Matt, I went back and checked around the address mode stuff. The PCB is hard wired on "rev k" for an address mode that works for eproms as far as I know. All the sigs were fine there and on every pin of every chip on the entire board. I wasn't interested in hacking the board to use 2 instead of 4 eproms mainly because I was trying to figure out what the specific problem is when all sigs are correct but a board still won't run. I had 3 boards in this state.
Dave hit the nail on the head. In the end, all 3 boards had bad ram. One board had 2 bad ram chips, the others had 1 bad each.
As you know, the ram chips are 2101/9101 22-pin x 0.4" DIPs that are soldered in, not socketed. Since it is labor intensive to swap them out, I avoid the task if possible. On my bad Rip Off board, I finally decided to try replacing the 3 ram chips a few weeks ago. I removed them all, installed new sockets, then plugged in what I thought were 3 new chips. I had bought about 20 new ones a few years ago. I just assumed they'd be good (I was wrong). When changing the ram didn't work, I was stumped.
Today I decided to install 3 new ram chip sockets into a known working Star Castle board. I swapped one chip at a time to test my so-called new chips. Five of twenty were bad. The rest worked. I dropped 3 of the "tested good" chips into Rip Off's new sockets and away it went. Worked great. I then replaced the ram chips on my last two remaining problem boards and they both came up okay as well.
So the answer to my original question "When all of the signatures are good and the eproms are good, what could the problem possibly be?", the answer is simply ram. Every one of my boards that had 100% good signatures in the end turned out to have bad ram and they all worked 100% once the bad ram was replaced.
Thanks to everyone who responded. It gave me things to think about and that helps a lot when you think you're at an impass.
If anyone is interested in buying one of my brand new reproduction CCPU exorciser boards, I have a webpage for them here:
http://www.biltronix.com/cinexor.html I would sell these boards without the wiring harnesses but most folks working at this level wouldn't have a problem making their own. I started using my CinExor board only a few weeks ago and with no prior experience with signature analysis, I have repaired 10 out of 10 of my CCPU boards. The ram was the only problem but now that I know what to look for, I doubt that it'll ever hold me up again.
William Boucher
----- Original Message -----
From: Matt Rossiter
To: Vectorlist
Sent: Tuesday, January 20, 2009 2:08 PM
Subject: RE: VECTOR: Cinematronics CPU board problem
On schematic page 4 of 7 at B8/7 - there are signatures related to "Address Mode". Have you verified those and is address mode set correctly?
You can use two 2532 eproms instead of four 2716 eproms without having to modify your board as outlined here. http://www.ionpool.net/arcade/cine/tailgunner-2532-replacement.txt
I've seen one case where the board would work with the two 2532 eproms but not the four 2716's. I never bothered to figure out why. :)
Are there any sections of the board that aren't tested with signature analysis other than parts of the clock and reset circuitry?
Matt
On Mon, 19 Jan 2009 16:43:18 -0500, "David Fish" <dfish_1@comcast.net> wrote:
Could be bad scratchpad RAM. Try replacing the 2102A’s @L-14, M-14 and N-14.
Dave
From: owner-vectorlist@vectorlist.org [mailto:owner-vectorlist@vectorlist.org] On Behalf Of Matt Rossiter
Sent: Monday, January 19, 2009 11:52 AM
To: Vectorlist
Subject: Re: VECTOR: Cinematronics CPU board problem
Importance: Low
Man, I feel your pain there. I have a couple of boards like that too.
Sometimes you need to reverse the clock position to get alternate signatures as listed in the manual.
Double-check the power-up circuit on Page 5 of 7
Make sure the JMI jumper is set correctly, which I'm sure it is.
You might have to scrap the signature analysis and start from the RESET line with an oscilloscope and work back that way.
Other than that - hopefully some of the experts on this list can advise you because I could use a little help as well. :)
Matt
On Sun, 18 Jan 2009 22:37:38 -0500, "William Boucher" wrote:
Has anyone ever been doing Cine CPU troubleshooting using an exorciser setup and found 100% of the signatures to be correct but the board still won't run? I'm having this problem with a Rip Off CCPU. It is not a mod'ed board (meaning it uses 2716 eprom's). I've tested or replaced about 30 chips including the eproms, a few bipolar d_roms, some 74LS163 counters that were messed up, a few gates, all the 74LS377's, the 74LS181's, the ram chips, all of the original chip sockets, and so on. Like I said, all the sigs are perfect. I tried the "load loop timer" test, the "spac" test, the "no-op" test, and the standard test. I've done them all 3x or more. The old original docs are loaded with numerical errors but I have sorted them all out already. The PCB is very clean, no open traces that I can find, no missing pads, no broken resistors or caps, etc. It is visually perfect. All the clock signals are perfect as well.
I had a stack of about 10 old CCPU's and I finally decided to repair them all. I whipped through the batch and fixed 6 of them within a couple of days. I'm now down to just 3 left. I probably should move on to the last 2 boards and come back to this one last, but I don't want to let it beat me. I'll keep at it even if I have to replace every last chip on the whole board one at a time if that's what it takes. At this point, I'm simply out of ideas.
One of my other boards that doesn't work also has all good signatures as well. I can only hope that it doesn't turn out to be the same nightmare all over again. I don't really understand how all of the signatures can be correct and the roms all verified and the ram replaced with known-good chips and the thing still won't run.
Any thoughts or suggestions?
William Boucher
http://www.biltronix.com/arcade_01.html
---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Sun Feb 1 20:22:49 2009
This archive was generated by hypermail 2.1.8 : Mon Feb 02 2009 - 01:50:00 EST