Creating a Sony BKM-129X compatible board

[skum]

Hi,

I've been spending some nights working on creating a BKM-129X compatible input card. I've (seemingly) reverse engineered the MCU and its workings, implemented it on an Arduino Nano v3 and my PVM-9L2 detects it as a BKM-129X, serial readout works and so on. You can read more about it here: https://immerhax.com/?p=422 (no ads) and find the Arduino code here: https://github.com/skumlos/bkm-129x-mcu.
Should you want to try it on a compatible monitor, please do and report back here, as the only monitor I have which is BKM-129X compatible is the PVM-9L2. To test it, connect a 100 Ohm resistor in series with a 100K Ohm resistor, with one end of the 100 Ohm tied to SS/slave-select of the Arduino, and the other end of the 100K to 5V. Between the two, connect the SLOT ID (pin 31a). Also connect PD2 of the Arduino direct to SS (which works as the MCUs SLOT ID). The rest is direct, MISO->MISO (PB4 -> 27b), MOSI -> MOSI (PB3 -> 28a), SCK -> SCLK (PB5 -> 28b). This then requires 5V coming from USB, otherwise connect 5V -> STANDBY_5V (5V -> 29a/b). In this case I would suggest implementing the reset circuit also (look in the service manual of the card which is available on the web).

I've at the same time started creating the "complete" board (is completely untested as Chinese new year is in the way of me getting some boards quickly ) Here's a preview:
https://immerhax.com/wp-content/uploads ... m-129x.png

The Kicad project, if anyone is interested, is here: https://github.com/skumlos/bkm-129x-compatible

The idea is that a front bracket (3D printed) will end up securing it, and of course fill up the slot gap.

Any (constructive) criticism or alike is welcome You're welcome to copy and change this, but if you do, please keep it open, and give credit where due...

[fernan1234]

This is a really commendable project, and may be very helpful to other people working on recreating other rare input cards.

[Dochartaigh]

Very cool.

Have you thought about the BKM-68x card? That's the one people routinely pay over $2000 USD for....would be hard obtaining a sample to reverse engineer though... (if possible...I believe that one is a LOT more complicated with integrated chips and such).

[skum]

Very cool.

Have you thought about the BKM-68x card? That's the one people routinely pay over $2000 USD for....would be hard obtaining a sample to reverse engineer though... (if possible...I believe that one is a LOT more complicated with integrated chips and such).

Yeah this request has come up every time I've mentioned the 129X project, so definitely there's a demand, but as you say, it's infinitely more complex, really high component count, FPGA and CPLD, it looks like a complete signal board in an expansion port. I guess the FPGA code can be read out a serial eeprom, but CPLD code most likely cannot, and if it could, just getting a card and a monitor that supports it would break the bank. Having to shell out $150 for the 129X was Bad enough

[SuperSpongo]

Very nice, thank you for your work! I first saw your JVC mod. A 129X clone will be super useful for potential D14H5U buyers!

EDIT: Great read! I think that after a successful test with an RGB signal a next step would be to design a SCART version, perhaps simpler circuits for the RGBS lines. Maybe somebody will design a through hole version as a DIY project Exciting stuff!

[skum]

Very nice, thank you for your work! I first saw your JVC mod. A 129X clone will be super useful for potential D14H5U buyers!

EDIT: Great read! I think that after a successful test with an RGB signal a next step would be to design a SCART version, perhaps simpler circuits for the RGBS lines. Maybe somebody will design a through hole version as a DIY project Exciting stuff!

Thank you! Yes when I've confirmed the whole board works, based on the original board, a redesign of the RGBS lines would surely be a good idea at some point, as it is quite complex. A SCART version would be easy, and I will make one for sure. A THT version would also be an idea, to make it more available to lesser experienced solderers.

[maxtherabbit]

SCART? gross!

make it with a DE-15 VGA input k thx

[skum]

SCART? gross!

make it with a DE-15 VGA input k thx

SCART ftw! But does DE-15 make sense when there's no dedicated HV sync circuit? Because there's no big issues in adding whatever connector on the input side...

