AR II Repair help

morphis72

Member
Joined
Nov 25, 2013
Messages
691
Reaction score
2
Location
Raleigh, North Carolina
So, my AR II got fried in a voltage spike. The audio channel is still kicking out the correct voltage but the +5 is gone.

So far I have:

Rebuilt the power brick - now getting about 16v DC on pin 1/2/3

AR II -

Replaced Q1, Q2
Replaced the voltage pot

I'm getting 15.5 vDC on Q2
I'm seeing 15.5 vDC on the case of the bottlecap transistor in the center of the heatsink.

I am seeing .45v DC on the big resistor in the heat sink.


update:
I have now replaced the bottle cap with no changes as well as Q5 which seemed to be putting out the correct voltage but the casing had melted a bit so swapped it to be safe.

Still no +5.
 
Last edited:
Your best bet is to get a hold of member andrewb (Andy) He will be able to help you. He helped me when the AR II in my Centipede failed and is very knowledgeable.
 
Check to make sure diodes CR1 and CR4 are not shorted, and are not in backwards (if you did remove or replace them at any point).

Make sure none of the caps are shorted.

Make sure you have a good mica or silicone insulator on the bottlecap transistor, and that you didn't accidentally put too much solder on the 2N3055 pins, such that they are shorting to the heatsink, under the heatsink. If in doubt, pull and re-solder it, and use as little solder as possible.

If you've replaced Q1 (LM305), Q2 (TIP32), and the 2N3055 with known good parts, then there's something else simple that is being missed, as those are the 3 active components that pretty much make up the 5V circuit.

Where did you get your parts? It's not a bad idea to suspect new parts as possibly being bad, depending on where you got them. It doesn't happen often, but it happens enough that you should always leave room for it being a possibility, when you're stumped. The TIP32 and 2N3055 are easy to check *out of circuit*, with a DMM. The LM305, not so much. (Or at least I don't know of a simple way to test it, so I just swap them out if they are suspect.)

Other than that, check all of your work, and make sure there are no shorts between traces or pins, and make sure none of the leads from the components are bent over and touching each other, on the solder side. (I always trim any that are sticking up too far, what could get bent over and touch something they shouldn't.) To look for shorts, hold the board up to a bright light source, and look through the PCB. Any shorts between traces will become much more visible.

You can also check continuity of the pins on each of the Molex connectors, to make sure they all are connected to the PCB. I had one AR recently that threw me for a loop, until I found that one of the pins had broken inside the connector, so it wasn't making contact.

Look for other broken solder joints, especially where traces connect to pins, like on the TIP32. You can get hairline cracks where the trace meets the solder blob of the pin sometimes, and not see them. Check everything with your DMM.

As a last resort, you can try pulling the audio amps (Q5 and Q7), and seeing if that makes any difference. I recall seeing one AR where a short in one of the amps was affecting the +5V, but I don't remember exactly how it was doing so, as I thought the audio was supposed to be independent from the +5, but apparently there's some way they can become coupled, via a bad TDA2002.
 
Check to make sure diodes CR1 and CR4 are not shorted, and are not in backwards (if you did remove or replace them at any point).

Make sure none of the caps are shorted.

Make sure you have a good mica or silicone insulator on the bottlecap transistor, and that you didn't accidentally put too much solder on the 2N3055 pins, such that they are shorting to the heatsink, under the heatsink. If in doubt, pull and re-solder it, and use as little solder as possible.

If you've replaced Q1 (LM305), Q2 (TIP32), and the 2N3055 with known good parts, then there's something else simple that is being missed, as those are the 3 active components that pretty much make up the 5V circuit.

Where did you get your parts? It's not a bad idea to suspect new parts as possibly being bad, depending on where you got them. It doesn't happen often, but it happens enough that you should always leave room for it being a possibility, when you're stumped. The TIP32 and 2N3055 are easy to check *out of circuit*, with a DMM. The LM305, not so much. (Or at least I don't know of a simple way to test it, so I just swap them out if they are suspect.)

Other than that, check all of your work, and make sure there are no shorts between traces or pins, and make sure none of the leads from the components are bent over and touching each other, on the solder side. (I always trim any that are sticking up too far, what could get bent over and touch something they shouldn't.) To look for shorts, hold the board up to a bright light source, and look through the PCB. Any shorts between traces will become much more visible.

You can also check continuity of the pins on each of the Molex connectors, to make sure they all are connected to the PCB. I had one AR recently that threw me for a loop, until I found that one of the pins had broken inside the connector, so it wasn't making contact.

Look for other broken solder joints, especially where traces connect to pins, like on the TIP32. You can get hairline cracks where the trace meets the solder blob of the pin sometimes, and not see them. Check everything with your DMM.

