As you can tell by my game collection, I love GamePlan
I brought Global Warfare to the Louisville Show last year for Roger Sharpe to play, and it went crazy when I turned it on. It is perfectly fine now (and will be at the Louisville show this weekend, and I will be bringing to the Chicago Expo this year for hopefully Roger to play it!).

What ended up being the problem is nuts. A staple that held the GI lighting wire actually went through and touched the post that connects to the spinner! When the spinner spun, there was enough conduction to send some AC goodness through the switch matrix! Good thing the MPU uses some 339's for processing the switch matrix I guess.
Anyway, after hours of trying to figure out the problem, it was the easiest fix. One pair of pliers to remove one staple.

Yeah I do the exact same thing as Robotworkshop does. I use a 4 AA holder, and solder a diode in the fourth position ( to make it a 3 cell).

Diode type is not much of a big deal since current draw is low. I always use a 1n4001 or whatever I have (same as a clamping diode on a solenoid).

Just take the battery off, and have the cathode (stripe side) of the diode to the "battery +" terminal on the mpu
. Negative side of the battery box to the - side.

PXL_20210415_002621147 (resized).jpg

You are correct. The Bally Grey legs are the closest. GamePlan had the single rib on the edge. Maybe a slightly different color grey. Tbh, I have been buying the pinball life black Williams legs. Yeah they are dual ribbed (uggghhh) , but I think they look way better. They are 28 1/2 as well.

https://www.pinballlife.com/bally-gray-legs-set-of-4.html

Link to the black Williams legs too:
https://www.pinballlife.com/williamsbally-black-legs-set-of-4.html

Lol! That is close !!!

Oh nice ! Totally agree with that !

Yeah GamePlan uses a 74154 for driving the switch matrix, so it is "Active Low" for the strobes. Basically when the strobe is active, it will be close to 0 volts, and the rest will be at 5 volts. I know other mfrs used an Active High instead (I think old Bally SS with a 6821).

Schematic is correct for the direction of the diodes. The cathode should be on the strobe wire, and the anode is toward the switch/return. The diode doesn't bring the return to 0 volts of course due to the drop (0.6v), but it is fed into a comparator to compensate and drive the 8255.

Hope that helps.

Greetings! I had a drop target break on my Sharpshooter the other night, and I am getting a little low on spares. I went ahead and sketched up a drop target and 3D printed last night. I just got it installed during lunch and tried it out a couple times and it is working perfect. I now have the fun to test this some more!

I marked it as a WIP, as I used a Gottlieb target as a starting point, and I think I am going to open up the back of the Spring loop, so that it is easier to connect the spring when it is in place (like the original had).

I ordered a spool of red as well, which once that comes in, I am going to print some spares.

Anyway, thought I would share a link to the thingiverse site for it: https://www.thingiverse.com/thing:5151387

pic1 (resized).jpg

pic2 (resized).jpg

pic5 (resized).jpg

It wasn't long after testing, I broke yet another of the original drop targets

To preserve what I had left, and to play without worrying about any more breaking, I went ahead and printed out two complete sets.
After the couple of minor revisions, I uploaded the latest drop target STL. So far they have been working great!

Same link as before: https://www.thingiverse.com/thing:5151387

PXL_20211219_183554097.MP (resized).jpg

Yep. I have Qidi Tech 1, which is a FDM. I have only made these using ABS fwiw (Qidi is completely enclosed)

It looks like it just may be a new switch w/some coating and a bit of a gap on the leaf switch. Red arrows added to your pic to what I am talking about. I would first try taking a non-glossy business card or 1000+ grit sandpaper, sandwich between the switch, and clean off to get the contacts flush. Then adjust the switch so that it is tight when closed and try that.

Wiring-wise, this looks correct. I don't have my Cyclopes set up currently (I rotate the lineup), but can do that Sunday evening if you need some pics.

airgap (resized).jpg

Ok awesome. So I think I see your problem. I took a pic of the schematic from Captain Hook before leaving for out of town today ...

Your green wires on the lower flipper are reversed.

The green wire from the anode side of the diode should go to the shared contacts of the leaf switches.

If these are backwards, I can see the power for the top flipper running through the hold coil of the bottom..which would make it act weak.