[maxtherabbit]

SCART? gross!

make it with a DE-15 VGA input k thx

SCART ftw! But does DE-15 make sense when there's no dedicated HV sync circuit? Because there's no big issues in adding whatever connector on the input side...

well you could add a sync combiner if you're feeling frisky, but otherwise just pulling TTL composite sync from pin 13 would be great

[SuperSpongo]

Nah, Scart is fine. People want to get rid of it. I'm not married to it myself, but it's what we have and I don't see the point in switching my SNES to a Multi-Out to VGA cable which doesn't even exist if I don't build it myself.

[maxtherabbit]

Nah, Scart is fine. People want to get rid of it. I'm not married to it myself, but it's what we have and I don't see the point in switching my SNES to a Multi-Out to VGA cable which doesn't even exist if I don't build it myself.

they do exist, there's already one on the market called the sync baby and I'm building a prototype of an original design as we speak

this nintendo multiout to VGA adapter dongle should be around the $25 price point

[fernan1234]

this nintendo multiout to VGA adapter dongle should be around the $25 price point

want it for nintendo multiout, sony multiout, genesis model 1 8-pin din, wii multiout, and saturn 10-pin minidin

[maxtherabbit]

this nintendo multiout to VGA adapter dongle should be around the $25 price point

want it for nintendo multiout, sony multiout, genesis model 1 8-pin din, wii multiout, and saturn 10-pin minidin

sony will not be supported with my design, nor any system which doesn't have c-sync on the multiout

no PAL Saturn, no PAL SNES - NTSC Saturn and NTSC SNES are good to go

[Listai]

This looks amazing!

I have a PVM20L2 so once this is a little bit more user friendly I'll definitely pick one up for some testing!

[skum]

This looks amazing!

I have a PVM20L2 so once this is a little bit more user friendly I'll definitely pick one up for some testing!

Thanks! As soon as Chinese new year is over, I should be getting some PCBs and then hopefully have time to solder them up quickly to "finish" the first prototypes, which hopefully just works

The current design of the RGB and sync circuits is a direct ripoff of the original card, with exception of a single transistor, the 2SA1462, which isn't readily available anymore apparently.

[skum]

So I just tried using a THS7374 as video buffer, since that's basically what the card does, and damn, it works a charm So I need to be able to switch the sync on and off (the whole video output can be done with the DISABLE pin of the THS7374), so I thought of simply using a ADG734 to be able to switch everything. This way the board will be extremely cheap, but of course include fine soldering... This requires a redesign of the whole board but its *much* simpler.

[RGB0b]

Excellent work. Remember to set the low-pass filter on the 7374 to "off", as cascading LPF's can soften the image.

[skum]

Excellent work. Remember to set the low-pass filter on the 7374 to "off", as cascading LPF's can soften the image.

Thanks, and yes! I was just going to put in a jumper to be able to set it to on if anyone wanted it, by having a pull-down. Never liked that type of stuff.

[maxtherabbit]

Excellent work. Remember to set the low-pass filter on the 7374 to "off", as cascading LPF's can soften the image.

Thanks, and yes! I was just going to put in a jumper to be able to set it to on if anyone wanted it, by having a pull-down. Never liked that type of stuff.

to the best of my knowledge there would be no reason to low pass the color signals in an analog monitor

[skum]

Excellent work. Remember to set the low-pass filter on the 7374 to "off", as cascading LPF's can soften the image.

Thanks, and yes! I was just going to put in a jumper to be able to set it to on if anyone wanted it, by having a pull-down. Never liked that type of stuff.

to the best of my knowledge there would be no reason to low pass the color signals in an analog monitor

Well cant think of any either, however, I'll leave it there, it can just be left out when mounting components.

[Listai]

Thanks! As soon as Chinese new year is over, I should be getting some PCBs and then hopefully have time to solder them up quickly to "finish" the first prototypes, which hopefully just works

The current design of the RGB and sync circuits is a direct ripoff of the original card, with exception of a single transistor, the 2SA1462, which isn't readily available anymore apparently.

I can't wait to see how this develops! - I was about to pull the trigger on some SCART cables for some of my component video capable consoles but with this in the wings I think I might wait it out and use the additional input for my other consoles.

[skum]

So a simplified board would be something like this:
https://immerhax.com/wp-content/uploads ... simple.png
Some of these resistors above and below the THS7374 will not be there in the final version, otherwise I guess components here are in the $20 range (maximum, BNCs being the most expensive).

