mentioned it),
this also would be a possible explanation, but I think in this case some more of the semiconductors in the HS-5 would be defective.

Hmm... back in the 90s when I did TV repairs for a living, we had that effect that electrostatic charge from the CRT
had piled up in the loudspeaker grill and eventually found its way into the microcontroller of the nearby control unit,
(the solution had been grounding the grill with a resistor), but the HS-5 has no CRT.

Hi everyone. I have finally found the second fault on my HS-5 organ. IC2 M50726 (XB827003) is faulty. I checked all switches and leds attached IC2 and no fault so I then swapped over IC2 and IC3. I refitted the RRDY transistor for IC2 and removed the same for IC3 Everything on IC2 now works. I need to try and find some parts now, a pedal keyboard CPU M50926 XC9901 and the DRV CPU above. I'll be keeping my fingers crossed.

Thank you to you all for your especially for the .lst.

I don't know how I can repay your efforts but I do have a lot of UV erasable eproms if anybody should need some. Not many small ones, mostly 128K up to 1M. They are free for the cost of postage.

Digging into an old/complex design built from obscure chips and trying to figure out how it works was fun.

If replacing the two microcontrollers (M50726, M50926) brings the HS-5 back to life again, please let us know.

Good luck.
Cheers,
Dieter.

Hi there team, I have now located spare parts from a company in the UK. They have offered the complete boards . I am hoping they arrive before Christmas but I have to say I am very confident now about the repair being completed and it should leave some spares for the future if needed. I will keep you posted with a sound clip when they are installed. Still very excited by this whole thing and having made contact with you guys.

Let me begin by wishing you all the very best for the New Year. Yesterday my spare parts arrived from an excellent company called Premier Organs in the UK. The owner Lynne has been super helpful. Today I installed the DRV pcb and suddenly I have all the lights and buttons working. Next came the PK20L pcb and now I have the pedals back as well. Special thanks go to Dr Jefyll and ttlworks for hardware ideas and dclxvi for software support. Thanks to you all.