Jump to content

midibob

Mecca Supporters 2018
  • Content Count

    789
  • Joined

  • Last visited

  • Days Won

    23

midibob last won the day on October 12

midibob had the most liked content!

Community Reputation

248 Excellent

About midibob

  • Rank
    Solid Silver Expert

Profile Information

  • Gender
    Not Telling

Anti-Spam

  • Are You A Spambot Again?
    No
  • Are You a Spambot ?
    Most definitely not.

Recent Profile Visitors

227 profile views
  1. midibob

    JPM system 80 cpu board repair

    Certainly a thorough job there Andrew. You'd never get that done with a commercial repairer that's for sure. I wouldn't even want to think what they might charge either. I do hate boards that have been got at as you end up with more faults to find and usually ones that don't make any sense.
  2. midibob

    Tales from Midibobs workshop, Sys80

    That's an idea. I can't really think of any other way of doing it to save build time and extra bits. Not sure what the JPM designer was on coming up with that one! I can't see any need to change the spacing?
  3. midibob

    Tales from Midibobs workshop, Sys80

    Thanks Paul, I did notice some chips that weren't obvious failures like no outputs were sufferring reduced and noisy outputs. How far they'd need to go out of spec to cause a problem is anyones guess but it's good to scope them just in case and swap as necessary. As you say the regulators definitely need a good overhaul too. Andrew, I assume you mean build it with changeable links to enable the correct pins? I could see the board getting put in the wrong position too ie. one pin out. Your right about the cost factor and demand. I did look into PCB buying and the breakpoint to get a reasonable quote was about 30 boards. One mistake in the layout and they all end up in the bin. Don't worry I've read your CPU board thread over and over.
  4. midibob

    Tales from Midibobs workshop, Sys80

    The board I use for RAM adaptors is double sided through hole matrix but with no tracks. It comes in various sizes and I cut what size I need. You can find them cheap on Ebay. For example:- https://www.ebay.co.uk/itm/Drilled-Double-sided-Copper-Prototype-PCB-Matrix-Epoxy-Glass-Fibre-Board-70x90mm/232524667208?hash=item36238cd948:g:CZcAAOSwu4BV6Xge:rk:30:pf:0
  5. midibob

    Tales from Midibobs workshop, Sys80

    Well, what a start to the morning. Connected the input port to ground, as I did yesterday, and of course nothing. Then I remembered it's got to be to the -12v, not ground. So what happened next. What can I say, all the CPU boards that I'd fixed blind all worked as did both the I/O boards. One point of note is the 12K board did actually get the 2.4KHz interrupt out to the CPU even without the link whereas the 8K's did nothing so that's solved that conundrum. Next step is to get some sort of test rig together to check all the I/P's and O/P's. That's the easy bit. Thanks to all for getting me off the bottom rung and Wizard with that snippet of code. So obvious when you know.
  6. midibob

    Barcrest MPU 3 reel problem

    That's not good. I assume the 2200uF 40v is good?
  7. midibob

    Tales from Midibobs workshop, Sys80

    Tomorrow could be a eureka moment then. I won't sleep tonight. Don't forget the different socket spacings on the 8K boards! I don't know how many variants there are or which is the most common. I can't remember where I bought my pin headers, probably the same as Wizards but I may have got them from RS, I'll see if I still have the invoice.
  8. midibob

    Tales from Midibobs workshop, Sys80

    Yep that's the johnny. Fresh head tomorrow and we'll see what happens although I did try it earlier this afternoon.
  9. midibob

    Sys80 PSU

    I've put all the info back in my workshop thread so that it all stays in one place. I assume the later PSU reg board was designed just to give a touch more of the 5v to the I/O and CPU as I was finding it dropping to about 4.8v before. It hasn't made any difference of course but I think TTL bottoms out at 4.75 so it's a bit close for comfort.
  10. midibob

