Lamp matrix test board design input needed

HHaase

Active member
Joined
Mar 12, 2010
Messages
1,384
Reaction score
10
Location
Sioux Falls, South Dakota
Hey guys;

Ok, here's one of the reasons I shelved the Bally Home boards... too many people were asking me to follow up on my previous diagnostic tester. The #1 request I'm getting is for a lamp matrix tester.

Well, I'm proceeding with it.

( For those not familiar with my solenoid tester, here's my info link for it. KLOV folks can PM me for lower pricing on these. http://www.siegecraft.us/tester.html )

Anyway, the lamp matrix tester will be fairly simple too. Basically the design will just be an 8x8 LED matrix on a small board that you plug into your CPU board for bench testing. Looking for some feedback from folks again, as I got some great ideas back when I did my solenoid tester, which ended up a lot more versatile due to input from you guys.

I'm looking at two ways of packaging it, either as an un-assembled kit for around $15 or a fully assembled board for around $35. This time around, I want to make it a bit more refined of an end product than just selling a bare circuit board with a URL to instructions.

What I really need most for feedback is how you guys can use this for boards other than what I am familiar with. I'm focusing these mostly on Williams boardsets, but want to make it useful for other manufacturers that use an 8x8 matrix for their controlled lamps. I know flat out that these will NOT work for the old Atari machines, they use a totally different concept on their controlled lamps.

-Hans

(Oh, and before anybody asks... I don't do pre-orders. I'll do a list of folks who are interested, but I don't take any payments until I'm tested and ready to ship.)
 
Last edited:
I'd definitely be interested, both in the kit PCB you have for the solenoid tester, and in the new project. :)

Don't really know what else I could use it for, though.
 
Here's a question to throw out there.

On my solenoid tester, I had it layed out for .156 molex headers. But they, of course, are pretty big. The resistor layout on that one, it didn't make a huge different, but on this one I can really shrink down the board about 20% or so. Measuring out at under 2"x2", which is great, half the size of my solenoid testers. I hate things taking up too much space on my bench and in my toolbox.

Would you guys object significantly if I put .100 headers on this board instead of .156's?

-Hans
 
Last edited:
If you're going to design it so you need a harness to connect it to anything you might want to include the connectors when you sell it. A lot of people are not going to have .100" or .156" pins.

Other than that it's a bunch of LEDs in a matrix on a PCB. Not a lot to consider really. Personally if I suspected a lamp matrix problem I would take a more direct approach to troubleshooting than connecting a bunch of LEDs to the outputs. Especially if I have to source a bunch of parts, assemble a PCB and build a wiring harness to do it. But people love blinky lights so I'm sure people will buy it. Interesting project all the same.

Question about the solenoid tester. I don't see any load resistors. Without those how is it any more useful than a logic probe? Slightly more convenient sure but not a true test of what the transistor is going to drive in the game. I would consider modifying the design with some load resistors otherwise it's not very effective for testing solenoid driver circuits.

Just my $.02

EDIT:
I thought of something else. You're going to need a resistor in series with each LED. I would use SIP resistor arrays rather than individual resistors, if you have not already considered that.
 
Last edited:
Yes, I'll definitely be including all the connectors and wiring needed for this one. I'll be selling matrix tester as an all-inclusive kit. I'm also going to be re-packaging the I/O tester as a complete kit as well, now that I've gotten enough feedback on how people are using them. I was pleasantly surprised at some of the variations that I've seen.

It's intended more for initial evaluations, but I found an LED matrix like this pretty useful when I started using Ed Cheung's all-in-one tester board for bench testing. It makes it easy to spot locked on or dead rows on a board, which you then dig into more in depth. But I hate the spaghetti pile when using Ed's board, since you're running ribbon cables back and forth all over the place. That, and the smaller individual testers let me use them when the boards are still in a machine.

I didn't put any load resistors into the solenoid tester mainly because people asked me not to, for data bus work (such as on Zaccharia) in addition to output circuits. I do have the individual resistors on there instead of SIP resistors mainly to make it more adaptable. For instance, using half of it as switches makes a quickie special solenoid tester for both the inputs and outputs.

You're right in both cases a logic probe can do the same job, but these do speed things up quite a lot.
 
Last edited:
It's intended more for initial evaluations, but I found an LED matrix like this pretty useful when I started using Ed Cheung's all-in-one tester board for bench testing. It makes it easy to spot locked on or dead rows on a board, which you then dig into more in depth. But I hate the spaghetti pile when using Ed's board, since you're running ribbon cables back and forth all over the place. That, and the smaller individual testers let me use them when the boards are still in a machine.

