Hi

Thanks for the post. Interesting findings. I'm producing my own design Squawk & talk sound board See here - https://pinside.com/pinball/forum/topic/mypinballs-squawk-talk-sound-board-gordons-alive I have also experienced various issues with the TMS5200 ics and i am using NOS parts with a brand new board and all new components !!

My view is that the chips themselves have not always survived the 30/35 years in storage well. I have had new chips that sound totally different in a brand new board. Some too low, some too high. I use various sound roms from games that i know well during my board testing to check the TMS chips is correct and working ok! It can be expensive though, as the chips themselves were over £10 each to me and sometimes are no good.

I am using the original spec 130k resistor for the timing. It would be interesting to experiment with other values as you mention to see what could be done with them, though i like the board bom to be standard. (Will note a possible change to a pot for future revisions)

One other idea for this chip i have is working out how to create an emulation of it using the pinmame source and an embedded controller. I think this is the only way going forward

Final thought is that its a real shame the TMS5220 ics isn't fully functional compatible as it is pin compatible and does work in my boards, but unfortunately the speech pitch isn't quite right and some parts are sometimes garbled. I wonder if a different resistor value here may help to??

Yes using silver for pin legs was a big mistake!! I guess we are lucky they realised soon and ditched that idea. The TMS5220 ic uses tin legs to. Damn those TEXAS bods

The dac is used on all boards. the ay-3-8912 (psg) is only used on some. If the psg is installed then it does the background sounds. I've always wondered why bally didn't make use of the ay-3-8912's music abilities on this era of board. If you look at gottlieb system 80b sound boards, they used 2 of these to make some really great 80s music

The ay-3-8912 are actually 3 channel themselves to, so the board is in theory 5 channel

Exactly what i was going to add. It seems quite a wide range and this is written in the data sheet!! How come such a large range can be ok? It almost seems like you had to buy your chips from texas in the same batch then work out what the correct frequency of them was??

Interesting. It is possible that the DAC isn't used for Flash Gordon although populated as specced by the schematics. Flash gorgon was the first game to use the s&t board, so maybe Bally already had all the sound effects done for use with the PSG chip (like the earlier sound board) and so hadn't started using the newer DAC idea. The DAC isn't part of any boot up test either (not sure it could be) so the board will boot without it on any game. Of course games that do use it will have all the sfx missing. The psg is part of the boot up and the board will not boot if this chip is used and missing.

To me it seems like the idea was to move away from the PSG entirely and do all sounds through the DAC. Probably cheaper at the time, but i still want to know why they didn't add music capability to the boards. That would have really blown every other manufacturer at the time out of the water, and the board already is the best example for the time

Not sure why this rev board has a 240k speech timing resistor either. My boards with 130k in, 10pf shunt and flash gordon roms sound correct.

Yes that is correct. The interesting thing here is Flash Gordon being the first game to use the s&t board may not actually need the DAC, as all sounds are possibly made still using the PSG (like the earlier sound board does).

This is my thought to from seeing such as large range of resistor values specced in the data sheet. I am also wondering about the TMS5220 ic and if a different resistor value will cure the occasional garbling here to. The way the voice sounds i think will be slightly different (not the pitch) as the internal tables are slightly different, but they may be possible to get working in a much better fashion. The dac non install on fg still intrigues me, as i would have expected no sound effects if this was being used as per the other games. I don't think you can rules out the fact fg was the first game to use the new board and the older generation of sound boards were pure psg sound gen boards for sfx.

Really great thread to be discussing some top notch pinball tech from the old days.

No, the sound plus board is the computer sound board with an extra interface for speech, all sound effects are psg generated. See my new version of the board that works for both eras here : https://pinside.com/pinball/forum/topic/mypinballs-sounds-plus-sound-board-in-development

Here are some videos on my experiments with a 470K pot. I adjusted from around 100k upwards and back down again with the same new board, but swapping the speech chips over.

TMS5200 installed

TMS5220 installed

Kind of cool to be able to change the pitch on the fly, but it doesn't really help much with getting a 5220 chip to work better on s&t boards. Such a shame as it seems the coding tables internally are just that little bit too different for the speech to sound correct. Speech seems to be slightly scratchy and bait muffled. Would be a really great future solution to, as they use tin legs and are still much more readily available. Oh well guess its back to buying seriously expensive nos 5200 parts again with unknown storage histories...

Date code on 2 batches on nos TMS5200 i have currently are:

8236
8109

The 8109 ones are the only ones i have had issues with so far

The TMS5220 date codes are:

8340