    Barcrest MPU 3 reel problem

    In my experience reel shakes are normally caused by bad solder connections on the red connector or corrosion in the crimps in the cable. A quick shake of the cable would confirm it. The transistors do fail but I usually find it's all or nothing. I can't say for certain that a chip hasn't misbehaved in the past although IC5 can throw up some funnies. The reel driver chips again are normally all or nothing. Might be an idea to scope the outputs to see what's going on. Other folks experiences my shed further light. Is this all reels or just one?
  11. midibob

    Tales from Midibobs workshop, Sys80

    Hmm, interesting. I did see that in the game manual and tried it to no avail. Wouldn't it be awful if that input line was buggered! I'll persevere on that front. Thanks for the tip. Thanks for the offer of extra bits Nick but the one thing about having all the faulty bits is I'm going to know this inside out by the time I've got it working. (even if it does mean jumping in the Thames.) If I get desperate then I'll let you know. Onwards and upwards.
  12. midibob

    Tales from Midibobs workshop, Sys80

    Well the saga continues trying to get a working system. At present I have two I/O boards, an early one and a version 3. I'm not 100% convinced these are OK even though I have found faults on them with bad chips. For now though they appear to behave exactly the same albiet the difference in I/O enabling of course. I also have 4 CPU boards, one 12K and 3 x 8K. The 12K is fitted with Top streak with 3 ROMs and the 8k's have been tried with Steppa nudge, Double up deluxe (only 1 Rom) and Mystery spin. The 12K board appears to be the nearest to running in that the watchdog pulses are present after a reset, from IC15, but disappear affter a few seconds. During this time there is a 2.4KHz o/p from IC15 which makes it's way to the interrupt on the CPU (pin25). The reels then power up (don't spin) and then the watchdog kicks in and a reset follows. Disabling the watchdog obvioulsy stops it resetting the system but the watchdog pulses never return although the 2.4KHz stays active. The odd thing is none of the 8K boards let IC15 on the I/O board send out a 2.4KHz signal although they do let it generate the watchdog but again it dies after a few seconds. CPU's and RAM have been swapped across between both types of board and there's no difference. This is certainly turning out to be a head scratcher and reminds me of the trials and tribulations with BFM Sys85 and Scorp1. Luckily no custom chips with this one. As I'm still much of a novice on this tech all I have at present is the reel deck with an I/O and CPU card and nothing else. There's no looms or anything else plugged in although I don't think I'm missing anything crucial that would stop it starting, unless anyone knows different of course.
  13. midibob

    Sys80 PSU

    I've lifted the 5v regs and put thermal pads under them and now, as expected, the pots do what they are suppposed to do. I've adjusted them to give about 5v on the I/O board. The other reg board (rev F) with the diode fitted (1N4001) was a bit to high (5v6) so I've swapped it for Schotty diode which only has a 0.2v drop. After that brief interlude it's back to the grindstone to see if I can get the system working.
  14. midibob

    Sys80 PSU

    Not arf! Flippin watchdogs, 100Hz pulse feeds, 2.4KHz and 100Hz interupt routines I think I've been through the lot. It's always the same when you have all faulty items and you're trying to get a working system, it could be anything. It won't beat me but might give me a damm good thrashing!!
  15. midibob

    Sys80 PSU

    Hi chaps, Still beavering away at trying to get a Sys80 system going and despite finding many bad chips and broken/corroded tracks I've yet to get any I/O and CPU combination to actually get the reels spinning. Looks like I might be jumping in the Thames again. I thought I'd have a breather as I've been bashing my head against a brick wall for the best part of a week already. I've started looking at the later type PSU's and noted there are at least two revisions F & J. We have docs for the rev F but not the J. One of the minor differences is that there is a 5v adjust pot. If you follow the info in the manual it states to ground both the +5v and +12v regulators to the heatsink. On a couple of J boards here that has been done but then it shorts out the ground pin of the 5v regulator so the pot becomes redundant. Surely that's not right? Also on a rev F board there is a wire link L8 which has been removed and a diode put in. This lifts the 5v ground rail by 0.6v which I assume is a mod to get a slightly higher voltage to the CPU as there is a bit of a voltage drop.
×