As a last resort, you can try pulling the audio amps (Q5 and Q7), and seeing if that makes any difference. I recall seeing one AR where a short in one of the amps was affecting the +5V, but I don't remember exactly how it was doing so, as I thought the audio was supposed to be independent from the +5, but apparently there's some way they can become coupled, via a bad TDA2002.

CR4 is shorted but was so long before I got the game. Someone desoldered one end of it and just lifted it up. It was working okay without it and to be honest I never even noticed it was like that until after I started trouble shooting after the voltage spike.

I checked the insulator on the transistor before powering it back up. I always check to see if those have continuity to the frame or heat sink after installing them.

The last time I rebuilt one of these I had two to do. I ordered kits from Bob Roberts and ordered a spare. It's been sitting on my shelf for about a year. So, his kit didn't have the TIP32 in it... It was an alternate. If I were buying now I would have picked up the TIP32C (i think is the better speced part would have to double check.) But, the one he included in his kits seemed to work just fine on the two I rebuilt when I bought the kits.

I did check all the caps in the 5v circuit for continuity but I didn't pull one end so your mileage may vary on that being a valid test.

I have a cheap but effective magnifying headset I use when soldering and if there is any question of bridges I check for continuity. So, everything is good there for those smaller close together pins.
 
Last edited:
Bob Roberts was known for having bad 2N3055's in his kits, which were counterfeit. If they are printed with Toshiba in red, they are likely junk. See here:

http://forums.arcade-museum.com/showthread.php?t=392021


As for the other stuff, we've covered just about everything there is to cover with non-working AR's, so it's a good idea to question yourself, assume you're wrong, and/or missed something, and go over everything again.

The recommended replacement for the TIP32 is 2N6107, though honestly, I've never seen one of these go bad. That's not to say it can't happen, but it's not common, and if it's bad, you should be able to verify that with your DMM, else it's ok. See Bill's list for more, regarding part subs and sources:

http://www.biltronix.com/arcade_electcomp_03.html


Also, try pulling Q5 and Q7, and seeing if that changes things. You can power the AR up fine with the audio amps removed, it won't hurt anything.
 
Bob Roberts was known for having bad 2N3055's in his kits, which were counterfeit. If they are printed with Toshiba in red, they are likely junk. See here:

http://forums.arcade-museum.com/showthread.php?t=392021


As for the other stuff, we've covered just about everything there is to cover with non-working AR's, so it's a good idea to question yourself, assume you're wrong, and/or missed something, and go over everything again.

The recommended replacement for the TIP32 is 2N6107, though honestly, I've never seen one of these go bad. That's not to say it can't happen, but it's not common, and if it's bad, you should be able to verify that with your DMM, else it's ok. See Bill's list for more, regarding part subs and sources:

http://www.biltronix.com/arcade_electcomp_03.html


Also, try pulling Q5 and Q7, and seeing if that changes things. You can power the AR up fine with the audio amps removed, it won't hurt anything.

Well, my first check of the kit parts show it is from Malaysia - so no red Toshiba to worry about.

I will go through it again top to bottom. Do I need to pull one end of the caps to check for shots? I figured even in circuit it would read a short...

Also, what about that lifted diode at CR4? Do I need to replace it?
 
To test for shorts, no, you don't need to pull one end. Only if you were going to measure them with a cap meter. But if one is bad, it would more likely be one of the small non-electrolytic ones, which you should be able to catch with a meter in-circuit. The electrolytic ones never really go bad, unless they've been physically damaged, stepped on, etc. (And even then, I've seen severely dented ones still work.)

Also, check for broken resistors, and cracked glass-encased caps. I'll gently push on each component sometimes, to make sure it isn't split in two in a way I can't see.

Pull Q5 and Q7.

And yes, replace the diode. Any 1N400x will be fine. Just make sure the direction is correct, as there are AR's with them in both directions, due to the trace layout being changed on some revs, so go by the silkscreen.
 
And yes, replace the diode. Any 1N400x will be fine.

I have some 1n4004's which I am going to use to replace it with. The specs look like they call for 1n4002's originally. Is this just a voltage rating thing similar to using a capacitor that has been rated for higher voltages than the original specs needed?
 
Yep. 1N400x's are all 1 Amp diodes, just different voltages. Higher voltage is fine, so the 1N4004 is ok.

So, CR1 was the culprit. Now I am back to the voltage being too high. Two different AR II's are reading the same as well. If I crank the voltage pot all the way down I am at about 6v DC.

I have rebuilt the power brick - I believe you were working with me on that in a different thread.

Replaced big blue, bridge rectifier, fuse block, and all the connectors to the fuse block. Pins 1, 2, 3 on the power brick that supply the unregulated 12v are all reading 20.5v DC. I believe something still isn't right with the power brick.
 
Last edited:
Back
Top Bottom