Millipede - voltages needed to get a picture...

modessitt

Volunteer: Encyclopedia Submission Moderator
Joined
Mar 22, 2007
Messages
33,414
Reaction score
325
Location
Austin, Texas
I need to test a Millipede board for someone. I can build a quick JAMMA adapter for it, but I don't need it to save high scores, use a trackball, play sound, etc. I just want to see if it boots and gives me a picture on the screen.

Looking at the pinout, I can supply the +/-5 pretty easily. I'm thinking the +/-22 is for the high score save, right? It's listed in 4 different places. Also, the 10.3 - is it for the sound? Do I need to supply it to get the board to boot?

I'm hoping I can just hook up the +5 and grounds, video wires, switch inputs and maybe the optional joystick controls - and get it to boot (if good) and give me a picture where I can test gameplay.

Comments?
 
I seem to recall that there was a ~+15V used to power the last stage video amps. Quick look in the schematic should be able to tell you.
 
I'm hoping I can just hook up the +5 and grounds, video wires, switch inputs and maybe the optional joystick controls - and get it to boot (if good) and give me a picture where I can test gameplay.

Comments?

You need more than 5V to get video on most millipede boards...
Final stage amp runs off 15V generated from the +22, but if you run +12 to the +22 (or jumper it to the +15 test point), you'll get enough voltage through for the amp to bias up correctly.

You need 10.3V unregulated for it to come out of reset, but +12 works fine for that too...

IIRC, the joystick inputs on centipede don't actually do anything.
 
Last edited:
You need more than 5V to get video on most millipede boards...
Final stage amp runs off 15V generated from the +22, but if you run +12 to the +22 (or jumper it to the +15 test point), you'll get enough voltage through for the amp to bias up correctly.

You need 10.3V unregulated for it to come out of reset, but +12 works fine for that too...

IIRC, the joystick inputs on centipede don't actually do anything.

Beat me to it! :)

I marvel at the video output DACs they kludged on the MP. It's a very cheap ($) way to do it, but it seems to work just fine! I bet it took a while to figure out the balancing act with the current-limiting resistors at the base of the transistor. Some mad catz /w skillz at Atari BITD! :D
 
IIRC, the joystick inputs on centipede don't actually do anything.

I hooked up the joystick inputs on the small connector for my Centipede adapter and it worked.

I guess I'll find out if it works for Millipede, too...

You need more than 5V to get video on most millipede boards...
Final stage amp runs off 15V generated from the +22, but if you run +12 to the +22 (or jumper it to the +15 test point), you'll get enough voltage through for the amp to bias up correctly.

You need 10.3V unregulated for it to come out of reset, but +12 works fine for that too...

Thanks again, Mark...
 
Beat me to it! :)

I marvel at the video output DACs they kludged on the MP. It's a very cheap ($) way to do it, but it seems to work just fine! I bet it took a while to figure out the balancing act with the current-limiting resistors at the base of the transistor. Some mad catz /w skillz at Atari BITD! :D

The transistors have high enough beta, that the base current is negligible... emitter current is limited to ~1mA buy the 1.2k, so the base current is only ~10uA.

The transistor stage is just a buffer, shifting the output voltage up by the Vbe of the 3906 and back down the Vbe of the 3904...
 
The transistors have high enough beta, that the base current is negligible...

Quite right. I think what I meant to say was, it must've taken a while to get the value of the resistors (in the effective D/A circuit) just right to manipulate the transistor correctly in its linear region.

..emitter current is limited to ~1mA buy the 1.2k, so the base current is only ~10uA.

Ahh we're looking at different schematics. I'm looking at the rev B which has only one tranny pulling current through a 270 Ohm resistor. If I'm analyzing it correctly, the transistor is acting as a current source which pushes current through the 100 Ohm resistor proportional to the voltage at the base. Thus varying the voltage at the video output.

I know there's another rev which uses two trannys per output.

I wonder why they did it this way, i.e. had the transistor buffer stage. IIRC with games like Galaxian each video output was just the wire-sum of the DAC. Could've been a reliability issue?
 
Ahh we're looking at different schematics. I'm looking at the rev B which has only one tranny pulling current through a 270 Ohm resistor. If I'm analyzing it correctly, the transistor is acting as a current source which pushes current through the 100 Ohm resistor proportional to the voltage at the base. Thus varying the voltage at the video output.

That's actually the earlier rev...

...and the 3904 is still just acting as a voltage buffer, dropping a VBE off the output of the 'DAC'... All you need to do make sure it works right, is ensure that the DAC output minimum code is above the VBE,sat of the device. With that circuit, you get 6.8k up and 1.8k || 1.8k || 1k || 470 = 236 down, and 15V*236/(236+6800)= .487V = off (just to keep the transistor biased on a little).

Since the current through the transistor is going to be proportional to the output voltage the base current (if you don't neglect it, although they probably did) just looks like another resistor in parallel with the 1.8k pulldown.

The 100 ohm is prob just meant as an impedance match (w/ the 3904 output impedance) to match the 75 ohm load.
 
Okay, I finally got around to building one of these, and I have a question.

HudsonArcade said:
You need more than 5V to get video on most millipede boards...
Final stage amp runs off 15V generated from the +22, but if you run +12 to the +22 (or jumper it to the +15 test point), you'll get enough voltage through for the amp to bias up correctly.

You need 10.3V unregulated for it to come out of reset, but +12 works fine for that too...

I Have the +12 connected to the 10.3 inputs, and have jumpered on the board between the 10.3 lug and the +15 lug. I'm now measuring about +12.94 at each. My +5 is measuring +5.02 at the chips.

I am testing a board of unknown working condition. I can't tell if the board is not working due to a fault on the board, or a problem with the adaptor. Right now, I get a very brief screen of jumbled letters and stuff, then it goes to a all green screen.

Q) Do I need to generate a Reset pulse for this to boot? And if so, is there an easy way to do so, perhaps with a button or jumper? I don't want to build a complicated circuit just to test one damn board that doesn't even belong to me.

Any help is appreciated...
 
So, if the 10.3 unregulated inputs (both of them) are connected to the +12 on a switching power supply, it should generate the Reset when I turn everything on?

Does the green screen sound like a board fault? Or could it be a problem with the adaptor? The only voltage I'm not putting to the board is the +22 (jumpered the +12 to the +15 test point) and the -22....
 
True, as I don't have an audio amp connected so I can hear the sounds. I just talked to the guy I'm testing it for, and he says he gets the same green screen in his cab, but I didn't think to ask if he was hearing beeps....
 
IIRC it should boot to 'D's on the upper left-hand side of the screen during a POST failure.

Yeah you probably also need the -15V via the -22V to hear sounds.
 
IIRC, the joystick inputs on centipede don't actually do anything.

Whoops... meant millipede... I know centipede works, although playing with a joystick is *terrible*.

Well, it turns out the joystick inputs DO work for Millipede (at least on this board), but it DOES suck....

Green screen could be going into self-test, clearing the screen, and getting a RAM error and making beeps you can't hear.

IIRC it should boot to 'D's on the upper left-hand side of the screen during a POST failure.

Well, the boards fixed. Turned out to be a bad pokey at D4. Replaced it and game came right up, so I know my adapter works, too. Have to wait until it's back in the game to make sure there isn't a problem with the trackball inputs or the sound.

Thanks for all of y'all's help....
 
Back
Top Bottom