I can still set up Cyclopes Sunday evening and send pics if you like as well.
-Jeremy

PXL_20220107_153344371~2 (resized).jpg

Awesome on the flipper!

Cyclopes uses 3 2732s (12k total), so a SharpShooter board would not work since that is typically strapped for 2716s. You are missing a rom though.

There are at least two revisions out there, but the zip file on ipdb has two of the room revisions in it.

I have an echo lake board in another machine (I think Andromeda but can't remember lol). I can create a rom for Cyclopes and try it out in mine, and then send you a rom after testing it.

Completely agree...just saying that Cyclopes uses 3 2732s vs 3 2716s in case if one was just going off of rom count

Do you happen to know what exactly doesn't work right with the Echo Lake board in Cyclopes?

I went ahead and burnt a rom (with the Dec 85 version) for the Echo Lake board I have and put in the game. Booted up fine and played a game. I will try it out this week, but didn't know if there was something specific that didn't work.

PXL_20220110_024034831 (resized).jpg

PXL_20220110_033159375 (resized).jpg

Thanks I will do a test with the glass off to see if I have the same issue.

Yeah I was thinking along the same lines. I have a Fluke 90 /z80 tester ,similar to the 9010a w/z80 pod, that I can do the comparison to the orig mpu-2 and now the echo lake w/Cyclopes rom.

I wasn't sure how the echo lake board was handling the additional address space of a 27128 that is uses, so I doubled the image by copying itself twice. Still doesn't use the last 8k of space, but thought any relative jumps could beyond the 2FFF program address space.

I will test later and report back!

I can't stress how much time the Fluke 90 has saved working on mpus, especially when it comes to finding a bad trace in the data bus, or diagnosing which ram/rom is bad!

So, the Echo Lake board inside Cyclopes works as expected to me. I confirmed:
1. Spot targets work when drop targets are down
2. Captive ball always registers
3. Drop targets reset as expected, saucer kickouts work, multiball etc.

The things I can think of that it could be on your game not working:
1. You do have an issue with your Echo Lake board
2. You need a new ROM
3. You have a playfield "issue":
3a. Fish paper behind drop targets missing and maybe shorting things out when they drop somehow? (pic attached)
3b. Missing capacitors on the standup targets (so they are not registering all of the time)
3c. Diode on the switch matrix is shorted that shares the targets (seems unlikely since that would be each diode on the targets)

3a. and 3b. would be the easiest to check. To help, I attached pics of the areas of my playfield.
For #2, if you PM me your address, I can send you a copy of the ROM I have in the Echo Lake board for Cyclopes.

So I didn't end up using the Fluke 90, since it seems to be working. I did take a pic of the MPU-2 that was in it from the bottom. It has some jumpers on it that look to be from the factory (consistent, doesn't look reflowed, same gauge / color -age as the 7th digit jumper). I will document the cut&straps for 3 2732s. The only schematic I have is for the MPU2-rev2 (pic attached, does have the reset circuit though), which most of the ones out there are the Rev1's.

left-drop-targets (resized).jpg

left-target-bank (resized).jpg

right-drop-targets (resized).jpg

right-target-bank (resized).jpg

mpu2-rev2-schematic (resized).jpg

mpu2-rev2-three2732s (resized).jpg

Not the drop targets, but the Standup targets that has the caps added. Actually all of the standup targets should have them (I think ) on this game.

Lol, yeah I agree! The important part of having it labeled "17" for a 7417 was helpful when I saw that the first time. After working on GP mpus for so long, these are definitely the most comfortable boards I work on

Hmm. So yeah my Cyclopes game works correctly. You probably know, but just for documenting:

1.As each drop target is hit, the arrow insert goes off.
2. Once all drop targets are down in the 4 bank, the inserts flash
3. As you hit each spot target behind the down drop targets, the arrow insert stops flashing
4. Once all spot targets are hit, the 4 bank drop targets reset

I will see if I have a proper switch matrix schematic. I should, but I haven't looked in awhile.

You are correct on the switch test mode in how it works. GamePlan only shows the "lowest number" switch on the display, and doesn't cycle through like other manufacturers.

With the drop targets up, and the balls out, can you confirm that the switches are the correct numbers?

