HB-F1XV black screen

Page 2/5
1 | | 3 | 4 | 5

By Pentarou

Master (226)

Pentarou's picture

24-10-2019, 02:20

Quote:

Is there are reason why would the Z80 leave D0 high after reset?

Don't know. Maybe it's in high-z mode (halt? /Busrq?)

Quote:

it was desoldered/repaired by the previous owner

Did you check if the blue tracks on the underside are damaged? I've read that those are very delicate.

Quote:

moreover it was 5.88V at the ROM-D0 pin, but Z80 max input is 7.7V so should be fine

No, that's NOT fine (unless you somehow took a wrong measure), RAM is damaged by going over 5V.

Quote:

I have no device where to test the Z80 so I will have to build something

I found this sometime ago:
GhettoFluke9010A
I wanted to use it to repair some arcade PCBs, but never got around to test it. It seems it's still at v0.1 don't know if it's usable.

And finally:

Quote:

tried also with megaram but no difference

What do you mean?

By fallen1

Resident (43)

fallen1's picture

24-10-2019, 18:38

Quote:

---Is there are reason why would the Z80 leave D0 high after reset?
Don't know. Maybe it's in high-z mode (halt? /Busrq?)

This are the CPU signals when everything is plugged back in:

Will do my homework and read up on the Z80 states...

Quote:

---it was desoldered/repaired by the previous owner
Did you check if the blue tracks on the underside are damaged? I've read that those are very delicate.

I did, and all seem to be visually OK. Only damaged trace that was somehow repaired was coming from the HIC + the cut grounds on A/V out connectors.

Quote:

---moreover it was 5.88V at the ROM-D0 pin, but Z80 max input is 7.7V so should be fine
No, that's NOT fine (unless you somehow took a wrong measure), RAM is damaged by going over 5V.

You are right, the RAM can handle 5.5V max. The 5.88V was Vmax on the oscilloscope (it's also on the screenshot I posted earlier - just not very readable). When measuring directly at ROM:D0 using multimeter, it reads 4.4V so not sure what to make of that

Quote:

---I have no device where to test the Z80 so I will have to build something
I found this sometime ago:
GhettoFluke9010A
I wanted to use it to repair some arcade PCBs, but never got around to test it. It seems it's still at v0.1 don't know if it's usable.

Great idea! Got only the small arduinos so will try to build something like that just for the Z80 + shift registers. Let's see

Quote:

And finally: --- tried also with megaram but no difference
What do you mean?

That was one of the tips I found on this forum - if your RAM is not working, use RAM expansion cart and Z80 will find it.
(I checked now and the correct name is MEGAFLASHROM, sorry for confusion)

So the plan is:
- Try to get image of the ROM + confirm it's still working
- Get Z80 out and build a Z80 tester - I'll rather order a spare z80 just to save the original until the tester works :)

Thank you for your time and help!

By fallen1

Resident (43)

fallen1's picture

25-10-2019, 18:30

I was able to dump the ROM (in 40 minutes using ESP-8266 Smile) and it looks to be fine. No stuck bits.
I sliced the dumped file into 16k slices and then tried to align it to ROM files from BlueMSX. This is the full ROM layout (IC12):

Files marked red had differences on very few places, but consecutive, so I assume no issues with the ROM's reading - most likely different SW version. After merging the two 32k slices, I was able to use all inside BlueMSX. That means my problem is elsewhere...now on to building Z80 tester

(Not sure if I can post the dumped rom here, but let me know and I will. Should be easy to merge it from the available roms at bluemsx homepage).

By Grauw

Ascended (8516)

Grauw's picture

25-10-2019, 19:36

That dump is something the openMSX devs are probably interested in!

By fallen1

Resident (43)

fallen1's picture

25-10-2019, 20:05

Grauw wrote:

That dump is something the openMSX devs are probably interested in!

Sure, should I post url here?

By Grauw

Ascended (8516)

Grauw's picture

25-10-2019, 20:07

I think that’s fine.

By fallen1

Resident (43)

fallen1's picture

25-10-2019, 21:37

By fallen1

Resident (43)

fallen1's picture

26-10-2019, 02:13

There is 0.0 resistance between VDP-D0 and IC6-D0 (same for VDP-D5 and IC6-D5) but other data bus lines have 6.6k there (measuring unpowered).
The IC6 is a 3-state octal latch (74LS373) and measuring it's data pins to ground gives constant resistance. The same goes for VDP's data pins and ground. Can I assume that there are 2 not working resistors on the bus somewhere? (CPU-VDP data pins have constant resistance of 10 ohms on all)

By fallen1

Resident (43)

fallen1's picture

26-10-2019, 11:50

took the 74LS373 out and it's outputs Q1 and Q5 have measurable resistance (~2M), rest is in high impedance/disconnected. Will get replacement on Monday and let's see...

By fallen1

Resident (43)

fallen1's picture

26-10-2019, 11:59

The Z80 tester won't be necessary for now, as there are no /BUSACK signals; /M1 + /RFSH + /WAIT + CLK + /MREQ and Address bus are busy; /HALT is high - I assume that CPU is working (reading stuff from memory and is getting nonsense back).

Page 2/5
1 | | 3 | 4 | 5