Jump to content
midibob

Tales from Midibob's workshop MPU4

Recommended Posts

Happy New Year folks.

 

With the tester now finished it was time to start on the MPU4's. Only a year later than planned!

 

Starting with a nice burn up...

post-6626-0-43918500-1515586144_thumb.jpg

 

It was a lot worse than it looked and once cleaned up was as good as new.

 

The next one suffered the usual corrosion around the battery area. A new rivet put that one rights.

post-6626-0-90261100-1515586144_thumb.jpg

 

The next couple had suffered the same fault of IC13+14 burning up. I still can't work out why this fault occured so if any one has any suggestions....

IC3 (68B21) had also gone.

Once the chips were removed.....

post-6626-0-48394300-1515586145_thumb.jpg

 

....and the after shot....only 1 rivet needed :)

post-6626-0-93808100-1515586145_thumb.jpg

 

This next one had a load of faults attributed to it so first thing was to give it a thorough inspection. First thing to note was this. I'm sure you'll all spot it.

post-6626-0-26229400-1515586146_thumb.jpg

 

There was also a high current drain on the board. With power on the 5v rail would go as low as 4v and vary up to about 4.9v. This was traced to the 6840 which was getting pretty hot!

 

The majority of boards are needing rivets in the battery connections as the main ground track either has no copper left or it just disconnects leaving C47 hanging in the breeze. The positive connection varies as to which one is live, depending on the board revision, and this again rots away from it's connecting track.

 

I've also come across an error in the diagram. IC24 (74LS122) has it's connections reversed ie pin 6 should go to pin 3 and vice versa.

This came about after trying to find out why motor drives C & D weren't working along with all the lamps and all the switches.  IC24 had failed and wasn't switching IC23 (74LS138) which controls the switches and lamps. The same output from IC24 also enables IC11 (74LS541) which drives motors C & D.

 

That's all for now but there will be plenty more. :headache:

 

 

 

  • Like 1

Share this post


Link to post
Share on other sites

Very nice work indeed bob and great to see the tester being of worth and being used for its intended use .is that a blob of solder under the ic socket ?🤔

Share this post


Link to post
Share on other sites

I was there when we were trying to get the machine to run. The burnt boards all caught fire due to a defective program card. I'm not sure what was bad in the program card but I suspect a dead short caused the mpu burn. Nick should send you the program card if he can find it LOL

Share this post


Link to post
Share on other sites

I was there when we were trying to get the machine to run. The burnt boards all caught fire due to a defective program card. I'm not sure what was bad in the program card but I suspect a dead short caused the mpu burn. Nick should send you the program card if he can find it LOL

will dig the cart out and pass it on to him paul .

Share this post


Link to post
Share on other sites

Thinking back there is more than 1 error on the mpu4 diagram if i remember or find the diagram i will hi-light them.

Share this post


Link to post
Share on other sites

Great work bob, another tec well on it’s way under your belt, brilliant work :)

Share this post


Link to post
Share on other sites

Hi Paul,

 

Yep, Nick related the cart being the problem. Can't see why but you never say never.

 

There's definitely more errors on the schematic but I hadn't seen this one mentioned before.

 

Another fault this afternoon. The last board when powered up had a short on one of the lamp transistors. Being as my tester has LED's they don't obviously glow brighter when you come across this type of fault but because there are 8 lamps inside the machine you can see when they glow extra bright. Of course you also get two rows coming on together but it's nice to see the background light up too. :)

 

Anyway, the fault turned out to be too much solder. It looked like every joint had been re-soldered on the board!

 

post-6626-0-91624400-1515619393_thumb.jpg

Share this post


Link to post
Share on other sites

I couldn't either but after burning up two boards we plugged the third one in with nothing but the red power plug. All was well. Then we put the program card in and nothing else. Powered on and bang, up she went. The ROMs were moved to another card and all was good.

Share this post


Link to post
Share on other sites

I have also seen those reel driver IC's fail with the reel plug on the wrong way round (due to a missing key pin) took three boards to realize that was the cause !

 

 

Paul

Share this post


Link to post
Share on other sites

Thinking back there is more than 1 error on the mpu4 diagram if i remember or find the diagram i will hi-light them.

 

You mean the circuit diagram commonly distributed & same one we've all downloaded from here at some point?

Share this post


Link to post
Share on other sites

You mean the circuit diagram commonly distributed & same one we've all downloaded from here at some point?

 

Yes, I've highlighted 3 here:

 

post-4024-0-75902800-1515669515_thumb.jpg

Edited by Wizard
  • Like 1

Share this post


Link to post
Share on other sites

Here's one that needed quite a bit of tack wiring for the Alpha and triac ccts. To try and do the tracks on the top side would have been a waste of time as they looked like they were corroded in more than one place. It's much quicker and more reliable to use tack wires in this case.

post-6626-0-56913700-1515698549_thumb.jpg

 

Next one was a bit deja vue as iC13 &14 had burnt again. I thought there were only two of these so here we go again.

Once the chips and resistor networks were removed and cleaned up there was no damage thankfully. It's likely IC3 has perished too.

 

post-6626-0-38347200-1515698550_thumb.jpg

 

One thing I didn't spot straight away, as I was focused on the IC's and resistors, was this...

 

post-6626-0-83358700-1515698550_thumb.jpg

 

I'm sure you'll all spot it!

 

After powering up it doesn't boot so that's tomorrows job.