[awbacon]

Awesome idea for a project. Best of luck with it!

[Dochartaigh]

Just wanted to add a note that you should be sure to test it on some of the weird Sony BVM models which have problems with sync. My 14L5, 20L5, 14L2, and 93L all work fine with the multiple 129x's I have...but it's those pesky BVM-D9's and BVM-D14's which can have weird, unexplained, sometimes totally random sync issues with the 129x RGB card on certain individual consoles (and by individual I mean one console on the same 129x card could work fine on one monitor, then that same exact console with same cable and same exact 129x could give you sync issues on another BVM!).

[skum]

Just wanted to add a note that you should be sure to test it on some of the weird Sony BVM models which have problems with sync. My 14L5, 20L5, 14L2, and 93L all work fine with the multiple 129x's I have...but it's those pesky BVM-D9's and BVM-D14's which can have weird, unexplained, sometimes totally random sync issues with the 129x RGB card on certain individual consoles (and by individual I mean one console on the same 129x card could work fine on one monitor, then that same exact console with same cable and same exact 129x could give you sync issues on another BVM!).

Well I can only test on hardware I have access to, and that's only the PVM-9L2. I'm trying to find someone in my country that might have compatible monitors, but none are the D9 and D14. It could be interesting to see if the card works worse or better than the original, as it really can't be called a clone in its current form, more a reimplementation.

[SuperSpongo]

Ooh, the THS is a great idea! Might be worth it to do a design for the JVC C01COMG clone cards as well.

[skum]

Ooh, the THS is a great idea! Might be worth it to do a design for the JVC C01COMG clone cards as well.

Yeah well, the THS7374 has the "issue" that it amplifies the signal, as it's a 2V/V amplifier. This means that in order to avoid overly bright image, I need to change the output resistors to match. I fear this can have an impact on other with a different input stage. However I looked at some of the other monitors compatible and it *seems* to be a non issue...

[RGB0b]

I just realized that the 129x card I use on my 9" BVM has the same exact sync issues as the BKM68x. Maybe a future rev of this card can include some kind of sync regeneration that ads the "VCR mode" the other BVM input cards use which fix the horizontal sync slice issue?

Now, I completely realize that's a big request and not something like "can you make it in red"...I'm just throwing the idea out there. Once again, thanks for this awesome project!

[skum]

I just realized that the 129x card I use on my 9" BVM has the same exact sync issues as the BKM68x. Maybe a future rev of this card can include some kind of sync regeneration that ads the "VCR mode" the other BVM input cards use which fix the horizontal sync slice issue?

Now, I completely realize that's a big request and not something like "can you make it in red"...I'm just throwing the idea out there. Once again, thanks for this awesome project!

Yeah I guess everything can be done, because as soon as the protocol for enabling the RGB is in place, which it seemingly is through the Arduino, then it's "simply" a matter of throwing in whatever components are desired. I am yet not strong enough in the intricacies of the sync signal to do it, but if I ever got a D9H, then I guess it would be an interesting project First tho, this coronavirus needs to gtfo, as a lot of production facilities in China are closed mandatory, including JLCPCB where I ordered the boards (at least where they produce the lead free stuff). So right now it's a waiting game, until they recover

[Dochartaigh]

I just realized that the 129x card I use on my 9" BVM has the same exact sync issues as the BKM68x. Maybe a future rev of this card can include some kind of sync regeneration that ads the "VCR mode" the other BVM input cards use which fix the horizontal sync slice issue?

Wait until you've owned 4 or 5 BVM-D9's and D14's)...and you'll see that they can each have individual issues with certain consoles. I've had one Sega Genesis for example work perfect on one D9, then plug it into another D9, with the SAME EXACT console, cables, game, and 129x card...and it'll do funky things. Think SNES (which is usually perfect) and of course TG16/PCE are other weird ones. And yes, I totally know this makes zero sense.

...so I don't know if a fix in this new 129x repro card would work for all of them as I don't know what exactly the problem is (or if it's the same problem on one D9 vs. another D9 or what).

Sidebar: VCR mode is actually in the BVM-D9's circuitry and menu (and assume D14's too), but is turned off when the 129x card is used - it does work on my BKM-127W cards just fine for composite or S-Video.