ukVac.com Homepage
Forum Home Forum Home > Technical > Fix logs
  New Posts New Posts RSS Feed - Chase HQ(s)
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Skin:


Chase HQ(s)

 Post Reply Post Reply
Author
Message
John Bennett View Drop Down
Senior Member
Senior Member
Avatar

5 Years of Supporting ukvac.com!

5 Years of Supporting ukvac.com!



Joined: 01 Jan 1998
Location: Newcastle
Status: Offline
Points: 18172

Feedback: 5
Post Options Post Options   Thanks (2) Thanks(2)   Quote John Bennett Quote  Post ReplyReply Direct Link To This Post Topic: Chase HQ(s)
    Posted: 07 Nov 2020 at 11:52pm

So, a (very) long time ago I offered to look at a pair of Chase HQ boardsets. Seemed a bit of fun – I thought I could do a bit of board swapping, score an easy fix and feel chuffed with myself.

Nope.

One boardset wouldn’t boot.

The other booted with a sound CPU ‘stop’ fault, the colours were wrong and the sprites were all messed up.

Swapping the video PCB over presented even more sprite faults.

So that’s 4/4 duff PCBs  

I looked at the non-booting board first. I checked the 68K reset line and I could see it was low. Whilst there is a typical reset IC, it passes through a custom – presumably for a watchdog or something. It didn’t seem to be getting out the other end. 

So I bypassed it . Naughty, perhaps, but I’m sure the owner can flip the mains if the game ever crashes.

Next, the lack of colours was simply a broken pin from the RGB DAC custom (that’s often snapped off completely, so watch for that if buying an ‘untested’ PCB).

After that, the CPU board worked great – sound, colours, the lot.

25% fixed…

Video PCB #1 next (dodgy sprites).

I stuck the scope on the address lines for the mask ROMs related to the sprites and I could see dead address lines. Looking at the board, there was noticeable track-rot. With two dead tracks linked, the sprites were fixed.  

50% fixed...

Next, the other video PCB. This was intermittent – the object layer corrupting itself over time.

I looked again at the object ROM address lines on the scope. When zoomed out to a long timebase, even though it was a blur of activity, I could see the underlying data starting to change as the intermittent fault appeared. The next job was to trace back to the source, which involved looking at outputs vs inputs of ICs – I think I went back through 5 pages of schematics before I arrived at a bad counter IC (not totally dead either, just to add confusion).

75% Fixed. 

Finally – the Sound Stop Error. This is where I gave up for a year

Anyway, at bootup, the master CPU talks to the sound Z80 to check it’s alive. This is done via the SYT custom IC. It’s not the easiest thing to debug as the main CPU sends a sound code occasionally and then otherwise the link is very busy sending FIFO checks, plus both sides are on CPU buses, so it requires a bit of triggering and bus deciphering (MAME helps).

Anyway, probing the Z80, it could be observed that the addresses were just cycling in an upwards-counting manner, so it wasn’t working right. No interrupts were occurring either (the YM2610 will send these once running).

The SYT custom handles the addressing for the Z80, allowing it to select ROM, RAM, YM2610 or cross-comms by accessing upper address locations on the memory bus. After a lot of probing I found that three address lines were broken from the Z80 to the IC and also to the RAM/ROM.

With these fixed, the Sound Stop error was gone and I could see far more normal activity (random-looking address accesses, interrupts firing from the YM and outputs from the YM).

But still no sound.

After far too many mis-steps (staring at the cross-comms, thinking no sound codes were going across the link), I realised that the OPO line was live from the YM2610 IC to the YM3016 DAC and there was an output, it was just rather small. The TL074 Op-amp afterwards was stuck at -5V for the two audio output channels. With this fixed, success!

4/4 boards fixed.

And 1 lesson learned – don’t fix boards for other people as it’s never as easy as you hope

But, to add one more, I bought myself a triple 68k proto Chase HQ (to try to emulate), which arrived with a video fault, however that was just a bad interconnect. But that's not relevant to normal boards as as the proto has DIP socket to ribbon cable adaptors everywhere as there's no mask ROMS, but dozens of EPROMs.

Anyway, might be something of interest. Such an ordeal, I had to post about it


Edited by John Bennett - 07 Nov 2020 at 11:58pm
http://www.philwip.com -sibling 'fix-off' (I’m losing by miles).
Back to Top
TheDaddy View Drop Down
User
User
Avatar

Joined: 25 Aug 2018
Location: uk
Status: Offline
Points: 4164

Feedback: 4.90909090909
Post Options Post Options   Thanks (0) Thanks(0)   Quote TheDaddy Quote  Post ReplyReply Direct Link To This Post Posted: 07 Nov 2020 at 11:58pm
Thanks for sharing John , i have 2 to fix myself i have had for ages. Hopefully get on them one day and once i am feelimg better.

