ukVac.com Homepage
Forum Home Forum Home > Technical > Fix logs
  New Posts New Posts RSS Feed - Time Pilot 84 Repair Using Slice
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Skin:


Time Pilot 84 Repair Using Slice

 Post Reply Post Reply
Author
Message
GadgetFreak View Drop Down
User
User


Joined: 13 Jul 2016
Location: Norfolk, UK
Status: Offline
Points: 897

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote GadgetFreak Quote  Post ReplyReply Direct Link To This Post Topic: Time Pilot 84 Repair Using Slice
    Posted: 03 Jan 2020 at 4:56pm
Right, I have to make this less scientific and simpler than my last one

I hate working on this type of Konami board as in the past I have had to replace so many of those high sided sockets due to poor connection and this board was really dirty. So first up was a wash & scrub followed by a good 24 hours for the board to dry off. It cleaned up nicely.

I then started checking all the mask roms and found that 3 of them had died, the character GFX at 2J, sub program at 10D and the sound program at 6A. Then I worked through all the customs checking them on my TP84 board (082, 2 x 083, 502 & 503) and also checked the two 68A09 CPU's which all worked fine.

On booting the board I just got a screen full of garbage:



So first thing was to use the logic probe on the CPU reset lines (pin 37), sure enough they both had the high to low change just after powering up and then stayed low so no watch dogging. I also then did a quick check of the program roms at 8/10J & 10D, all of them were showing address & data line activity.
My initial thoughts were work ram or address / data bus issues so I checked the schematics and started on the bus buffers 11G/H/J and 11D/E/F.



So I changed it out:


So a definite improvement but the board still wasn't running so I moved onto the work ram which is a 6116 @ 9F but doing a power up grab to get the ram clear writes to the ram, proved it was fine.



Next I decided to check out the four 2114's at 8B/C & 9B/C. Three of them checked out fine using the same power up method to catch the ram clear.



But 9C failed the test:


Finally I now had a board that booted though the display wasn't right and it would frequently go into a reset frenzy.



At this point I was out of obvious ideas and decided to work through the board in a logical manner, adding each location & chip to Slice then doing a few grabs to see if the chip passed. It only takes around one and a half minutes per chip so is quite quick.

I found the 74ls377 @ 12D was failing:



Replaced it, Slice then passed it, but no difference to display so more faults to find....
It was at this point that the board stopped running again. So I concluded another chip had just failed and so started going back over the chips involved in memory addressing and clocking. Just about the last chip I checked was the 74ls74 @ 14G which gets fed by a horizontal clock and then via some other logic, controls a couple of the 74ls138 enables. I already had a results grab which showed it working but when testing again the outputs were static.

First Test:


When Re-Tested:


After replacing it the board fired up again so I carried on checking chips with Slice. I then found the 74ls174's @ 3C & 3G failed a Slice test. One had a single stuck output and the other had a lazy output.



Then yet again while doing more tests the board stopped booting! This time I decided I had better check the two mask roms @ 8J & 10J. Sure enough the one @ 8J had died. I burned a new one and the board booted again for about 5 minutes before it just changed over to giving me solid colour over the whole screen. The colour did change every so often so I was pretty sure the board was still running correctly.
So I started checking the chips directly involved in the character display and soon noticed that the clock signal was not present @ 2E/F/G. This lead me back to re-check the 74ls04 @ 6F which Slice then failed showing that pin 10 was stuck low instead of pulsing at 6.141Mhz.
Changed it out and the display came back to life but I still had this reset issue which would happen at the start of the second demo run through. My first thought was timer but by checking the 74ls10 @ 13G I could see that the reset was not a timer issue. I then checked the 74ls32 @ 15J and could see it was either the MAFR or SAFR signal.

I also noticed the two solder pads between 8G & 9G and that shorting these would disable the reset. So I did this and rebooted the board, now when it got to the second demo run through, the sprites froze on the display but the background carried on scrolling and the game carried on playing. So I started to check the remaining chips at which point I found the 74ls244 @ 10A was failing Slice, I also noted it was the same Fujitsu batch as the one I had already replaced! So I replaced it and the board stopped resetting.

I still had no sound, just a constant tone, so having already proven the sound cpu was running with the logic probe I decided to check the 2 x 2114 ram chips being used for work ram. Both came up as faulty.

Both failed with the first location:


After replacing they both tested fine:


Now the board was fully working and I hadn't actually needed to replace any of those sockets that have given me so much trouble in the past.

Anyway, fast forward 24 hours and I fired the board up again to make sure it was stable now. It was for about 50 minutes! Then it started flickering the display, it was like every other pixel was being randomly turned off and also some of the background gfx were corrupting.



Checking the schematics I noticed there was another 74ls174 that controlled some of the address lines on the colour proms. On looking more closely I also noticed it was the same batch as the two Fujitsu 74ls174's I had already replaced. So I connected the clip and Sliced the chip.



So as I expected faults on every output. Once I changed it out the board was working again but for how long I don't know.

In total I have replaced:
2 x 2764 Mask Roms (10D, A6)
2 x 27128 Mask Roms (2J, 8J)
3 x TMM2114 (9C, A2, A3)
1 x SN74LS04 (6F)
1 x SN74LS377 (12D)
2 x MB74LS244 (10A, 11D)
1 x MB74LS74 (14G)
3 x MB74LS174 (2B, 3C, 3G)

This repair was made more difficult with the fact that as I was repairing it new faults were developing. But considering the board had probably developed a fault some 30+ years ago and been shelved, then kept in cold, damp storage ever since it's amazing that it even works now.
It is always a challenge to get a board back up and running after so many years. I always find it's a good idea to run the board for several hours after repair at random times over a couple of weeks to allow any faults to develop and show themselves.
Back to Top
virtvic View Drop Down
Senior Members
Senior Members
Avatar
The Cabaret King

Joined: 04 Jul 2010
Location: Whitton, Midlsx
Status: Offline
Points: 7296

Feedback: 4.764705882352
Post Options Post Options   Thanks (0) Thanks(0)   Quote virtvic Quote  Post ReplyReply Direct Link To This Post Posted: 03 Jan 2020 at 6:07pm
Nice work for a good game!
THE PENCE IS MIGHTIER THAN THE SWORD!

Check out the podcast

http://tenpencearcade.co.uk

Got cabarets? > CALL ME!
Back to Top
Digimon View Drop Down
User
User


Joined: 16 Sep 2016
Location: Hamburg - GER
Status: Offline
Points: 250

Feedback: 0
Post Options Post Options   Thanks (0) Thanks(0)   Quote Digimon Quote  Post ReplyReply Direct Link To This Post Posted: 05 Jan 2020 at 10:04am
Good Job!
Back to Top
strykr View Drop Down
Senior Members
Senior Members
Avatar

Joined: 10 Nov 2010
Location: Eastbourne
Status: Offline
Points: 11419

Feedback: 5
Post Options Post Options   Thanks (0) Thanks(0)   Quote strykr Quote  Post ReplyReply Direct Link To This Post Posted: 05 Jan 2020 at 10:09am
Nice fix! And thanks for posting
Wanted - Electrocoin Wheels Runner Cabinet http://tinyurl.com/oq6dkau
Check out my 80s Synthwave Music :)   https://247retro.bandcamp.com/album/naeon
Back to Top
 Post Reply Post Reply
  Share Topic   

Forum Jump Forum Permissions View Drop Down



This page was generated in 0.268 seconds.