So this I did yesterday and found no continuity. Please don't worry about asking basic stuff. Nothing is basic to me. I appreciate the help and am learning a tremendous amount from you guys. I did make sure the continuity was working on the meter before I started.

This I didn't do, but it is a good idea and will try it.

So to test the caps I was going to use resistance.

To test continuity, I was using the setting with the audible indicator as it was too difficult for me to touch the leads on the connector and look at the meter at the same time. If this could give me a false negative depending upon the orientation of the leads on the meter, then maybe I didn't do this right. But I would think putting it on the resistance setting would do the same thing. Perhaps I should run through the test twice, switching the orientation of the leads between the two rounds? What do you think?

Hey, winteriscoming, I took another look at the schematics, but I don't see the connection. The keypad looks like it is on a different matrix and goes to a different place on the board than the switches in the matrix I'm dealing with. Am I missing something? Thanks.

Ok, I have a feeling this might be important, but I don't know what I'm looking at. I found this other schematic in the manual. Note the upper right and how Bomb O and Rocket O and Bomb B and Rocket R are next to each other. Can anyone tell me what I'm looking at and does it shed any light?

Schematic switches.JPG

You're absolutely right. Mine too. I pressed 5 on the keypad, and it registered as rocket O. However, it registered as rocket O and only rocket O. I then pressed actual rocket O, and it registered rocket O and bomb O. I then pressed 5 on the key pad and rocket O at the same time, and it registered only rocket O. I also unplugged the key pad and pressed rocket O, but I sill had the same problem. For what its worth, here is the keypad schematic.

Schematic keypad.JPG

I just started going through and testing the caps, and I have a question before I continue. My multimeter is set to resistance. Some of the caps start low and quickly max out (go to infinity) the resistance reading on the multimeter. Some of the caps peak at a maximum resistance of 4 mega ohms. Should I consider 4 mega ohms as being sufficiently high as to indicate a functioning cap, or is there something wrong?

Do I need to get the leads of on the multimeter right on the legs of the cap, or is it sufficient if they touch the lugs that the caps are soldered to since with an open switch, the cap should be the only connection between the two points?

Big clue everyone!!!

I was doing the switch test again after checking some things, and I noticed that when I held the rocket "O" target only that target registered. I then moved to try something else, and rocket "O", bomb "O" and some other switches in that send row registered. I moved again and only rocket "O" registered. What could it be about my movement, I wondered. [Pause for suspenseful build up. . . .]

I noticed that every time more than one switch registered, I was leaning on the metal side rail of the pinball machine. Every time I took my arm off the metal side rail, only one switch (the proper switch) registered. I repeated this enough times to be sure.

So, I need your help again. Does this mean I have a short to ground? If so, how do I go about finding it? Or, does it mean something else? Perhaps I just need to play with rubber gloves from now on . . . .

No metal ramps or habitrails. I'll spend my next round of free time checking all of the switches for something along the lines of what you describe.

I did check all of the caps, and they were fine. But that may be a moot point now.

Hey guys, thanks for the continual follow-up.

I have more clues based on some tests I ran after reading your comments. First, I measured resistance between the cabinet side rail and the ground braid throughout the machine. In all cases, there was continuity, so the cabinet appears grounded. Then I measured resistance between various switches and the cabinet side rail. Again, in all cases continuity. That seems wrong. Finally, I removed CJ4 and measured resistance between the switches and the cabinet side rail. No continuity that time. I plugged CJ4 back into the board and measured resistance again between the switches and the cabinet side rail. Continuity again.

This seems bad to me. How do I interpret this? Are we back to this being a board issue?

There is connectivity between every pin for CJ4 on the board side and the ground braid for the cabinet. I also found the ground braid that goes to the board. It appears to be separate from the cabinet braid. I checked connectivity between the board ground and the cabinet ground and found connectivity.

I haven't removed the board yet. I can, but would like to know what I should be looking for. Please feel free to talk in very simplistic terms to me. Thanks!

There was definitely resistance. Not a dead short if that means no resistance.

I wasn't touching the leaf switch with my fingers, just the tip of the probe on my multimeter. The other probe was on the cabinet rail. I had resistance but connectivity with CJ4 plugged in. Infinite resistance, no connectivity, with CJ4 unplugged.

Next steps? I'm happy to take off the board but would like to know what to look out for. Thanks again everyone.

Hmmm. I think that might require me to have three hands . . . .

Oh, I see. I was pushing on the targets with my finger. The targets are plastic, but tomorrow I'll mess around some more and push them closed with a plastic rod to ensure there is no connection there.

Yep, they do have a rivet in the center. That was probably it. I'll report back after trying again.

