Recently bought a dirty/halfbroken/old Williams System11c Rollergames. (Pictures at the end of the post, saved on imgur.) Besides a broken display (I ordered a replacement, http://imgur.com/a/pl2EL ) the machine sometimes powers up into attract mode, but sometimes it gives 4 (or 5, counting is hard) 'error sounds' (not the gong). If it does power up with the one-gong I can play the game just fine but the sound (music and sfx) goes totally silent after 2 or 3 minutes. (Record time is ~5 minutes.)

Three questions:

1. I think it's one of the EPROMs that are bad (according to http://techniek.flipperwinkel.nl/wms11/index2.htm#sndled ). But I'd like to know if I can test(measure something) this further before ordering replacement EPROMs?
2. In regards to question1, I'm interested in learning to burn my own EPROMs. Any suggestion what brand and type I could buy that's entry level, for pinballs mainly and runs on Windows 10?
3. Pictures (http://imgur.com/a/pK7wq) of the CPU board that had a leaking battery, damage is "ok". Should I do something specific to clean it up?

I'm new to all this, and I bought the Rollergames to learn something about repairing pinball machines. Hope you can help me out with some nudges/tips/answers to my 3 questions.

Thanks for the fast reply. Looks like I need a burner to check before ordering. Doesn't look like I can source a GQ-4X locally though . I'll have a good look and clean for the CPU board.

I guess the EPROMs are broken with the problems I'm having and I just should order replacements?

Small update: the game always boots into attract mode now I fixed a proper batterypack. So that's good news. The sound still stops working after 2-3 minutes though. Not sure why.

With my limited knowledge and a logic probe I checked some ICs on the sound board. I started with the resetline of U13, U8; then went to check data on U8, U3 and U1. On the latter IC i only measured the data (which seems to be present). I did not measure the output side of U1. Not sure how to do that.

For a more complete description of what I noticed (including a peculiar pattern on U3), see the images and descriptions on this imgur link: http://imgur.com/a/hl8dZ

Any tips or idea's what to do next is appreciated .

Otaku and @yeoldpinplayer: awesome suggestions. I combined your two suggestion and started a game (cold) with sound. The moment the sound stopped I immediately fiddled with the J21 connector on the CPU board (ribbon cable to the soundboard). The effect being I had the sound return for about 3 seconds after it was gone. Fiddling after those 3 seconds didn't matter anymore and I was unable to make the sound return.

So I guess I should reflow the pins on both boards and perhaps try to find another ribbon cable? (I might have a spare PC IDE cable I could test with.)

Yeah, I hope so. I'm just trying things I think might help you guys give me good advice on the matter . I'm all very new to all this. Curious though, as I saw the data pins on the ICs pulse. I guess that's not a good way to measure if it's generating sounds. (Again, very new to this.)

Not super much. I did solder the remote batteryholder to the CPU board and that looked ok. I have a 25w soldering iron, will that suffice for reflowing? Any tips for the reflowing process in general?

My display is out and I'm awaiting arrival of the new display, so getting to the sound test is a test on itself . (And I'm likely to slow for that in regards to the problem "heating up". Good tip to fiddle with the connector, looks like I should reflow both connectors and check the cable and hope that solves it.

Thanks for the repair/pinballgames offer! Afraid mailing it would likely cost me more than a new board + would not match with my instant gratification person I am . (I'm located in the NL, takes about 20 days+ for a parcel to come my way from the states.)

The ribbon cable isn't PC size, but since the boards do not have a bounding plastic around the pins I can abuse a slightly larger ribbon cable to test. I'll test that first (good tip). Haven't checked it any further though(busy days), but I'm going to go at it this weekend. I'll post an update on how it goes / and if it fixed it around Sunday.

I reflowed one side (soundboardside) to no avail. The comment about compressed air made me feel the heatsinks, they were cold to the touch. I did notice that U3 is somewhat "warms up".

U3 isn't hot, but it's noticeable warm (while the other ICs are just cold). You can keep your finger on it no problem (so it isn't hot).
Is that normal for that IC?

@barakandl: The cpu feels cold to the touch.

I reflowed both sides of the connector now (on the soundboard + cpu board). I also changed the cable (ended up using the display cable ). I replaced 4 caps near the end of the soundcircuit, suspecting they might be broken. (See pictures, the brown one was a little curved at the bottom: http://imgur.com/a/zNq1m )

Alas to no avail.
The music still stops as evident in my phone recorded movie:

In the schematics I cannot really find what voltages should arrive at the soundboard. I wanted to measure that, but i don't really think that's the issue as I guess more stuff would stop working.

Suggestions / things I could try or test for?

Good call. I cannot read the voltages either (regrettably the pinball came without its manual).

I'm guessing it reads: +5, grn, -12, key, +12, +12.
I've measured: +4.94, x, -11.8, x, +13.91.* (Disconnected J3 to measure.)

*) Measured it 5 minutes after shutting down the machine (leaving J3 disconnected), and powering it up again at 14.45 .... is that within tolerances?

No clue if it's within tolerances (see above post). But it seems that I tend to get 14.00 volt at most lately, not anything higher. (Currently it's 13.4 volt.)

Anyway, with my epromburner still in order (I'd love to check the eprom data) I started experimenting again. Reminder: I'm new to this. So any tips / tricks are welcome! I ran a scope on U3 (the one that gets warm/hot, the YM2151) on the serial output pin (the one that goes to the DAC) and a "random" (d0 tru d7) data line. (Used by the chip for addresses and data, both read and write.) Results in the two short (few seconds each) youtubes below.

Youtube1: Turn up the sound if you want to actually hear the music/sound stop, but it's apparent from the change in output. You're looking at the serial output that goes into some form of "repeat/steady" mode.

Youtube2: This is pin D5 on U3, a random selected dataline. It kind of looks like there is some data being send/read/addressed? Any second opinion on this? (U3 tends to be warm, approaching "nearly something near hot" by this time. Perhaps important.)

I put the scope into triggered mode. This is the first time I connected my (new, old) scope to an actual device. New to this scope stuff as well.

PS Loving this adventure this Rollergames is setting out for this. Learned a lot so far on pinball machines!