EDIT: It turns out the missing sounds are speech ("Yo, Taxi!") and only some of the drum sounds in the music. Most likely the bass and snare drum samples, such as the beginning of multiball.

I just repaired my Taxi, which had a bad bridge rectifier that popped capacitor C9 on the power board.

https://pinside.com/pinball/forum/topic/taxi-boot-problems

The game now boots and plays, but some of the sound channels are missing. There's no buzzing to go with the warning when the machine is turned on without batteries, and none of the music tracks have a drum channel.

I don't know much about the sound system in System 11B machines, but I wouldn't be surprised if something got fried when that rectifier started leaking. Where do I start testing to see if I can figure out what's going on with the sound?

I've re-seated the MPU-to-sound ribbon cable twice. It's not off by a pin.

It turns out the missing sounds are speech ("Yo, Taxi!") and only some of the drum sounds in the music. Most likely the bass and snare drum samples, such as the beginning of multiball.

This makes things different.

Well, now I've done it. I made the mistake of using the word "leaking". Now everyone's going to want to see pictures of the MPU.

I meant it was leaking current through one of the diodes. Also, the MPU board doesn't have a bridge rectifier on it. Trust me, the MPU board is fine.

What should I check for on each of the pins of the ribbon cable? High? Low? Pulsing?

I've done some more poking around, and have found that the -12v on the power supply actually reads in the high -14v range. It looks like the MC1408 Digital Analog Converter (U2) runs on -12v. When I run the sound test (not the music test) and I check the sound inputs on the U42 PIA (which are on Pins 2 through 9) I don't see that they're pulsing. Rather, they're stuck high. The pins that are high change with each sound in the sound test.

Sound 1: 2, 3, 4, and 9 are high
Sound 2: 4,9 are high
Sound 3: 4,6,9 are high

... and so on.

I don't trust my logic probe to tell me if the Sound Input Select pins on U2 are behaving as they should. I also don't know anything about how the sound/speech stuff works. When the system decides to play "Yo, Taxi!", how does that whole process actually work? From there, I can start testing the different points of failure.

I'm kind of stuck without help from you guys, so anyone who knows anything would sure be doing me a favor by helping out.

My bad. It was pins 2-9 on U9 that I tested. I'll test the pins on the chips you suggested as well.

I tested the the pins you suggested while in Sound Test mode, letting the system cycle through the eight or so sounds ("Yo Taxi", "Ooh", etc).

Here are the results:

U-41

Pin 18: Stays high the whole time
Pin 40: High the whole time

U-42:

10-17: Low all the time
18, 19: High all the time

39, 40: High all the time

What am I looking for when I test these pins? What state should these pins be in when the sound test is cycling through its sounds?

It looks like U3 is the CVSD, which according to The Internet, produces the games voices. From the schematic, pin 3 is "A Out". Analogue out, maybe? That goes to the sound/speech mixer.

I don't see any activity from pin 3 during the sound test at all. Maybe that's where the speech and sound are getting lost.

I kind of feel like I'm flailing, here.

If I'm reading this right, it looks like some of the "sample-y" sounding stuff is generated on the MPU, and some on the sound board itself. The 4-pin molex that goes from J16 on the MPU to J1 on the board has 5v running through it. No problem there.

I'd hypothesize that the sound from the MPU isn't properly being mixed on the sound board before going out to the speaker, but I don't know which sounds are generated by the "sound ROM" on the MPU and which sounds are being generated from the "music/speech" ROM on the sound board.

I've also tested the ribbon cable by swapping it with the Jackpot score display ribbon cable. No change.

On a related note, which pin is pin 1 on the Jackpot score display?

I have to say, this thing really has me stumped. Is there anyone here who can help me?

Guess not.

Despite the fact that nobody here is willing or able to help me, I'm going to document my progress anyway, in the hopes that my work can help someone else in the future.

DumbAss loaned me his working sound board and power supply from a Black Knight 2000. When I put my ROMs into his sound board and I use his power supply, I get drums underneath the music both in-game and during the music test.

With that in mind, I've looked at the schematics and found that the drum samples are likely sent through the MC1408 Digital-to-Analog Converter at U11 on the sound board. U11 has (what I think are) inputs on pins 5 through 12. During the music test, I can see pin 8 pulsing on and off on my logic probe in time with the drums. Specifically, music test 02 pulses during the drum intro of Multiball, and then goes dead when the drums stop.

