Lee ... That's the same ebay store I bought mine from .. But at a better price, Seems this board comes from a laser CNC machine, Because one of the other boards from ultratech is listed as a 5 axis laser focus board..

Tony ... I heard Williams arcade uses allot of 6502's in their boards.. Do you know anything about them... Seems allot of their sound boards ver 6 & 7 and 11 & 12 end up on ebay and look interesting as proto boards, especially the ones with 32 pin memory sockets.. But the Photo's are to fuzzy to see what chips are on them...

_________________
When falling from a high place, You might as well try to fly

Williams never used the 6502 series. They preferred the 6909 for their video and pinball work.

Atari video and pinball used 6502s. their later 16 bit 68000s still used 6502 sound boards.

gottlieb pinball system 1 was a Rockwwell 4 bit. Their system 80 was 6502 and the system 3 was 65C02.

Bally loved the 6802 for their pinballs.




as an aside, I used to work in the arcade industry for Romstar and SNK. I am presently designing a pinball hardware and software using the 65C02.

_________________
"My biggest dream in life? Building black plywood Habitrails"

Tony .. Nice versatile driver board, I will keep it in mind, But suffer's the problem I'm trying to get away from, Using NPN driver transistors with the ULN2803's or any open collector pre-driver requires they be turned on to turn off the tip122's.. It seems you solved that by putting the 74HCT595's in tri-state with pull-up resistors to turn the transistors off... Just noticed the Opto to serial then back to the parallel latches .. Pretty cool..

_________________
When falling from a high place, You might as well try to fly

it isnt a solve for that. the board allows for a stuffing of either ULN2803s OR TIP 120s. You don't use both. the 595s have latching outputs to allow the off or on state. The pullups are for the tip120s only. The Tip120 circuit is based on pinball solonoid work.

The opto to serial is its own circuit. There are two completely isolated serial chainse, one to output to the driver transistors or uln2803s. The other one inputs from the opto isolators.

The only really cute gig with this board allows jumpering to either 12 volts or 24 volts for the pullups for the optos or the diode pullups for the output side.

My only regret with the board, it doesnt handle super much current, I made the power traces at .50 instead of larger.

It was designed for Halloween for running an interactive haunt game setup, so it is is pretty generic.

If you want, I will post the schematic to it tonight in my folder for you to dissect some more.

_________________
"My biggest dream in life? Building black plywood Habitrails"

Tony .. Yea I'd like to scope out the schematic..

Got the Ultratech board already, And it is not quite what it appeared to be.. It in fact has tip122's not 127's, Each emitter has two 2w 16 ohm resistors in parallel equaling 8 ohms to ground which charges a 220uf 50v cap also on the emitter, Providing a current surge at turn on, But then current limiting with the 8 ohm resistance, And also causing emitter degeneration and reducing base drive which may be a problem.. Each base is driven from a MC14050 non-inverting hex buffer capable of 45ma per buffer????, Through a 220 ohm resistor to the base of the 122's, Beside the triple row connectors on the back, There are 2 sip connectors one for input & one for outputs.. And although the board is hackable, The big negative is, It also has .050 traces limiting current capability.. Plus it was listed as a stepper driver, It is etched on the board as a solenoid driver board, The cap providing the pull in current and then the resistors providing hold current.. The board is designed for 24vdc and has a LM340T5 regulator for the IC's

_________________
When falling from a high place, You might as well try to fly

http://www.nightmarepark.com/6502/Fear%20IO.pdf

K, let me know what you think

_________________
"My biggest dream in life? Building black plywood Habitrails"

Tony .. I know when you designed it you had a use in mind and it wasn't pumping out 48 amps, So don't take my observation to seriously..

The circuit is nice and versatile.. Only things I would change on the layout is to put the TIP120's next to the output pins, Or move the output pins inboard next to the transistor, To keeping high current path as short as possible, Then you could have used wide traces without weaving small traces between them... It could use a few more caps for the drivers... 24 tip120's at 5 amps each is a 120 amps.. De-rating it to 2 amps for use without a heatsink is still 48 amps, And all 2803's turned on de-rated to 400ma ea output would be 9.6 amps CORRECTION ! (175ma = 4.2 amps) , That makes for a bit of switching noise..

Of course this would be a worst case use of the board, But gives the board the most versatility when worst case scenarios are taken into account.. The power connector should be beefed up by adding 3 or 4 more ground pins or add a large spade connector for GND, Being all the current has to flow on the negative connector.. It would be best to have two separate grounds one high current for drivers and another for all the rest of the circuit... Thus avoiding the GND current from modulating the voltage on the rest of the board or trying to sink current on signal grounds..

_________________
When falling from a high place, You might as well try to fly

It was originally designed for running the LED clocks. I went more with the TIP120s to run single air valves as a just in case. Was never designed to run everything at once. Definitely can do those if I make a second run of these.

the 2803s will be running low, bout 45 mA per output I don't think it is possible to get 400 mA out of each 2803 driver as I have heard from some others. I think maximum is really 80 ma. Each one will drive 3 LEDs for the clock. I also figure 1 amp for brighter LEDs to use the TIP120s.

