Repairing a Yamaha TG77 synth: where do I begin?

[armandhammer]

Hey gang, long time lurker coming out of the shadows

I have a Yamaha TG77 rackmount synth that I want to repair. But I'm really unskilled when it comes to troubleshooting digital stuff that has CPUs, RAM, ROMs, etc. The symptoms are basically that there's no response -- no text on LCD, none of the button LEDs light up when pushed, only the LCD backlight comes on because it's directly connected to the power supply. It looks to me like it's just not booting up and I have a gut feeling that if I can get it to boot then it's likely that everything else (the audio processing side) will work fine. It's a super clean unit with no visible damage, no indication that anyone opened it up before me, and not a speck of dust inside.

I'm looking for some guidance to help me learn more about troubleshooting complex digital systems like this, I am not looking for someone to "do the work for me". I just really don't know how to narrow it down to figure out exactly what's not working.

The Service Manual PDF (17MB, includes the circuit diagrams) and a Circuit Diagram PDF (3MB) are available here on my Google Drive:

https://drive.google.com/file/d/16OhDoH8hmu581E5xF7D5CqaLF0IAEzyu/view?usp=sharing

https://drive.google.com/file/d/16wUeHEjnGlGrHILRjDuttnZNjOAdEh73/view?usp=sharing

Here's what I've done so far:

1) Replaced all of the electrolytic caps in the power supply and the main boards, checked all voltages coming out of the PS. Checked VCC voltages at most of the ICs. Checked two crystal oscillators on the board. Everything looks good.

2) Replaced the internal CR2032 battery.

3) Replaced the LCD (I knew this likely wasn't the issue, it's just known that these particular LCDs fail over time and the text becomes hard to read so I wanted a new one in there so when it finally boots I know I'll be able to read the display).

The next step is to test all of the ceramic capacitors but I'm waiting for my Anatek Blue meter to arrive in the mail. I might just go ahead and replace them all because it's cheap and there are only about 60.

Anecdotally, someone who has a SY77 (the keyboard version of this synth with 2 main digital PCBs instead of just one) said theirs exhibited the same symptom and it was determined that a 74HC245AP IC had blown. They got it replaced and everything worked fine. I have no idea how one would determine if one of these ICs is bad, though...

Here are the questions that I have:

How do I narrow down what might be wrong? I've spent hours looking at the schematic and looking at the data sheets for the "decoder" and "transceiver" ICs and have a vague idea of how they're supposed to work, but no idea how to actually test them.

What's not clear to me, using IC105 (SN74ALS245AN) as an example, is what the voltages at pins 2-9 are supposed to be. They're connected to RA106 which says +5V, but I measure about 0.2V at all those pins. Other of these same ICs, like 103, I measure 2.5V on 2-9. And yet on 106 I get 90mV on pins 5 and 6, 2.35V on pin 7, and 5V on 2, 3, 4, 8, and 9. But to my untrained eyes, the circuits (resistor arrays and capacitors) around these ICs all look nearly the same. Is it significant that I get different voltages on these i/o pins?

Should I start my troubleshooting somewhere else?

Also, is there any way to check the CPU, RAM, or ROM to find out if that's where the trouble is? I think ROMs can degrade over time, maybe?

I have the following equipment:

Rigol DS1054 scope, ZT302 DMM, FY6800 Signal Generator / Counter, Anatek Blue capacitor meter (arriving in the mail in a few days), one of those cheap transistor/diode/capacitor testers from Aliexpress, Hakko 888 soldering station and FR-301 solder pump.

I'm willing to invest $150 or so into other equipment which can help me troubleshoot this digital stuff.

Thanks for the help. I wish I could explain all of this better, but with only 1 year of experience and this being my first attempt at repairing something that's mostly digital, I still lack the proper terminology and skills.

If this was your synth, where would you start, what would you do?

[helius]

Yes, you can easily test components like the CPU and RAMs to see if they are working. Probe any of the data or address lines (make sure you don't short them out to neighboring pins) and check to see if they are behaving normally (switching states from low to high). In particular, if any of the bus lines adopts an intermediate state (like 2V) you know something is wrong. Also check the reset lines: they should go active when power is applied and then go inactive about a second later, without constantly resetting. These measurements will tell you a lot about how much of the device is working.

Other diagnostics include using thermal imaging to look for failing components, and using stimulation (logic pulsers) to check if signals are propagating.

Bus transceivers are a reasonable suspicious part. They tend to take the brunt of any ESD events on exposed circuits like the expansion card slots. Another thing to watch for is defects in the PCB itself: either from leaky capacitors or just accidents, traces and vias can be broken or shorts created. Sometimes these are hidden underneath components which makes troubleshooting difficult. If you find an area of interest where components aren't behaving, check every trace nearby for continuity.

[ozcar]

IC105 is a transceiver for the 8-bit data bus. RA106 is just a bunch of resistors that would pull the "A" side lines up to 5V when/if not actively driven (RA114 does the same on the CPU attached "B" side).

I guess you were measuring the voltages with your DMM? Otherwise, as Helius said, a voltage like 2.?V would be a bad sign. 

Pin 19 (/OE) of IC105 is grounded, so the outputs are permanently enabled (only the direction changes, depending on R/W) - if you were to see all the "A" side lines all low, but not all the "B" side lines low, then you could be on to something, but I doubt it will be as easy as that!

[armandhammer]

Thanks for the responses so far. I need to do some googling / youtube video watching to make sure I know how to probe the data lines properly with my scope -- at this point, I'm still apprehensive about frying chips if I somehow touch them with the probes in the wrong way. Fortunately on this board it seems like every IC has test points coming out from each pin so at least I don't have to worry too much about touching the IC itself and having a probe slip.

Lots to learn. Give me a few days and I'll report back what I find. In the meantime of course feel free to reply with any further help. Thanks again!

[Renate]

If you checked all the power supply voltages, the next thing is just to get a sense of what is happening.
That means looking at the address/data/control lines.
Are they dead? Are they live for a second when you turn it on, then go dead?
Some sort of bus error might stop it dead.

If it churns for a second then dies you may need an easy way to reset it so you can see something.
IC169,  a PST518B is the reset detector.
Pin 3 is the open collector output.
You can safely ground it to reset. Just be careful.
(Have you already checked that this pin is high? It should be?)

For advanced testing you may even need to connect an open collector pulse to that to repeatedly reset it while you look with the scope for life on the board.