Gauntlet 2: Speak to me!

[DarrenF]

Hey Darren,

Just need a clarification, I posted an image from the board here:



I think you mean the chip labeled "A", but I wanted to make sure -- chip "B" is about the same coordinates, and is also an LM324. So, A or B?

The picture is just the bottom right corner of the board.

Thanks!

Well, as I mentioned in my 1st post, I don't have a Gauntlet PCB. I also can't see the letters and numbers around the edges of the PCB in that photo, to use them to locate an IC.

What I can do is look at the ICs in that photo and compare them to the locations given in the schematics (you do have a copy of the schems, right?). The TMS5220 is at 13R, the TM2151 is at 15R... so that's clearly the "R" row. The LS169 is 14S, and the LS273 is 15S... so there's the "S" row. The YM3012 is at 15T, so presumably the one you have labled "A" is at location 14T. The LM324 @ 14T is used for the YM2151 sound chip (not the speech).

The next row down should be the "U" row, and it looks to me like the IC you have labeled "B" is at location 13/14U (i.e. in row U, straddling columns 13 & 14). That's the amp for the speech sound chip.

 

[Vongoosewink]

Well, as I mentioned in my 1st post, I don't have a Gauntlet PCB. I also can't see the letters and numbers around the edges of the PCB in that photo, to use them to locate an IC.

What I can do is look at the ICs in that photo and compare them to the locations given in the schematics (you do have a copy of the schems, right?). The TMS5220 is at 13R, the TM2151 is at 15R... so that's clearly the "R" row. The LS169 is 14S, and the LS273 is 15S... so there's the "S" row. The YM3012 is at 15T, so presumably the one you have labled "A" is at location 14T. The LM324 @ 14T is used for the YM2151 sound chip (not the speech).

The next row down should be the "U" row, and it looks to me like the IC you have labeled "B" is at location 13/14U (i.e. in row U, straddling columns 13 & 14). That's the amp for the speech sound chip.

Thanks for the clarification, that's what I was afraid of -- looks like we did the wrong chip. The good news is that the sound still works, so my soldering must not be too bad!

No, I don't have the schematics, what I thought was the shems were just the joystick shems, alas.

Back to the shack, for another LM324!

 

[DarrenF]

http://arcarc.xmission.com/PDF_Arcade_Atari_Kee/Gauntlet/

The one that starts with "SP" is the schematic package.

page 7.

 

[Arcadenut]

Here is a High Res Photo of the board:

http://www.arcaderestoration.com/temp/gauntletfullsize.jpg

Try replacing the POKEY (if you have another Atari game, you can borrow one from it). It's the one numbered 137430-001 just above the Sound Chip.

 

[DarrenF]

Here is a High Res Photo of the board:

http://www.arcaderestoration.com/temp/gauntletfullsize.jpg

Try replacing the POKEY (if you have another Atari game, you can borrow one from it). It's the one numbered 137430-001 just above the Sound Chip.

Thanks for the photo. That seems to confirm my reverse-estimation of IC locations.

I doubt a bad POKEY would effect the speech, though. Attached is the schematic of the audio section. The POKEY is pretty much completely seperate from the speech IC (TMS5220), until their outputs are mixed and amplified. So unless the POKEY is jacking up the data bus, I don't really see how it could cause no speech.

 

[Arcadenut]

Thanks for the photo. That seems to confirm my reverse-estimation of IC locations.

I doubt a bad POKEY would effect the speech, though. Attached is the schematic of the audio section. The POKEY is pretty much completely seperate from the speech IC (TMS5220), until their outputs are mixed and amplified. So unless the POKEY is jacking up the data bus, I don't really see how it could cause no speech.

You're probably right, but according to the schematics, the POKEY goes to the 72LS273 at 13P which feeds into the TMS5220. So in theory, it could affect the Speech. I figure though that it's a lot easier test to do since it's socketed.

Could also be a bad ROM.

 

[DarrenF]

You're probably right, but according to the schematics, the POKEY goes to the 72LS273 at 13P which feeds into the TMS5220. So in theory, it could affect the Speech. I figure though that it's a lot easier test to do since it's socketed.

Could also be a bad ROM.

Yes, both the POKEY and the TMS5220 (thru the 74LS273) are connected to the sound data bus. That's not an 'output' of the POKEY, though... more of a shared input. It is possible, I suppose that the POKEY is holding lines on the bus high or low, but I'd expect that to mess up other sounds and the music (all 3 audio devices share the data bus). The OP only mentioned missing speech, not that other sounds are wrong or missing... so that leads me to think that the data bus is likely OK.

Of course, the best way to diagnose it would be to take a scope and work your way backwards thru the audio circitry until you found the speech output. But didn't get the impression that the OP has access to one, nor the ability to use one... so I was suggesting what I thought to be likely culprits.

