Omega MSX2 - function trouble

Supporter (12)

07-10-2021, 21:39

Hi. I'm building Omega MSX2 step by step using the instruction from here
https://msxmakers.design.blog/proyectos/omega-home-computer/...

I'm in step 8 and I'm having trouble displaying the characters. There are other characters in the words or some are missing (after reset is display different - different errors in words). Keyboard works only a few seconds and then system freeze.

I tried to replace the circuits: U1, U2, U7 and U26 but the result is still the same. Circuits are new from Mouser or from MSX Makers (they should be tested and programmed by MSX Makers - GAL, ROM).

I did not try to continue in the next steps. Can't it just be a problem that not all circuits are connected and is there any interference or signal reflections from the empty sockets?

Images previews

Full images
https://ibb.co/njdvNG0
https://ibb.co/vscXDZV
https://ibb.co/wLMTz8n
https://ibb.co/P1XzPVg

Aangemeld of registreer om reacties te plaatsen

Van Grauw

Ascended (10681)

07-10-2021, 23:06

It seems a VRAM issue to me.

Are the CAS / RAS timings of the RAM chips you used fast enough for the V9938?

Van mig01

Supporter (12)

08-10-2021, 08:47

I'm using V9958. I looked at other designs and they use VRAM with 10ns timing as well. Why would the boys from MSX Makers sell me incompatible VRAM?

What about capacitor trimmer C89? Any informations are here but I don´t speak Spanish and I don't understand.
https://www.youtube.com/watch?v=0QRl3Vd0eok&t=6560s

SRAM I have this:
https://eu.mouser.com/ProductDetail/913-AS6C4008-55PCN

I have an oscilloscope so I can measure the required waveforms according to the link:
https://msxmakers.design.blog/proyectos/omega-home-computer/...

It's weird to me that the keyboard freezing a few seconds after the reset. Reset sometimes does not work and it is necessary to turn off the computer, turn it on. I have noticed that the MAX691 has short pins and does not fit perfectly in the socket. I ordered a new MAX691.

Van msxmakers

Champion (501)

15-10-2021, 10:59

mig01 wrote:

Hello Mig01.
MAX691 can be the cause, try just starting the computer wile pushing the MAX chip in. You can also try to overlap a second socket over the first one.
Please, let me know if that chip was supplied by us, we will do what you need to help, but send us an email because I cannot identify you as Mig01.
The computer can work with C-BIOS without an RTC working properly. Do you have how to program your BIOS with it? Did you ordered us a Halt & C-bios ROM chip?

Tell me how it was after doing all that.

Van Edevaldo

Master (154)

16-10-2021, 20:11

Quote:

I'm using V9958. I looked at other designs and they use VRAM with 10ns timing as well.

They are 100ns, if I recall correctly even 120ns would work fine.

Post the the part number of the VRAM chips here. There are minor subtleties between the 64kx4 memory models and the VDP was not compatible with all of them. It may be worth double checking.

It is also good to have spares to test as well. Best case those chips are 30 years old NOS. But more likely used and not handled with ESD protection. You may swap them into different positions and see if the problem changes.

Van Edevaldo

Master (154)

16-10-2021, 20:49

It almost looks like the VDP is being written to too fast. Like when "auto" is replaced by "auoo". A first letter ends up being replaced by the next. But the "ggto" case is the exception. Hum...

Do you see the same behavior every reset? Are you able to reset the board without powering it down and compare?

It is interesting to see that this motherboard has a proper reset circuit / supply monitor (the MAX chip). Most MSXs I saw just have some sort of delay pulling reset low while the supply is powering up. This makes me think the problem could be your 5V supply. If the connection is too long, or the 5V is not well regulated, of there is too much capacitance on the board and the 5V supply was not expecting it. Also the MAX chip used has a 50ms delay to release reset (what is good but potentially small if your supply is not really up to it). All this can cause a situation where the supply is not really stable when reset is released. MAX695 is a similar chip with a longer 200ms delay.

If you are in doubt about C89/C90, just leave them sort of in the middle position. You can make minor (tens of ppm) adjustment on the frequencies with them. But if C89 is left close to their minimum/maximum the circuit may not work properly and make the system unstable. Middle should be ok even if the frequency is a little off.

Van Grauw

Ascended (10681)

16-10-2021, 21:35

Edevaldo wrote:

It almost looks like the VDP is being written to too fast.

Is the M1 wait circuit functioning correctly? You should be able to check with the scope.

Van mig01

Supporter (12)

18-10-2021, 09:11

Thank you all for your suggestions. The problem seems to be solved. The first one... I replaced the MAX691. The original circuit had short terminals and did not fit well in the socket. I bought a new one from www.tme.eu (they have cheap transport) and it already fits well in the socket. This solved the reset issue (when I press the reset button sometimes it worked and sometimes it didn't - computer not booting, I see black screen and they do nothing) and the computer stopped freezing. When I installed the rest of the components the problem with displaying text was solved. Signal reflections were probably created there as the paths on PCB were not closed with components. I now have on the board all integrated circuits and so far it seems that everything is working fine. I've been playing with it for over half an hour and I haven't any problems yet. I still bought a spare VRAM on ebay, but they probably won't be needed.

Thank you.