RFC: Williams CPU Tester Project
beaver
board tester.
After I wrote about my System 9 Playfield tester, I received a few
e-mails about building a tester for the CPU board. Since my playfield
was drying, I decided to explore this some more and have developed the
following concept:
http://www.edcheung.com/album/album05/pinball/cpu.htm
Questions:
- Have I left any interfaces out?
- Changes you would make?
- Any ideas on the best cabling concept?
- Would you buy one?
Thanks to Capt Neo for his encouragement in this project.
Edward Cheung (donning flame suit)
Chester Pete
to support you for when you do your next project!
But IMHO, you would have more "customers" if your device(s) were a PLAYFIELD
tester. Be aware (don't take this the wrong way - what you are doing is
great) that a CPU tester may be more a "vanity" project for yourself rather
than something that others want. Just be sure of what you are trying to
achieve.
Consider this, the people who really want a CPU board tester are the small
number of board repair people out there. There are not many of those around
and lets be honest, they are quite cheap to use - Most of them probably
have their own board testers already - you did have a comment from you!
If you are aiming this for the pinball owners then I'm not sure it is
useful. The pinball owner normally only wants to know if the board works OK
or not.
If it doesn't, they will send it in for board repair or buy another one. Why
hook up your tester to find out something they probably already know - the
board doesn't work properly.
Now if you made a PLAYFIELD tester that would be a different story. I know
it sounds the wrong way round but what do pin people mostly do? - they shop
the playfield! A playfield tester that ensures a board isn't damaged on
switch on would be a definate plus for the majority out there and I think
many people would buy/want one!
To me the responsibilities of such a device would be to protect the CPU
board from damage (before you find out what is wrong on the playfield)
If all is OK, the machine diagnostics would then be used to find the
problem.
In other words, find the gap and fill it - don't replicate something that is
already OK.
N.B. Most of the pin problems I have found have been on the connectors,
wiring or playfield devices (if you discount the classic battery connector
problem)
and lets be honest, It is a playfield problem that usually causes board
failure. If it is anything else, then it really is the domain of the
dedicated repair specialist.
Perhaps a playfield tester could be your second project? Personally I would
prefer it to be your first!
Best of luck
"beaver" <e...@edcheung.com> wrote in message
news:1131713796.4...@g44g2000cwa.googlegroups.com...
kenn...@yahoo.com
System 3-7 CPU/Driver/Power Supply/Sound tester. It's modeled after
the factory Williams one and works much the same way. It was almost
ready to bring to TPF, but didn't make it. I'll have it ready for the
show in Houston in the spring, depending on if/when that happens.
K2
beaver
ago, I wrote about my playfield tester project:
http://www.edcheung.com/album/album05/pinball/playfield3.htm#tester
When I wrote about it on this forum a few weeks ago, a well-known
member stated that I was wasting my time on a playfield tester. I was
pretty surprised by that comment, and it sounds like you would agree.
Here is that thread:
I build these projects because I find it fun to do, but I am also eager
to contribute something that others will find useful. I hope to hear
more feedback...
Edward Cheung
Chester Pete
No probs. Wish you the best success. As I say I would definately but your
CPU tester but I'm not sure I am typical !?!
I'll repost what I said then in your original thread. The point I'm making
is that a simple set of devices might be all that is needed to fill the gap.
N.B. The playfield tester you made before was not mobile and small enough to
fit between the CPU board and the playfield - that to me is the clincher.
What do you think?
If I don't make sense then please tell me why or ask for more of my mad
ideas!
If you make it - it saves me from doing it!
----------------------------------
That is bizare! I have been thinking about building something similar for
myself.
Heres the idea:
instead of making one big test rig, make several small dedicated boxes. They
have to be portable enough to take to a machine and plug them in-situ.
Each box is dedicated to a particular connector for a model (e.g. system 7)
and they monitor the signals going to AND from the solenoids, switch and
lamp matrices.They have a battery and an earth strap
Each box has LEDs which check for particular things.
The solenoid driver connectors
- Check for always on (transistors burned)
- Check for output short circuit
- Memory for an output being grounded (turned on by cycle test and reset by
pushbutton)
- Test for short circuit
The switch matrix and lamp matrix
- check for double triggers and light appropriate LEDs (the cycle test
shows this) - If more than two LEDs fire then an "error" LED fires and
remains on.