 

[Vongoosewink]

Yes, both the POKEY and the TMS5220 (thru the 74LS273) are connected to the sound data bus. That's not an 'output' of the POKEY, though... more of a shared input. It is possible, I suppose that the POKEY is holding lines on the bus high or low, but I'd expect that to mess up other sounds and the music (all 3 audio devices share the data bus). The OP only mentioned missing speech, not that other sounds are wrong or missing... so that leads me to think that the data bus is likely OK.

Of course, the best way to diagnose it would be to take a scope and work your way backwards thru the audio circitry until you found the speech output. But didn't get the impression that the OP has access to one, nor the ability to use one... so I was suggesting what I thought to be likely culprits.

Yeah, the sound is fine, it is just the speech affected. It does make some sounds though -- when there is supposed to be speech ("Welcome, Elf" and "Use Keys to unlock doors") there is a faint popping and crackling, for the duration of when the machine should be speaking. It is very quiet though, I didn't even notice it at first. I would guess this is the result of the speech signal coming through incorrectly.

Darren is right, I'm not sure what a scope is, hopefully it won't come to that! Hope to be able to get a new chip on the board later this week (I have to go impose on my very patient uncle again!) If that doesn't work, we'll try another tactic.

Thanks guys!

 

[Vongoosewink]

Progress Report

Ok, here's the latest update. I've replaced the LM324 at location "B" above (now both A and B have been replaced), but the game will still not speak.

One change though, I don't know if this is good or bad, the game used to make a faint popping & crackling when it was supposed to be speaking. Now it does not. Normal game sound effects continue correctly, just no speech.

Any ideas?

 

[DarrenF]

There's also a bilateral switch (4066B) at 14/15U, that the voice goes through; as well as a coupling cap (C41). But I still think the LM324 is a likely culprit.

If it wasn't the LM324... there's still the 4066B (essentially a switch, the game uses to turn on and off the path for the speech) and the C41 cap.

I suppose you could try shorting the 4066 's input to the output to see if it's never closing.

The coupling cap: if you can't check for a signal before and after it with an oscilloscope, just replace it.

Or, of course, the problem could lie further upstream (on the digital side).

 

[Vongoosewink]

If it wasn't the LM324... there's still the 4066B (essentially a switch, the game uses to turn on and off the path for the speech) and the C41 cap.

I suppose you could try shorting the 4066 's input to the output to see if it's never closing.

The coupling cap: if you can't check for a signal before and after it with an oscilloscope, just replace it.

Or, of course, the problem could lie further upstream (on the digital side).

I'm not sure how to short an input to output, and I don't have an oscilloscope (anyone out there in Indy have one?)

 

[copious]

I have a similar problem.....with mine the speech is there it's just choppy.


i.e. Welcome Elf sounds like Wel.........C.......E

I'll thought the amp might be causing the probelm but I thought hte sound might be mixed there. I can replace the POKEY as I have a SW that I can borrow it from. I'll post after I do that and I might flip that CAP if I have the same value in my spare cap kit.

 

[copious]

It's not the Pokey. I replaced it with a known working one and the same results followed. I think it might be that switch; I set the menu to speech no and it is still using speech. I need to find a replacement locally if not I'll order one from the usuall culprits.

 

[Vongoosewink]

It's not the Pokey. I replaced it with a known working one and the same results followed. I think it might be that switch; I set the menu to speech no and it is still using speech. I need to find a replacement locally if not I'll order one from the usuall culprits.

Thanks for the input, please let me know how it turns out.

We may have the same problem, but mine is a more 'advanced' case -- no words at all, just crackle.

 

[copious]

Didn;t work....I'm going to order a new Speech ROM

 

[Vongoosewink]

Just for reference, a new speech chip did not fix mine. I think I need to get this into the hands of someone with an o-scope to figure it out.

 

[copious]

OK I have a new cap too I'll try that. The last part I can think of is the audio amp maybe one of the caps on it is dried out? I thnk the speech goes through one channel and the music through the other?

 

[copious]

Does anyone know how the audio amp on this works? Could that be the probelm? It has 2 channels could the voice be getting caught in one of them?

 

[DarrenF]

Does anyone know how the audio amp on this works?

Yes. Gauntlets apparently exist in two varieties. Those with a switcher for the 5VDC have a stand-alone "Audio PCB" for final audio output amplification. Those with only a linear power supply have an ARIII, which includes the audio amps. In both cases there are two TDA2030s; one for each channel.

However, speech is not exclusively on one channel. It is fed into both left and right channels equally back on the PCB (thru R87 & R88).

If the speech is present, but doesn't run normally, I might suspect the TMS5220 itself, or whatever drives its clock... which looks like an LS169 @ 14S.

 

[copious]

Wish I knew what happened? I moved the game from one side ot another and now the speech works fine. must be a solder joint?