Korg Forums

[luckyluca]

I have a Trinity Pro.
All of a sudden, today, I got a system error printout over the disk menu when loading a combi/prog .pcg file.
I rebooted the Trinity and tried loading a different .pcg, another system error printout.
After that, when turning it on, it stays on the blue screen, frozen, with no led lights, nothing works.

Using enter+0 +5 +9 on startup changes nothing. It doesn't enter diagnostic either.

Video here: https://youtu.be/omznZPu5A30

Please note it was emitting a faint smell of warm electronics/plastic when using it for a while, but nothing serious, and it worked flawlessly until now.

Help!
I can flip it upside down open it and look for obvious signs, also I'm not great at electronics, so I won't be soldering/desoldering.
Any tips/suggestions?

I'm based in London UK, if there's anybody local who would be happy to lend a hand please PM me I would be super-grateful!

thanks for looking

[voip]

The most likely cause of the problem is the power supply, so that might be the place to look first. The board should be labelled something like KLM-1875. There could be other causes, e.g. the floppy drive may have sent corrupted data to the main board.

Test mode is entered by holding down "5" and "ENTER" whilst powering on. But, before doing this, first try an Initialisation, by holding down "0" and "ENTER" whilst powering on.

It might be worth contacting a keyboard tech to have a look at the instrument.

Cleaning the floppy drive heads, using a floppy disk cleaner, is also worth carrying out.

.

[luckyluca]

[voip]

2.8V across a CR2032 is low, but probably not the likely cause of the issues. Having said that, it might be worth replacing it, to remove it from the list of possible causes, and because it's a relatively simple thing to do. I would replace a CR2032 showing 2.8V, as a matter of routine. The Trinity doesn't contain a realtime clock or calendar, which is often the reason for a CR2032 to be fitted. Instead, the CR2032 is used for maintaining RAM data, when the mains power to the Trinity is switched off. Shorting the connections that the battery makes contact with, during replacement of the backup battery, will clear the RAM which contains the PCG data, so it may resolve a potential corrupt data problem.

It's also worth taking a close look at the electrolytic capacitors on the power supply, to check for case bulging or leakage. Although "capacitor rot" is reckoned to affect electronic devices manufactured between 1999 and 2007, and the Trinity's manufacturing window is before this period, general ageing of capacitors also leads to reduced capacitance, so the power supply may be functioning out of spec. The smell of warm electronics from the power supply board could be indicative of this. Replacing the electrolytics on the KLM-1875 board is worth doing. Excessive ripple on the 5V supply can result in a continuous stream of reset signals being generated, resulting in the Trinity's digital electronics never getting the chance to boot up.

.

[luckyluca]

[voip]

Shorting the battery directly is not a good idea, although the internal resistance of a CR2032 typically is around 10 Ohms when new, so this would limit the short circuit current to around 200-250mA, but that could drop the battery voltage somewhat, after the short is removed. There is a better place to short to ground, and this makes use of a 1k resistor (R206) connecting the battery +ve to the rest of the circuitry. The junction between this resistor and R72, R77, and D1, can be shorted to ground without drama. There is a track on the topside of the board that is part of this net. With the board facing you, so the battery is towards the lower left corner, there is a track just to the right of the battery that starts next to a small triangle of three other vias, turns through about 220 degrees, then runs horizontally, between the track with DI-TRI screen printed on it, and ends with a slight upward stroke, and the copper area with C12, C13 ,C14... on it. Check with a multimeter that there is around 2.8V on this track and, if this is the case, short it to ground for a few seconds.

Fluctuations in the power supply rail voltages is likely to lead to problems in proper circuit operation, so needs to be fixed.

.

[samriccijr]

Hi (Un)luckyluca ....

Voltage drift suggests power supply issues....

Get the Trinity service manual
Replace the BR2032 battery
Clean and re-seat the power connectors
Unplug power to the USB floppy and restart. (this uses one of the 5V rails)
Check for cracked solder joints by the voltage regulators
Check the power supply photo transistor
Check the power supply transistors Tr1 and Tr2
Check the power supply voltage regulators
Check the capacitors for swelling, smoke, failure
Lastly, Check the display connections and fuse resistors

Unplug the USB floppy and restart. IIRC the floppy used 5V.
--- If the USB stick gets warm, you could be straining the 5V power rail...
---Or try a different USB stick....

Replace the Br2032 battery. 2.8V should trigger a Low Battery Voltage message. (I had a Kurzweil PC2 that would not boot until I replaced the 0.1v battery with a 3.0 volt battery).

Clean and re-seat the power and ribbon cables. Then re-test.

Check the Power supply voltages. At the power supply harness test points, you should see:

Pin2 ground
Pin 1 12V DC
Pin 3 -12V DC
Pin 4 ground
Pn 5 5v to 12v Vcc

If you don't have voltages here, then it is something in the power supply:

Check the 12V voltage regulators 78m12 79m12
(broken solder joints are common here)
Photo transistor PC1 (be careful!)
Transistors TR1 and TR2 (be careful!)
IC 3 (zener diode board) (be careful!)
Look for burned, smoked or swollen capacitors, diodes, zoners, etc.

Moving on to KLM 1825 1826, you should see across

C1 5V to 12V DC Vcc
C6 12V DC
C7 -12V DC
C9 12V DC or less
C11 -12 VDC or less
C15 5V DC

the culprit here would be the 5V voltage regulator, fuseable resistors and capacitors.

Power off. Check the continuity of the Fuseable Resistors

R15 (10 ohm or beep)
R17 (10 ohm or beep)

Inspect capacitors for smoke, bulging, smell, etc.

