What's new

Project: Helper board for Mitsu's Repro Batman Forever Sound amp board

mathewbeall

mathewbeall

Enlightened
Joined
Nov 6, 2017
Messages
2,376
Reaction score
2,030
Location
Mission Viejo, CA, USA
Hi Folks,

This is a the beginning of a thread where I start to gather names for people who will want to purchase one of these from me. Here is the current render of what the boards coming for V1.1 will look like - Still need to do full end to end testing, so its a possibility that this won't be the final version. The final version of the board will be BLACK to match the sound amp board, its blue below just because it stands out more and is easier to see.

This is a recreation of what Derrick2k started to build - so I can't take credit for much of this, other than learning how to use all the PCB creation software and get this out the door!

I do intend to opensource everything for this, including gerbers and a BOM with links - so anyone can build this on their own, and doesn't need to get it from me - edit: here is the github link: https://github.com/mathewbeall/batman_forever_jamma_helperboard

I don't know exact cost yet - but am guessing it will come in around $65-$70 USD plus shipping, which would be break even for me. My initial order is 20 boards, so 19 after I get mine built. If there is more interest past that I can make another board order.

Screenshot 2026-03-14 at 8.41.44 AM.png
Screenshot 2026-03-14 at 8.42.06 AM.png


pcb_front.jpeg
​


pcb_back.jpeg



Features​

  • Full JAMMA passthrough — Video, controls, coins, test, service all pass through untouched
  • Power distribution — Taps JAMMA 5V/12V/GND to power the sound board, fan headers, and voltmeter
  • Audio switching — Single 4PDT slide switch selects ST-V native audio or Batman sound board output
  • Stereo-to-mono downmix — Sound board's amplified stereo (BTL) output is summed to mono for the single JAMMA cabinet speaker
  • Button remapping — Two 4PDT slide switches shift P1/P2 button mapping (Batman uses buttons 2/3/4 instead of 1/2/3)
  • Line-level audio — Independent path with dual-gang volume pot, RCA jacks, and pin headers for external amplifiers
  • Voltage monitoring — Slide switch selects 5V or 12V rail for a 0.28" voltmeter display module
  • Fan headers — Two 3-pin fan connectors with 12V power
  • All through-hole — Zero SMD components, zero passives (no resistors or capacitors). Hand-solder friendly.

Please respond if you are interested in one of these, and I will start to build a list here as is normal so that when the testing is all done, I can start building these (as time permits outside of my normal job) and get them out to folks!

Final V1.2 completed builds.



Interest List (updated 2026-05-30)
24 boards available

1. RepairGuy95 — x1 — USA - Shipped
2. Dex74 — x1 — Germany - Shipped
3. Darksoft — x1 — Poland - Shipped
4. HardAzRockz — x1 — Tucson, AZ - Shipped
5. Orakelschaf — x1 — Germany - Shipped
6. clintkolodziej — x1 — Illinois - Shipped
7. DRAGONKZ — x1 — Sydney, Australia- Shipped
8. Mitsurugi-w — x1 — New Bern, NC, USA - Shipped
9. MudGuts83 — x1 — Oregon - Shipped
10. bodgit — x1 — UK - Shipped
11. VJP — x1 — Bern, Switzerland - Shipped
12. djsheep — x1 — Brisbane, Australia - Shipped
13. dillingerradio — x1 — TX, USA - Shipped
14. dlbenito — x1 — Spain - Shipped
15. 9999 Damage — x1 — San Francisco, CA, USA - Shipped
16. raffy92 — x1 — Japan - Shipped
17. bender3455 — x1 — TN - Shipped
18. XtraSmiley — x1 — Washington DC - Shipped
19. plasticfactory — x1 — Arkansas - Shipped
20. FLEX 3D - x1 - United States - Shipped
21. Pega$u$ - x1 - Washington - Shipped.
22. LoneWolf24 - x1 - San Bernadino, CA - Shippped.
23. Khatru311 - x1 - Las Vegas, NV - Shipped
24. chrisapple - x1 - California - DM Sent

​

Total: 24/24 claimed​

 
Last edited:
You can put me down for one. Been itching whether or not to upgrade my st-v multi. Wouldn't wanna miss out on this.
 
