Can't adjust +5v on Space Invaders power supply

Spyridon

Well-known member
Joined
Nov 1, 2007
Messages
7,849
Reaction score
242
Location
Chester Springs, Pennsylvania
I have a Midway Space Invaders and the +5v is reading at +6.21. Adjusting the pot has no impact on the voltage. I replaced the pot (as well as all the caps) and no change. It is still reading at +6.21 and turning the pot doesn't change the voltage.

Any thoughts?

IMG_9673.jpg
 
Probably a bad 2N3055 transistor.

I replaced the 2N3055 transistor and no change.

I did checked the LM305 and it was taking in +12.92v unregulated and spitting out +5.82v. I replaced it and now it is much closer to +5v (I think around +5.05v). However, now I'm down to +4.78v at the test point and the pot still doesn't change the value.

I need to look at the schematic and see what is between the LM305 and the test point.
 
Just a thought from my foggy memory,do you have the game pcb hooked up during your test?
I am thinking these act goofy if they are not under load.So i was curious if maybe you unplugged the edge connector from the game board to protect it until you got the voltages right.I know i worked on something like that anyways lol.
 
Just a thought from my foggy memory,do you have the game pcb hooked up during your test?
I am thinking these act goofy if they are not under load.So i was curious if maybe you unplugged the edge connector from the game board to protect it until you got the voltages right.

Yes, I had the PCB disconnected until I got the voltages correct.
 
Okay, here is where I am at on this.

Without the PCB hooked up, the power supply at the +5v test point is reading 4.94v. Turning the pot had no impact on the voltage.

Hooking up the PCB, the +5 voltage drops to 0.651v. +12v and -5v are all good.

Thoughts?
 
How thick are your wires between the PSU and the PCB? If they are the same gauge as those TIIIINY wires soldered onto the board then that will be a major problem.
 
Check the +5 pin on the pcb for a short or leakage to ground. With the game off and edge connector disconnected. Put your red meter lead on the +5 pin and black lead on any gnd pin.

18 PIN Edge:
1 +5V
2 +5V
3 +12V
4 +12V
5 -5V
6
7
8
9
10
11
12
13
14 GND
15 GND
16 GND
17 GND
18
 
Last edited:
You're looking for high resistance. Just checked two different boardsets and I got 2K+ on one and 900+ on the other. This was done with the sound board removed. Keep in mind the sound/daughter board uses the +5 rail as well. So basically if you have low resistance then suspect the pcb. The tantalum capacitors (standing tear drops) used on the supply rail are notorious for becoming leaky or shorted.
 
Okay, I tested that out and I got 189.7 ohms when I use ground pins 14 or 15.

Also, you listed pin 18 as ground and I think it should be for the monitor
 
Last edited:
Corrected the pinout listing above. Video and GND were swapped. Removed the video pin text as we're not interested in that for this test. Any GND pin can be used for testing purposes. They are all connected together.

I just took a resistance reading from +5 pin to GND on a fully working Space Invaders Deluxe CPU board and surprise! this one read 5K+ ohms. So I would say if you are touching the right pins with your meter then you have something leaking to ground on your board.

What you do to narrow down the issue is remove the 8080 cpu and take a reading while it is out. If the reading is still low then do the same with the eproms/roms. It's normal for the reading to get a little better as you remove and retest, but you will be looking for a significant difference. If it doesn't improve then unsolder one leg of the tantalum caps (tear drop shaped) in line with the rams. Do one tantalum at a time and retest until you find the bad one or replace them all at once if the problem is there.
 
Also, I just got a spare board in the mail last night and it read about 2k ohms. I haven't plugged this board in to see what it does yet.

I'll test the new one out but also follow your advice on the current one to see what it does.
 
First thing I would check are all the various ceramic and tantulum capsi to se if any are bent over and touching a chip leg or something.

Also check the back to see if there's any scratches or anything where one trace is touching another.

If you happen to come to the Philly area anytime soon let me know we can take a stab at it.
 
Last edited:
Corrected the pinout listing above. Video and GND were swapped. Removed the video pin text as we're not interested in that for this test. Any GND pin can be used for testing purposes. They are all connected together.

I just took a resistance reading from +5 pin to GND on a fully working Space Invaders Deluxe CPU board and surprise! this one read 5K+ ohms. So I would say if you are touching the right pins with your meter then you have something leaking to ground on your board.

What you do to narrow down the issue is remove the 8080 cpu and take a reading while it is out. If the reading is still low then do the same with the eproms/roms. It's normal for the reading to get a little better as you remove and retest, but you will be looking for a significant difference. If it doesn't improve then unsolder one leg of the tantalum caps (tear drop shaped) in line with the rams. Do one tantalum at a time and retest until you find the bad one or replace them all at once if the problem is there.

I'm having a similar issue with my SI board (detailed in this thread). The reading from the +5 pin to ground is 533 ohms, so something is obviously leaking to ground. I pulled the cpu and the roms, no change, then unsolderd 1 leg of each of the tantalums, 1 at a time, and still no change. I even unsoldered the disc caps and checked, no change. If electronforge was taking repairs at this time, this board would already be on it's way! ;)
 
I'm taking on SI boards if you're interested. Are you getting that reading with just the motherboard or is that with the daugherboard on there as well?

Edit - if you're checking between the 5v pin (pin #1) and ground (say, pin #17), I'm getting 65ohms with a Deluxe daughterboard installed, ~90ohms without. Thats coming from a board that I literally just powered up and had functioning fine. I'm definitely seeing it fluctuate some (the 12v and -5v readings were all over the place, indicating I'm reading across lots of caps).
 
Last edited:
Back
Top Bottom