I've also been able to reproduce the symptoms by grounding out pin 14 of U11, which I believe takes the DAC out of the mix entirely.

From here, I need to do a bit more testing to really understand how the sample part of the sound board works. Then, I'll hopefully be able to isolate the failed component and replace it on my board. After that, I need to do the same thing with the speech, which comes from the MPU board.

More test results. The part I'm working around is here:

SoundCircuit (resized).jpg

On the working sound board, the following voltages are read at each pin that doesn't deal with inputs, which are pins 5 - 12. Those are consistent across both boards, and are the pins that are in time with the drums. This is probably because the PIA that drives UI11 is working fine.

1: -5v*
2: 0
3: -14v
4: -.6v
13: 5v
14 & 15: 0
P16: -12v

The difference between the working board and the non-working board is Pin 16. On the non-working board, pin 16 gives me -0.6 volts. That's weird. So, I checked the schematic, and found that pin 16 goes to C6. C6 goes into into R10 on its other side. On both boards, the R10 side measures -14v.

I also took measurements at different points for all the other various components in that area: C8, R9, R10, C5. Nothing jumped out at me, unless I missed something on the non-working board.

Because C6 reads differently on the U11 Pin 16 side on the two boards, my current working theory is that I have, indeed, fried my 1408 DAC. I'm going to see if there's a way to use a clever configuration of alligator clips and a known good 1408 to see if I can test the working one in-circuit with my board somehow.

*Also: Pin 1 reads -5v on the working sound board. That's really odd, because Pin 1 isn't actually connected to anything. It reads almost nothing on the non-working board. That certainly doesn't make any sense, but at worst it's just another sign that my U11 1408 chip is busted.

I've ordered a replacement 1408 chip, so let's hope that's the problem. I was going to get one from Ed at Great Plains, but he's not taking orders anymore, so I grabbed his part number and found one online. Fingers crossed!

That's next, I'm sure. If you've read the thread, you'll see that I have two sound issues. One on the sound board and the other on the MPU. I haven't even begun checking out the MPU problem yet.

If a bad CVSD on my MPU is causing both missing speech ("Yo, Taxi!") and missing drums, then why do the drums magically reappear when I put in a known-good sound board?

I really don't think you understand the nature of the issues I'm having.

First off, there is no CVSD on the sound board. That's on the MPU board. The stuff generated by the CVSD leaves the MPU board through J16 and goes into J1 on the sound board.

For all the tests I've run, J1 on the sound board has been disconnected. The only connection between the sound board and the MPU is the ribbon cable, which is only used to tell the sound board what songs to play. I can start Song 1 in the music test, and then disconnect the data cable, and the song still plays with drums on the good board.

Unless the CVSD is able to beam its signal across time and space, there's no way the CVSD is able to keep the drums going under the music track when the MPU is not physically connected to the sound board in any way.

I'm not saying the CVSD on the MPU isn't also a problem (especially when it comes to the missing speech) but I'm working on one problem at a time. Right now, the problem I'm working on is the missing drums, which aren't present on my board, but are present on the known-good board, completely independent of the connection to the MPU.

I'll also make sure to replace every capacitor on both boards just to make sure.

That's why I like you. You don't really do sarcasm.

And we have drums!

Replacing the 1408 DAC (as well as transistor Q1) on the soundboard has restored the drums to the music, as well as the "bong" sound when the machine boots. (So much for "every sound on the machine runs through the CVSD".)

The irony is that I suspect the CVSD on the MPU is the reason I'm not getting any speech. I kinda figured that from the start, though. Time to get acquainted with that piece of business.

Well, I've taken a giant step back. I put the known-good sound board in to see if maybe my fixed-up sound board just wasn't mixing sound from the CPU properly, and I've discovered that the known-good sound board has stopped making drum sounds. I put my fixed sound board back in, and got drums again, but while poking around with the logic probe, I think I shorted two pins of the 1408 DAC I just replaced (possibly pins 3 and 4). Drums are gone on both boards now. I replaced the DAC with two different spares (easy to do, since I socketed it when I replaced it originally) and the drums are still gone. So I've gone and messed something up, and have no idea what, and I don't have a known good board to compare it against anymore.