Dave.
AKA Wonderboy. If there is such a thing as reincarnation , knowing my luck , I will come back as me !!
Back to Top
stevebm1 View Drop Down
Senior Member
Senior Member
Avatar

Joined: 05 Sep 2011
Status: Offline
Points: 6534

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote stevebm1 Quote  Post ReplyReply Direct Link To This Post Posted: 08 Nov 2020 at 7:55am
nice job,I know this game is a bit marmite for some people,but I love it
WANTED:

BALLY SOLID STATE PINBALL COIN TRAY
Back to Top
John Bennett View Drop Down
Senior Member
Senior Member
Avatar

5 Years of Supporting ukvac.com!

5 Years of Supporting ukvac.com!



Joined: 01 Jan 1998
Location: Newcastle
Status: Offline
Points: 18172

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote John Bennett Quote  Post ReplyReply Direct Link To This Post Posted: 08 Nov 2020 at 10:07am
Thanks. Yeah, I can’t actually play the PCB here, but very fond memories of the game (was in our local ‘swimming baths‘).
I’d love to see the DLX ride-in variant in the metal one day. It looks brilliantly absurd.
http://www.philwip.com -sibling 'fix-off' (I’m losing by miles).
Back to Top
neil1637 View Drop Down
User
User
Avatar

Joined: 30 Apr 2014
Location: cumbria
Status: Offline
Points: 3333

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote neil1637 Quote  Post ReplyReply Direct Link To This Post Posted: 08 Nov 2020 at 1:14pm
Amazing work JB. Seems a shame you can't boot it up and play it after all that!

Extra points for perseverance!
Please visit my Arcade Diary Blog at https://neilsarcadediary.wordpress.com/

Grail search - Donkey Kong and Pole Position upright.
Back to Top
biglouie View Drop Down
User
User
Avatar

Joined: 19 Nov 2015
Location: Kent
Status: Offline
Points: 1676

Feedback: 0
Post Options Post Options   Thanks (0) Thanks(0)   Quote biglouie Quote  Post ReplyReply Direct Link To This Post Posted: 10 Apr 2021 at 10:44pm
John, maybe I should ask this question elsewhere, but with the scope output, what are we looking at. Each of the address lines being monitored and you want them all matching their pulse, so things happen at the same time?
Back to Top
John Bennett View Drop Down
Senior Member
Senior Member
Avatar

5 Years of Supporting ukvac.com!

5 Years of Supporting ukvac.com!



Joined: 01 Jan 1998
Location: Newcastle
Status: Offline
Points: 18172

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote John Bennett Quote  Post ReplyReply Direct Link To This Post Posted: 11 Apr 2021 at 9:45am
Actually it was fairly basic.
Graphics hardware is really 'acceleration' on most arcade stuff. So the CPU doesn't access the address bus of the graphics ICs (or indeed the graphics ROMs). 
The CPU dumps codes into graphics RAM and the hardware/customs takes care of cycling the address lines of the graphics ROMs and taking the data out and putting it into the next custom (e.g. the layer mixer hardware). Ghosts and Goblins does all this too, just with discrete components.
Without that, the CPU would be reading a ROM and plotting a pixel and you'd get about one frame a second.

So, anyway... the graphics ROMs will be busy all the time and the lower address pins should all have activity if it's running a game and sticking tiles/sprites on the screen. So it was really just finding the address lines that were dead on a ROM IC that I knew was involved with sprites. So something you could do with a logic probe (for the dead tracks anyway - you'd struggle to see the fault I saw with the malfunctioning counter IC without putting a scope on the count lines out and watching them distort as the picture on the screen changed).

With a crashed CPU, you have to go a level further and look at the address lines to see what address it's getting to, which you can then use the memory map to help work out what it's trying to read from when it goes screwy.  

http://www.philwip.com -sibling 'fix-off' (I’m losing by miles).
Back to Top
terriblefire View Drop Down
User
User
Avatar

Joined: 29 Jun 2021
Location: Glasgow
Status: Offline
Points: 135

Feedback: 0
Post Options Post Options   Thanks (0) Thanks(0)   Quote terriblefire Quote  Post ReplyReply Direct Link To This Post Posted: 29 Sep 2021 at 6:55pm
Originally posted by John Bennett John Bennett wrote:

But still no sound.

After far too many mis-steps (staring at the cross-comms, thinking no sound codes were going across the link), I realised that the OPO line was live from the YM2610 IC to the YM3016 DAC and there was an output, it was just rather small. The TL074 Op-amp afterwards was stuck at -5V for the two audio output channels. With this fixed, success!


I have this exact issue but its unclear (at least to me) what you did to correct it. Was it the OP amps the dac or some other line that you fixed. 
Back to Top
John Bennett View Drop Down
Senior Member
Senior Member
Avatar

5 Years of Supporting ukvac.com!

5 Years of Supporting ukvac.com!



Joined: 01 Jan 1998
Location: Newcastle
Status: Offline
Points: 18172

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote John Bennett Quote  Post ReplyReply Direct Link To This Post Posted: 29 Sep 2021 at 7:21pm
Heh, it’s not clear for me either reading it back.

Pretty sure it was the opamp in the above case - I didn’t need anything exotic that wasn’t off the shelf for any of it.
http://www.philwip.com -sibling 'fix-off' (I’m losing by miles).
Back to Top
terriblefire View Drop Down
User
User
Avatar

Joined: 29 Jun 2021
Location: Glasgow
Status: Offline
Points: 135

Feedback: 0
Post Options Post Options   Thanks (0) Thanks(0)   Quote terriblefire Quote  Post ReplyReply Direct Link To This Post Posted: 29 Sep 2021 at 8:52pm
OK thanks. I'll swap the OP amp out and see what happens. That was my working theory that the OP amp was dragging the sound to saturation. 
Back to Top
 Post Reply Post Reply
  Share Topic   

Forum Jump Forum Permissions View Drop Down



This page was generated in 0.228 seconds.