I should get the boards tomorrow - and I work this weekend to get all the testing done EXCEPT for the line level in/volume/out - those parts are still pending from Digikey and don't have a scan since 3/17 in North Dakota - hopefully will see movement on them tomorrow!

Should have an update by sunday evening of the testing I have been able to perform.
 
I got my last digikey package today - so was able to do a lot of testing tonight.

12V fan headers - good
button remap buttons - good
sound switch from passthrough (st-v) to batman forever (downmixing amplified audio to mono and out the jamma edge) - needs more testing
power harness - good
line level section (input, volume adjustment, output) - need to build another test board to test more (I bridged a channel somehow - the parts placement has to be done in the right order).

IMG_2539.jpegIMG_2540.jpeg
IMG_2541.jpeg

Video of the audio coming back to the jamma edge, through my CBOX supergun, through my tink4k and on my lcd.

View: https://youtube.com/shorts/BGhyZRRIkQw?feature=share



The one thing that I noticed so far at the beginning, that I can't really reproduce again right now after multiple reboots is a couple of times there was some low frequency "noise" or feedback that started and didn't back down - I was able to flip the switch to "audio bypass" and then switch it back and it cleared up. I do have a V1.2 of my board that has some resistors on the audio downmixing output to the JAMMA edge - because working with claude and doing some research, this is a better than that just essentially "connecting the wires together". This is what Claude had to say regarding the feedback I was hearing:


This is likely the exact issue V1.2's series resistors are designed to solve. When you directly parallel two BTL amplifier outputs (L and R tied together), small gain differences between channels cause one amp to fight the other — one tries to source current while the other sinks it. This can manifest as:

- Low-frequency oscillation / "motorboating"

- Intermittent onset (depends on signal content, temperature, and how the gain mismatch drifts)

- Clears when you break and re-make the connection (switching to ST-V and back resets the state)

The 0.22 ohm ballast resistors in V1.2 (R1-R4) are specifically there to prevent this — they limit the cross-channel current when the amps disagree, damping the oscillation.

I will try and finish up the line level testing today or tomorrow. Any thoughts on the downmixing issues I have seen? It would really suck to throw away 19 boards because of this, but I would also hate to send a board out that doesn't do the downmixing correctly.

Matt
 
I can reproduce it pretty consistently when the audio amp is cold - so from a bootup, go into the menu and go to the audio volume adjustment and just let it sit there, it will quickly have the audio issue, which can be alleviated by switching the audio bypass off and then on again - after the heatsinks warm up, the audio is fine. This is what Claude had to say (sorry if I am triggering folks who don't like the AI tools - but they are quite useful to me and I do challenge it if it doesn't sound right):

That's a really useful data point. It strongly reinforces the gain mismatch theory — when the amp is cold, the two channels have slightly different gain characteristics, and as the transistors warm up, their operating points converge and the mismatch shrinks below the oscillation threshold.

BTL amp gain is temperature-dependent, and the two channels won't heat up at exactly the same rate (different thermal paths to the heatsink, slight component variation). Cold start = maximum mismatch = oscillation. Warm = matched enough to be stable.

The V1.2 resistors should fix this regardless of temperature — they provide enough impedance to keep the cross-channel current in check even at cold-start worst-case mismatch.

At this point - I think I should order the boards with the resistors there to see if resolves the problem - and I might as well order the 20 boards again, doesn't make sense to order just 5 and then 15 later.

Once I get the line-level tested on the boards that I have - assuming thats ok, would work for anyone who doesn't care about downmixing back into the JAMMA edge.

More to come after the line level testing.
 
The line level control works as well - so I think I just need to order the new batch and see if that fixes the downmixing issue I am seeing.

Everything is ordered now for the final run of the new V1.2 board and all the parts - will have enough to make 24 of these.
 
Last edited:
You must log in or register to reply here.
Back
Top