Starting with the bottom left spot "M" target I get:
M = 33
O = 34
N = 35
S = 36

Spot targets on the top right I get:
T = 37
E = 38
R = 39
S = 40

Drops I have:
C = 17
Y = 18
C = 19
L = 20

Top right:
O = 21
P = 22
E = 23
S = 24

Good luck!

I really want to create your issue on my machine to figure this out.

Can you take a pic of your echo lake board, a couple close shots?

Maybe there are some differences in how they are stuffed, using straight up TTL vs LS, causing a timing issue or something.

sethbenjamin , can you check what MPU you have on your Echo Lake board?

I looked at mine, and I have a Z0840004PSC (pic below)

Looking at an archived Echo Lake board that was for sale, it has the Z84C0006PEC:
https://pinside.com/pinball/market/classifieds/archive/102793

Typically this would be fine, but The "C" after the 84* part number means that is a CMOS vs NMOS chip, which does have different timing characteristics (I checked the the datasheet here: http://www.zilog.com/docs/z80/ps0178.pdf, page 31).

I don't have any CMOS Z80s, but have at least 30 of the good ol' Z80, and Z80As lol, so I can't test if this makes a difference. I went ahead and ordered a Z84C0008PEG from Digi Key, that I plan to socket my Echo Lake board and swap in just to see if I can recreate.

Like I said, my Cyclopes works fine...heck with the old MPU-2 board too. I just hate to see another Cyclopes down

PXL_20220120_014335129 (resized).jpg

Dang it, so that is not it either. I will compare my board to your pic. It might just make things either to let you try my board in your game as well, just so we can eliminate if the prob is on board, or elsewhere in the game.

I can take a video of my game w/board too prior to sending, to show how it is working in mine too.

***PSA on getting an MPU-2 or Echo Lake board working in Cyclopes***

sethbenjamin and I have been troubleshooting the Echo Lake MPU board in Cyclopes, sending boards back and forth, videos of what it is doing now etc.

There is a change that was made to the Cyclopes board with 2 resistor values that are different from other MPU-2 Rev2 boards. The 6.8k pull-up for the comparators is 470 ohms, and the 10k pull-down resistor resistor is 1k. This works out for the reference voltage to be ideally 3.06 volts vs 2.97 volts. Being just a little higher of a reference voltage, I think this is safe to do for all games, and should not cause any errors in switch registrations.

Changing out the two resistors on the Echo Lake board, allowed for the drop target banks to work exactly how they behaved with the original MPU-2 installed.

If you do have a Cyclopes and the drop target banks are not working correctly, I would double check your MPU board for these values as well. The other MPU-2 boards I have, do have the values as labeled in the schematic.

For the Echo Lake board change the following:
1. Resistor circled in Blue (immediately below U23, top of the 100k resistor row) from 6.8k to 470 ohms
2. Resistor circled in Green (immediately below U22) from 10k to 1k ohms

PXL_20220217_185124138~2 (resized).jpg

PXL_20220217_185301392~2 (resized).jpg

I am going to start a new thread once I get some more testing done, but I have a multi MPU design now and will open source the files when complete for people to make their own if they like. I am not going to build and sell them myself, but do want to contribute back to the community. I have a working proto board now in my Cyclopes. I am going to test it in Andromeda, SharpShooter, and Global Warfare before making the repo public. It should be this weekend.

Quick video of it working here:

More info on the progress at: http://www.epinball.com

-Jeremy

Yeah it is just a lot of soldering, but not too bad

Other than the parts, you will obv need to burn the eprom as well. The image will be uploaded/shared too.

No clue what forum category choose for open source hardware, but here is the link to the forum for the multi MPU board:
https://pinside.com/pinball/forum/topic/build-your-own-gameplan-multi-mpu-board

GamePlan uses AMP Mate-N-Lok connectors for the connection between the head and the cabinet.

I have ordered replacement pins from Digi-Key myself.

Example for the male.pin:
https://www.digikey.com/en/products/detail/te-connectivity-amp-connectors/60620-1/15674

The 12 pos housing is 1-480278-0.
Mouser should carry as well.

Hope this helps.

That would be a 3amp slow blow fuse for playfield.

Yep! That would be me
I created a 3d sketch of a Gameplan drop target. I actually have a few of these in my games and work well. I have had to use an xacto to knife to ensure the slots were pretty clean. Here is the link:

https://www.thingiverse.com/thing:5151387