The other aspect is that you can make another "set" for ingame play. This
would be used for checking the action of solenoids with a fast acting trip)
What do you think?
"beaver" <e...@edcheung.com> wrote in message
news:1131751219.6...@g47g2000cwa.googlegroups.com...
Chester Pete
I have a problem right now that I want to solve. It doesn't affect
gameplay but it is anoying in that "I want it to work perfectly". One day I
will look at it!
Its a Fire! machine that has a switched lamp which faintly glows when
another lamp is supposed to be on. It would be so much easier if I could
install an "in-line" device which showed the ins and outs - I could quickly
identify if the problem was on the playfield or the driver board when going
through the tests.
"Chester Pete" <pREMOVEf...@aol.com> wrote in message
news:dl3c2n$754$1$8300...@news.demon.co.uk...
Dallas Overturf
spare set of boards available!
Most PF tesing is covered in the game diags. And that woud be much more to
your advantage AND cheaper
than a PF tester. It's bringing up the MPU and Driver circuits that is a
pain.
Regards,
Dallas...
--
Home Page: www.geocities.com/d_overturf
Replies regarding News Group messages to me may get a challenge verifier
back. (CHAP).
The Sender may be >
Subject: [Challenge Response] Confirm your email by visiting this URL
https://smtp.stowetel.com/challenge/... SomeURL.
If so you need to Click on the link in the challenge and do what it says.
This is usually a one time deal.
---------------------------------------------------------------------------------------------------------------------------
"Chester Pete" <pREMOVEf...@aol.com> wrote in message
news:dl3deq$8tn$1$8300...@news.demon.co.uk...
Fred Kemper
I don't know what need I would have
for it myself, though.
--
Fred
TX
CARGPB#8
******************
"beaver" <e...@edcheung.com> wrote
Chester Pete
Yes you're right. Currently I have a spare set of boards for system 7 which
I am keeping for a rainy day.
The thing is, this problem is on a system11A (Fire).
If I had this "in-line" tester, it would be so much easier.
1) Fire! Machine has problem - like a lamp matrix problem.
2) Remove connectors for lamp matrix and plug in two (row and column)
dedicated "in-line tester" boxes. Each box has 8 LEDs (16 in total).
3) Go through the machine tests and see which pairs of LED lights come on.
If different from two, then there is a problem on the board!
5) Plug the original connector into the "in-line" tester box. In other
words, the machine is fully connected up again but with these tester boxes
inbetween the board and playfield.
6) Repeat tests.
7) when you hit a problem you can see which wires are needed to be checked -
without having to look at a manual!
4) initially thoughtneed to repair but thi.
"Dallas Overturf" <over...@fastdial.net> wrote in message
news:JWbdf.163$Kf3.117@trndny02...
beaver
improvements that are missing.
Edward Cheung
beaver
head. I thought you were addressing it to someone else and that is why
I did not originally reply. It seems to me that the CPU tester would
hit most of the issues you mentioned. Like lines stuck high or low.
Also, the playfield tester was meant to be able to hang off the
sidewall of the machine, so it is somewhat portable. The cables are
long enough to connect through the opening of the backglass, but of
course it does not fit in-line. I will have to think about that one.
Regarding your problem of the lamp being on faintly. I think the CPU
tester could replicate this condition and allow you to hunt down the
cause. The LED matrix just emulates the lamps on the PF. So ONE lamp
glows faintly when another particular lamp is on? That sounds tricky
(but fun) to diagnose. If you are in the DC area, I wouldn't mind
coming over to take a look.
Anyway, I will work intermittently on the CPU tester.
Edward Cheung
Chester Pete
it but I'm in the UK.
The advantage of an "inline" tester is that you do not need to remove the
CPU board - Conceptually, it is a device which is to test the "interface"
between the CPU board and the playfield. Really, I should have called these
devices "interface signal" testers (catchy huh!) but then again a
"playfield" tester is easier to say.
The idea is that the extra information given by these "interface" testers
help someone to "debug" the problem to the CPU board or to the playfield.
All this without having to have a manual.
In the end, the pinball owner is only looking for a situation where he/she
DOESN'T see a pair of LEDs. What could be simpler than that?
Good luck
"beaver" <e...@edcheung.com> wrote in message
news:1131818103.7...@g49g2000cwa.googlegroups.com...