Definitely handy for a full blown test bench. It wouldn't be my first choice for problems in the machine but still a cool option. I can see it being handy to just drop in there and get an overview in lots of cases. I've built a few boards like this for that very purpose. I did one for the switch matrix as well. I don't intend to sell these. That might be something to consider next. A switch matrix board. I did mine with SMD diodes to make it nice and compact but for the switch matrix you don't even need the diodes as long as you're only pressing one switch at a time. The diodes were a PAIN to solder by hand.

I didn't put any load resistors into the solenoid tester mainly because people asked me not to, for data bus work (such as on Zaccharia) in addition to output circuits.

Definitely would not be my first choice to troubleshoot address or data bus problems but I see where you're coming from. Still limits the usefulness for troubleshooting solenoid driver circuits. You could use an external load but suddenly the convenience factor is gone.

I do have the individual resistors on there instead of SIP resistors mainly to make it more adaptable. For instance, using half of it as switches makes a quickie special solenoid tester for both the inputs and outputs.

That makes sense for the 20x LED board but not for an LED matrix (IMO). 64 resistors are going to take up a ton of board space and be a real pain to deal with. The options for use are far less with a matrix. I don't see people using it for anything other than an LED matrix where the 20x LED board is broken out into individual circuits leaving all kinds of options for use. It's actually ideally suited for testing stuff other than solenoid driver circuits.

You're right in both cases a logic probe can do the same job, but these do speed things up quite a lot.

If the price is right a small time savings is worth it. It's really just a matter of how to best use it so you're actually saving time. Ultimately that's up to the individual.

EDIT: I had an idea when I was originally designing my LED boards for an LED board using bi-color LEDs geared more toward testing of address and data bus lines. Basically just a crapload of logic probes on a common board. That's another idea...

EDIT2:
You could also consider using "resistor LEDs" with built in resistors but that would somewhat limit the usability.
 
Last edited:
That makes sense for the 20x LED board but not for an LED matrix (IMO). 64 resistors are going to take up a ton of board space and be a real pain to deal with. The options for use are far less with a matrix. I don't see people using it for anything other than an LED matrix where the 20x LED board is broken out into individual circuits leaving all kinds of options for use. It's actually ideally suited for testing stuff other than solenoid driver circuits.

LOL got myself confused there. In the 20x LED board I had a resistor per LED, but for the matrix you only need 8, one per row in this case. I thought about using SIP's, but the size difference really wouldn't make much change to the board cost, maybe $0.10 to $0.15 per board is what it worked out to. Discreet resistors are pretty cheap already, dissipate heat better, and are easier to source and replace if the end users need to.

After I get done with this, I'm considering a switch matrix input board too. Still toying with a few different concepts, haven't nailed it down yet.

-Hans
 
LOL got myself confused there. In the 20x LED board I had a resistor per LED, but for the matrix you only need 8, one per row in this case. I thought about using SIP's, but the size difference really wouldn't make much change to the board cost, maybe $0.10 to $0.15 per board is what it worked out to. Discreet resistors are pretty cheap already, dissipate heat better, and are easier to source and replace if the end users need to.

You could do that and minimize the resistors. There are pros/cons to either. I wouldn't use SIP in that case either.

After I get done with this, I'm considering a switch matrix input board too. Still toying with a few different concepts, haven't nailed it down yet.

A bunch of switches in a matrix. Not complicated :p

The one I designed is specific to Bally/Stern and connects directly to the MPU. It didn't take more than a couple hours to design. Assembling is kind of a pain though. Especially if you use diodes. These I might actually sell at some point but who knows when that will be. Too many projects right now.
 
A bunch of switches in a matrix. Not complicated :p

Until you try to solder in 64 switches and 64 diodes :) (and have to pay for them!)

I'm more thinking about if I want to have tactile pushbuttons, dip switches and such. Right now I'm leaning toward 8 pushbuttons for the rows and an 8-dip block for the columns.

-Hans
 
Until you try to solder in 64 switches and 64 diodes :) (and have to pay for them!)

This is an ideal use for a solder pot if you were doing large scale production. I've already assembled several of them. I know exactly how much of a pain in the ass it is assembling them by hand. Still an extremely basic design. You should be able to build them using all tactile buttons in prototype quantities for $10-$15 each fairly easily. This is the kind of thing that might be more suited to a "kit" and leave the soldering to the buyer.

Right now I'm leaning toward 8 pushbuttons for the rows and an 8-dip block for the columns.

The buttons are not that expensive if you buy them out of China on ebay. I wouldn't go with DIPs personally. The trade off in convenience for a minor savings isn't worth it (IMO). Also limits flexibility. I would probably use an alligator clip lead before something like that.
 