I am still not happy with my mistake of using only .50s for the power leads now on thinking about it still. Thanks on the observation ideas. Sadly, the weave to the 2803s and resistors is still needed though. My thing is on the time constraint since these have to be up and running for the 2nd week of October. Right now, all 6 boards I need for this are stuffed for the 2803 outputs, the inputs arent stuffed. For some other room ones, they will be stuffed as needed and I will add heavier wire jumpers to make up for my trace width mistake.

_________________
"My biggest dream in life? Building black plywood Habitrails"

Regarding trace width: A PCB layout book I have here has a graph of cross-sectional area of a trace versus current-carrying capability at 20°C and 30°C temperature rise, and then gives this example:

Obviously that's a constant current. It could carry far more in pulses that are too short to give the trace time to heat up, pulses that have plenty of cooling time between them.

To find shorts in inner layers of a multilayer board, I made myself a gizmo that puts pulses of a couple of amps through traces smaller than that. (.006 is common, so yes, I'm putting two amps through a six-mil trace.) It does about seven pulses per second with less than 1% duty cycle, and it puts it into the net that's shorted to ground (or other plane), and then I use the oscilloscope on the 5mV/div setting to find where the voltage drop disappears, since that's where the short will be. Since the pulses are so short, the trace doesn't have time to heat up. The voltage drops I'm seeing a lot of are in the 20mV range. I might change it to further increase the current.

Sorry I had allot of driver datasheets open, And was comparing them, I must have been looking at the wrong datasheet when I quoted 400ma,, The 2803's are only capable of 175ma all drivers on..

But anyways you have to remember the collectors are the output pins and also where the heat goes.. To run them near their rated output, They need sufficient traces to dump their heat into, Which is rarely the case.. Or use stick-on dip heatsinks... But by putting drivers next to the output pins helps,, With a large enough trace to the connector, The pins and the wires leaving the board actually help sink the heat..

_________________
When falling from a high place, You might as well try to fly

http://www.circuitcalculator.com/wordpr ... alculator/


http://www.frontdoor.biz/HowToPCB/HowTo ... Space.html

http://home.comcast.net/~pcb.george/trace.html

I need to dig into these pups sometime...

_________________
"My biggest dream in life? Building black plywood Habitrails"

Handy PCB reference... Haven't designed to many PCB's in my life (mostly wire wrap or point to point vector-board), But repaired several thousand that smoked... But once my PCB CNC is finished I'll be using them for ref...

You only get two design methodologies .. Overkill, Or do the math... Its a shame that positive side switching didn't catch on.. Its main benefit is you can tie all the collectors to a big power buss plane to dump heat.. But being TTL is all negative ground based... Almost all driver chips are negative side switching.. Open collector instead of Emitter follower...

_________________
When falling from a high place, You might as well try to fly

Got the Star MFG Board, It's a Vending machine board, It has is a R65C02P2 2 mhz, Uses PAL 16L8 address decoder, 32k Eprom 2k Ram and Two 68B21 PIA's, I/O Two PIA's and four buss drivers and one transciever .. 14 OC outputs,, Two 74LS138 Decode Addressing for off board locations

l VALUE CARD SHOP VERSION COPYRIGHT 1993 BY STAR MFG., INC. THANK YOU $0.00 VENDING EMPTY PLEASE CHOOSE ANOTHER SELECTION EMPTY PLEASE CHOOSE ANOTHER SELECTION OUT OF SERVICE CARD VENDOR SPORT CARDS... THE GOOD STUFF! UNSEARCHED... BEST BRANDS... THE ROOKIE CARDS AND SUPERSTAR CARDS ARE IN THIS MACHINE. SPECIAL COLLECTOR CARDS INCLUDED RESET CT? SAVED BUTTONS MT12 COIN/BILL

I dumped the Eprom to a BIN file

Lee ... Tony ... I Sent you a copy

_________________
When falling from a high place, You might as well try to fly

Disassembled and, as I can't access my website and its mirror has gone AWOL, put on my FTP server in pub as Vending.zip

I've done next to no commenting and there's possibly address bytes still to resolve but it assembles correctly.

Lee.

Thanks Lee ... That PAL makes it impossible to figure out the memory map with a multimeter and truth tables... I imagine its not quite as interesting to disassemble as the robot code was.. Best I can figure it had a bill changer and some kind of alpha numeric display.. I think the 24C02 serial flash is used to store any coinage inserted in case of power failure, And/Or keep a record of sales...

I figure to make it a robot board clone, I can make a daughter board with a 40 pin wire-wrap socket going through the daughter board into one of the PIA sockets and put the PIA in the wire-wrap, And then socket 2 ACIA;s and a VIA next to it and connect data and control... Then with a few extra address leads and some decoding I can map them into the I/O,, It has enough buss drivers and I/O on the main board for driving the stepper motors with minimal circuit hacking.. And I can either change the main crystal to 3.6864MHz, Or leave it at 4mhz and use a 9.8304 crystal and a 74HCT4040 to generate baud rates..

_________________
When falling from a high place, You might as well try to fly