Safe travels. Both you and Robert, as well as the others who have chimed in, have been great. My guess is that I'm still at this when you get back as I may only have a few minutes each night or every couple of nights to test things during the work week.

Robert, once I have a chance to run this test, I'll try to lay out a summary of what is happening and what I've tried.

So, I did some more testing tonight, and here is a summary of where we are:

**The problem lies with multiple switches registering in a single send row. It is never switches in a different send row. So, rocket "R" can cause it or rocket "O" can cause it, or some other switch in another row. The key is there needs to be a connection to the ground. When I was testing in switch test mode, my finger on the metal rivet and my arm on the rail created this connection. When I tested in game, me pushing the metal ball into the target with my hand on the rail also created the connection. If I push a switch without having my hand on the metal rail, or push a switch with a non-conductive plastic rod, then only that switch registers, no others. (Interestingly, the problem seems worse / more pronounced when I am testing in an actual game instead of switch test mode, but that could be my imagination.)

**If I unplug CJ4 and test resistance of the board pins to ground, I get around 50 k ohms.

**If I unplug CJ4 and test resistance of a switch lug to ground, there is no continuity.

**If I plug in CJ4 and test resistance of a switch lug to ground, I get continuity with a resistance of about 50 k ohms.

**I decided to try to play an actual game since it had been a while since I had done anything but tests. (This is where it gets embarrassing.) The game, I think, functioned normally. Even though my hands were on the rails near the flipper buttons, I didn't notice any weird switches going off. This is probably because my body wasn't making a direct connection between the switch and ground. The embarrassing part is that I can't remember the issue I was having that caused me to do the switch tests in the first place. I know it was something along the lines of a switch misfiring, but I can't remember exactly what.

I don't want to leave this alone, because it seems like everyone agrees that this is not normal behavior for a pin. So, I guess the question is where is the short -- on the board, on the playfield? What should I do to find it or to test further.

I hope this was a good summary. Let me know what else you would like to know.

Well, I'm really embarrassed if that is the case. So, measuring some connectivity between a single switch lug and ground is normal? I thought Terry said that was wrong. I'll play a bunch of normal games to see if any problems occur, or if the issue I experienced way back when reoccurs.

If everything is normal, I feel really guilty. That said, you guys have been great! If nothing else, I learned a tremendous amount from both of you and am now much more skilled with these things.

I'll report back after getting a bunch of games in to just let you know how this is going.

Thanks for being so understanding.

For my own edification, what exactly is happening? The current is coming in through the send row and down the receive column corresponding to the switch, so that the switch registers. The current is also coming down the send row, going through me, and into ground. Why does that then register some of the other switches in the send row? Somehow the current going through me is hitting some of the other receivers in the receive column?

I'm a very visual person, but I'm not sure I get it by looking at the matrix.

I've now played a few games and determined that the pin is playing fine. So whatever the issue was that made me check the switch matrix in the first place, it may have just been a weird one time thing. The pin is old, and sometimes things happen. So, I guess I was really creating the problem when I tested the matrix by closing the switches with my finger.

If I can indulge your patience for a bit longer, I think it would really help me and any other newbies following this thread understand what is going on. So, I'm trying to follow what you are saying above, Terry. I get the basic idea of the voltage at each column and row being kept high. The machine then pulses the columns low one by one. If a switch is closed, what do you mean by pull the row low? How does that physically take place, and what does the ground have to do with it? Also, I'm not really sure how a column can pull anything with a blocking diode in the way, but I'm clearly missing something. Perhaps "pull" is just a term of art.

Also, if I interject myself by touching the switch and ground what do you mean by pull the row low? Somehow connecting something held high to ground makes that row go low? I kind of get that, but I'm having trouble seeing how it then affects the whole column. Perhaps it is how the CPU is programmed to interpret things?

Anyway, if you have time (obviously no rush), it would be great to close this thread out with a bit more explanation. I read your webpage on the subject and found it very informative, but there is still some basic concept that I'm missing. I'm feeling dumb, and I hate that. Thanks again!

I think I know why I'm being a bit slow in getting what is going on. I was imagining the switch matrix as a simple tool where current is sent down a column one by one and the rows are read to see if any switches are closed that correspond to the column that is pulsed. While that simplistic idea is accurate in general terms, there appears to be a lot more going on. I just googled pull up resistors, for instance, and found a whole other layer of complexity behind this that I wasn't aware of. I'm making my way through some basic electronics texts but haven't gotten to that level yet. I'm a fairly scientific person, but also need to imagine what is going on (like visualize the electrons traveling along wires, etc.). I haven't been able to get my head around how the ground, pull up resistor and switch matrix all work together yet. I'll keep working on it and studying up. Once I think I get it, I'll post a simplified explanation and let you guys tell me if I got it right. In the meantime, thanks again for all of the feedback!