This is an ideal use for a solder pot if you were doing large scale production. I've already assembled several of them. I know exactly how much of a pain in the ass it is assembling them by hand. Still an extremely basic design. You should be able to build them using all tactile buttons in prototype quantities for $10-$15 each fairly easily. This is the kind of thing that might be more suited to a "kit" and leave the soldering to the buyer.

The buttons are not that expensive if you buy them out of China on ebay. I wouldn't go with DIPs personally. The trade off in convenience for a minor savings isn't worth it (IMO). Also limits flexibility. I would probably use an alligator clip lead before something like that.

And posts like this one are exactly the reason I ask for feedback, I'll look into a layout for a full 8x8 matrix and see what kind of pricing works out on it.

If the market was big enough, this is just the sort of thing that an SMD design would be perfect for, but there's no way would I sell enough to justify outsourcing the assembly work. If I sell more than 100 I'd be shocked.

-Hans
 
Adding the whole thing together, looks like a full 8x8 switch matrix board, including board, buttons, diodes, connectors and cables in kit form.... would sell at around $25 per kit. That's assuming 6mm square tactile pushbuttons.

Board size would end up at 3.5" x 3.8", and a lot of soldering. 130 components on there, total of 400 solder joints. The buttons will snap in place though, which will help a lot.

I wouldn't go any smaller than the 6mm square buttons, the 6mmx3mm size would make it hard to manipulate without using a stylus.

-Hans
 
Oh, one more question.....

Wiring.

6" long harness enough for most of you guys? or would you want enough wire for a foot long harness?

-Hans
 
Yes, I'll definitely be including all the connectors and wiring needed for this one. I'll be selling matrix tester as an all-inclusive kit. I'm also going to be re-packaging the I/O tester as a complete kit as well, now that I've gotten enough feedback on how people are using them. I was pleasantly surprised at some of the variations that I've seen.

It's intended more for initial evaluations, but I found an LED matrix like this pretty useful when I started using Ed Cheung's all-in-one tester board for bench testing. It makes it easy to spot locked on or dead rows on a board, which you then dig into more in depth. But I hate the spaghetti pile when using Ed's board, since you're running ribbon cables back and forth all over the place. That, and the smaller individual testers let me use them when the boards are still in a machine.

I didn't put any load resistors into the solenoid tester mainly because people asked me not to, for data bus work (such as on Zaccharia) in addition to output circuits. I do have the individual resistors on there instead of SIP resistors mainly to make it more adaptable. For instance, using half of it as switches makes a quickie special solenoid tester for both the inputs and outputs.

You're right in both cases a logic probe can do the same job, but these do speed things up quite a lot.

This stuff sounds great. Please let me know when you have the kits available. For those reasonable prices I'll get one of each one you offer.
 
Just waiting on sample LED matrix blocks, then I can get to prototyping the lamp matrix board.

Finished up a parts list for the solenoid tester kits, will be ordering parts in a couple days to re-package them as complete kits. Price will be $15 per kit. I'll have 21 of the double LED style, and 10 of the special solenoid hybrid to sell (Not bad, originally did a run of 100 boards)

So I figure in about two weeks I'll have the solenoid tester kits ready to sell, and hopefully the LED matrix will be ready by end of November, maybe sooner.

-Hans
 
Working on the pre-order list for the matrix tester kit, switch matrix kit and the solenoid tester kits now. Solenoid testers are in short supply, only have 31 of them left. Will be packing 21 kits of regular output 20-led testers, and 10 kits of "special" solenoid testers.

Quses and Frax, have you on the list for one each (lamp matrix, regular and special solenoids, switch matrix).

If you want on the list, shoot me a PM. No money needed at all yet, just working on a list at the moment.

Still taking suggestions for tweaks, changes, ideas and suggestions.

-Hans
 
Still waiting on the LED modules, so no prototype work yet on the lamp matrix tester.

The solenoid kits should be ready to go in about a week or so, everything is already en-route. Felt odd ordering things in quantities of 500 again, I haven't done that since I ran my paintball shop.

I've also got another small board project in the works, while I was waiting on things to arrive. A small daughter board for Firepower. Puts all the EPROM and PROM's onto a single 2764 EPRPOM, and fits under the CPU chip. No modification necessary to the board, plugs right in.

I just need to confirm the layout and data sequence before I send off an order for the prototypes. This one is too complex to go straight to production without testing. If I can fit it in there, I'm also going to try and put LED indicators for the +5v and Reset lines too.

-Hans
 
Back
Top Bottom