I've got to say, I'm pretty bummed out that aside from the guy who wants me to replace the CVSD on my MPU board to fix the problems with sounds generated by my sound board, nobody else has offered their assistance on this. That means either nobody here is knowledgeable enough to help me, or nobody wants to help me. The first, I guess I can understand - though I find it hard to believe that there is exactly one person on Pinside with any knowledge of the sound system of a System 11 machine.

The other option is that nobody here wants to help me. If that's the case, then I don't know what I've done to deserve being shunned.

In any case, if nobody's going to help me, then I think I'll just end up parting this machine out. It's not doing me any good sitting here, and I'd get more of my money back doing that than trying to sell a broken machine.

I've read just about everything there is to read about this stuff. PinBot (System 11A) and Taxi (11B) are different enough in the sound department that a root cause for one isn't necessarily a root cause on the other.

I've de-soldered and tested all of the components that sit between the 1408 and the Op-Amp at U17. They all check out. I put the board back in, and no drums. This machine is screwed.

I'm not spending 35 dollars on a hard-to-find part without a reasonable chance that it could be the culprit. As I told the other guy, replacing the CVSD on the MPU isn't going to fix the problem caused by areas of the sound board that have nothing to do with the CVSD. I want to get that working again, and then I can focus on the MPU-related issues.

At this rate, I may end up buying you a new sound board.

Thanks for being the example of why I'd rather part it out.

I've already taken one known-good sound card from working to not working. I'm not spending a hundred dollars on a board that could go bad as soon as I put it into my machine.

I've read Clay's guides. I've read threads here, and on rec.games.pinball, and on KLOV. Most of them involve sound issues on High Speed, Grand Lizard, or PinBot, which are different iterations of System 11 - a generation of pins divided up mostly by how sound is handled. Issues with one sub-type of System 11 doesn't necessarily mean those problems will carry over to the other.

I have the tools, and I have the schematics. I just need someone to explain the circuit to me. I have a ton of questions about it, and no way of learning the answer on my own aside from going back to college.

I've picked up an oscilloscope, and I've done some initial tests of the sound board. The Yamaha synthesizer at U3 feeds into the DAC at U1, which outputs on two channels. Each of these channels goes into an input on the op-amp at U10. The op-amp outputs each go into their own 10uF capacitor (C3 and C4) before going to their respective resitors before going into the op-amp at U17.

The 1408 DAC at U11 is driven by the PIA at U12. I've no reason right now to believe the PIA is bad, because when I turn the volume up really, really loud, I can hear the drum intro at the beginning of the multiball music on Music Test 02. The output of the 1408 DAC ultimately runs through C8, though I still have no idea how the thing actually works.

This oscilloscope is two-channel, so I can hook up a channel each to C3 and C4, set the vertical knob to .5 volts per square and see the waveforms jamming along with the music. When I hook channel two up to C8, though, I get a pretty flat line.

IMG_3736 (resized).JPG

If I zoom in further, I can start to see some activity on channel two, but at that point, the zoom level is 50 mV per square, and that's some really sensitive testing. Who knows what incidental noise I'm picking up?

So, the fact that I can hear the output if I crank the volume up means the DAC is clearly able to generate some kind of really, really weak analog signal. Since I don't understand how the output of the 1408 is supposed to work, though, I can't tell if the weak signal is a fault of the chip or some component along the way. The output from C8 runs through R16, which the schematic says is a 6.6K resistor, but the part list, my DMM and the resistor itself says is a 10K. This is consistent with both of the boards, and both of these boards worked fine with this resistor being a 10K.

So, I now have some good data, but until I really understand how a signal makes its way from the 1408 to the op-amp input at pin 2 of U17, I'm still stuck. Now would be a great time for wayout440 to show up and give me some data I can use as a reference.

It can't be that minor if it's taken this long to get any help.

So far, I've had someone derail my thread to work on his own problem, a guy tell me to replace a part that's not on the board I'm working on, and a guy who wants to buy the whole machine from me for 500 dollars. That's not including you advocating sending the board to a guy who I've been told won't touch a board that anyone has ever done any work on.

The only person who's offered any help is LateCentury, and I'm going to dig in on that when I get back home. So, for that, I'm thankful. Thanks!