This should get you started. Make sure you have power before you tear apart the touch display.

Please reply or PM if you have questions. Best!

Retired after rebuilding
Four 01/Ws (failed floppy drives and leaky 16v capacitors)
two Karmas
one i3
two Kurzweil PC2s

Sam in NJ USA

[luckyluca]

voip, sam thank you so much, great stuff.
It’ll take me some time to digest, and it’s a bit daunting. But still, great stuff.

Voip: I followed your suggestion and shortened the battery at the point in the motherboard you mentioned.
Unfortunately it didn’t help.
The battery line read -2.93v. by the way.

Sam: I understand that the battery that prevented the kurzveil from starting up had 0.1v rather than 3v, but mine was reading 2.93v by checking the board and 2.8 last time by checking the battery directly.

I found a klm-1875b replacement on ebay, quite costly, but accepting returns, it might be a worth thing to buy and test.

Regarding the klm1825/26, are those tips for the analogue audio board, or did you mean the motherboard 1820?
My gut feeling is that the issue is with the motherboard.

Before failing, the screen printed out system error: c11 (can’t remember exactly) and hex something something followed by garbled characters printed over the disk menu page.

Do you know if the trinity would still power up without the 1825/26 and floppy connectors plugged in?
It doesn’t give signs of life, aside from the display being backlit blue.

Otherwise, all boards so far look to my uneducated eye brand new and clean.

Sorry about my ramblings.
Best!

[samriccijr]

Hi luca!

Trinity should still boot with 2.8 volts of backup battery.

Yes the Trinity should boot without a floppy or audio/midi board connected. There is no harm in unplugging and trying.

There are 4 power connectors. You can unplug 3 and check the voltages, then connect one and recheck, then connect the next and re-check.

On the 1820 mainboard, the voltages should be

C5 5V Vcc
C6 12V
C7 -12V

Then clean the cable connections and take a break.

The touch screen is over 20 years old too....But we are not there yet...

Sam

[luckyluca]

Hi Sam and Voip,

Ok, so here it is, thanks for looking into this with me:

Power on, from 1875b just the powercable plugged to 1820. All other cables are plugged into 1820 and 1825/26. Aside from solo-tri.
1820:
1-2: 12v
2-3: -11.9v
4-5: -5v
C6: -11.97
C7: 0.01
C5: 0.01

Power on, from 1875b all powercables plugged in. All other cables are plugged into 1820 and 1825/26. Aside from solo-tri.
1820:
1-2: 11.97v
2-3: -11.92v
4-5: -5.07v
C6: -11.92
C7: 0
C5: 0

1825/26:
C1: -5v
C6: -11.9v
C7: 0v
C9: -11.31v
C11: 0v

Battery: 2.93v (or 2.98v this time, I can't remember)

And here below how I took the measurements, please let me know if I did anything wrong.
https://ibb.co/XYVtSF7
https://ibb.co/4N1pdyn

[voip]

Looking at the measurement points shown in the linked images, the reference ground point circled on the KLM-1820 image is the digital ground, and is, strictly speaking, only valid for making the +5V measurement across C5. C5 is a through hole component, and the measurement point shown in the image is not associated with C5. It is one of the lines to the aftertouch circuit board. Also, on the KLM-1820 board, the correct measurement point on C7 would be the -ve terminal of the capacitor.

It's not really very easy working from schematics and board layouts, without the board in front of me.

.

[samriccijr]

Moving discussion offline.

We are checking the power rails to each board.

Sam

[samriccijr]

HI Luca and voip!

Power supply voltages to 1820 and 1825 appear ok.

We are looking at the 5V and -12V voltage regulators on the power supply board.

Maybe a solder touch up will fix the 12v fluctuation.

We are also considering the 5V supply to the the display.

Also: Hold Enter-Reset and try to reboot. If you get a prompt, then the OS is ok.

Sam

[luckyluca]

Resuming after a couple of months of hiatus.
First of all thank you Sam for all your help, apologies for not getting back to you any sooner, real life too precedence.

To recap, the Trinity is still not powering up. The screen receives power and can see it getting brighter, otherwise, there are no other signs of life.

To bring this post up to speed and have it all in one place, I measured voltages in several places, and cleaned the powerboard by removing the gak/flux at the bottom.
I can't visually see cracks or other obvious issues but I'm not so familiar with electronics.

All images can be found here:
https://imgur.com/a/0YSgXDi

I have all of the measurements taken by probing the pins inside the connectors. Results are in Volts of course:


See the powerboard before and after:

[voip]

The solder joints on the PSU all look OK. The readings given of the voltages on the power supply pins are all inverted. Pin 2 of the 5-pin connectors is the ground or 0V point, so the testmeter black lead should be placed on this pin for the +12V and -12V measurements, and likewise pin 4 is the 0V pin for the 5V supply (this assumes that the black lead is plugged into the multimeter COM connection).

Just to clarify. In the last two groups of measurements in the table, are the 4-pin and 3-pin connectors, respectively, the only connectors plugged into the PSU?

There are signs of corrosion damage at the right hand side of the keyboard, to the case and to the PSU "tray". Are there any signs on the bottom cover that might indicate where this corrosion may also have gone?

Also signs of corrosion of the metalwork nearest to the backup battery on the 1820 main board.

Also on the 1820 board, in the diagonally opposite corner of the PCB from the backup battery, near CN34 and CN33A, it could be a trick of the light, but the PCB tracks look a little "crusty".

Going any further with the investigation is really going to need a scope to determine the power supply voltage ripple, and to observe signals on the digital boards.

.