After doing so many MPU3's which the majority were no boot it makes a change to see one that doesn't. I'm sure I'm going to regret saying that. :clown::lol:

 

I had a few reel sensors and back boards to check. Using the Test rig in 'start mode' you can plug the optic lead from the MPU4 into the reel back board and then plug the sensors in and monitor the operation via the lamps and bleep on the rig. The amount of time I saved was a real bonus and not only does it check the sensors but the on board chip (LM339) too. :)

Share this post


Link to post
Share on other sites

To be honest I thought the burnt ones would be scrap. Just shows what can be done.

Share this post


Link to post
Share on other sites

Good to see so many repairs already Bob, you are whizzing through them 😊🤔

 

Keep up the good work!

  • Like 1

Share this post


Link to post
Share on other sites

You mean the circuit diagram commonly distributed & same one we've all downloaded from here at some point?

 

Yes

Share this post


Link to post
Share on other sites

See post 12

 

There may be more errors but those are the most obvious

  • Like 1

Share this post


Link to post
Share on other sites

Nicely done bob .definatley a broken track there id say !

Share this post


Link to post
Share on other sites

The sound circuit tant can also burn. I had that the other day. Lots of smoke and it melted the sound pot.

 

It's actually made me a lot more cautious about leaving them on unattended to the point where I'm pondering a co2 extinguisher and an additional smoke alarm.

Share this post


Link to post
Share on other sites

Yep, those pesky tants do make a mess when they go!

 

I'm still having fun and games with this board. IC3 was swapped out and this enabled it to boot. It wasn't stable however and any flexing of the board would drive it potty. The alpha was also corrupt although it was now functioning.

So there was still a way to go and as it had already had remedial work done around the battery area it's likely there are still some tracks damaged somewhere. That can wait till tomorrow.

 

post-6626-0-65583200-1515780934_thumb.jpg

 

Moving on I noticed one of the motor driver IC's had been changed and under one of the pins it looked black.

post-6626-0-78798200-1515780933_thumb.jpg

 

Hmm, this had better come out and I'm not sure I'm going to like what's underneath!

post-6626-0-01780300-1515780935_thumb.jpgpost-6626-0-42378800-1515780935_thumb.jpg

 

All that carbon has got to go otherwise it will eventually track across. For those of a nervous dispostion look away now. :lol:

post-6626-0-84375200-1515780935_thumb.jpgpost-6626-0-27030100-1515780936_thumb.jpg

 

It all has to be dug out, luckily it hadn't reached the adjacent hole. Once that's done it needs to be filled with epoxy.

post-6626-0-73148100-1515780936_thumb.jpg

 

When that's dry it can be drilled and riveted. A bit of the old green stuff and it's good to go. Won't need that strap wire again either. :)

post-6626-0-32212900-1515780937_thumb.jpgpost-6626-0-20437600-1515780938_thumb.jpg

 

I'm also having fun and games with another board but more of that tomorrow if I get time.

Share this post


Link to post
Share on other sites

Fixed!! :clap:

 

This was quite a bizarre fault in the way it exhibited itself. It appeared that flexing the board was driving it potty even though it wasn't booting correctly in the first place. As the initial triac lamp test wasn't running correctly I swapped out IC8. This didn't make any difference but what I did notice was just lightly touching the top of IC8 would stop it going bananas??

 

Knowing what I know now it was obvious what was happening but I couldn't see the wood for trees at the time. After a fair bit of head scratching and meter prodding I found that touching pin 23 of IC8 with my finger was enough to stop it going mad. Sure enough the track was broken from IC14 (74LS138) and therefore the 'enable' pin (23) of IC8 was floating in the breeze! Putting in a tack wire and all was good.

 

Another misprint on the diagram here which I think has been mentioned somewhere before but the outputs of IC14 are all one out compared with the designation of all the chips they feed ie Y1 should be IC2 (not IC1) Y2 should be IC3 (not IC2) etc.

  • Like 2

Share this post


Link to post
Share on other sites

Well found bob nice work and what we've come to expect now !maybe a revamped diagram is on the cards at some point .

Share this post


Link to post
Share on other sites

Good work Bob. 

 

It shows how difficult some of these faults are to find.

Share this post


Link to post
Share on other sites

This next one has been a long drawn out process.

 

Most of these MOD 2 boards (if not all) I'm replacing all the triac driver transistors as all the legs are usually corroded, the same with T20 and 21.

 

On the first power on I was greeted with a puff of smoke and the sound of a bowl of Rice Crispies! C55 10uF tant had exploded!! :wacko:

 

Once replaced, on switch on all the triacs were firing and staying on but going no further.

This was down to IC8 (68B21) being faulty. Once changed the triac lamps on the tester ran their sequence but then stopped on #1 which signifies a RAM problem.

In this case it wasn't the RAM chip itself as I jumped in before thinking and swapped out the RAM which of course made no difference. At least it was socketed now though. Attention was then turned to IC19 as it's legs were corroded being right next to the battery. This was also bad so in with a new one.

 

Still stuck on triac lamp 1?

 

Next thing was to check the RAM voltage and lo and behold it's only about 1.2v although when I first touch the meter on it's about 3.5v? The track from T20 base to R116 had gone open circuit due to corrosion. Once a tack wire had been fitted it now booted. :)

 

First row of lamps flash on the tester and I'm thinking yep all done. Well no, it froze on the first row. All the switches were inactive and a power cycle would do exactly the same again. At leat it was consistant. :lol:

IC33 (4049) was found to have a bad gate and as this controls the switch matrix it was locking up the program possibly due to it thinking the CANCEL switch was always operated. Once changed I was greeted with a full set of rows and columns as normal.

 

Next thing was to check all the switches. The 4 switches (3rd in from the right) 14,6,6,15 didn't bleep when operated. This was down to a bad gate on IC34 (4049).

 

Checking through all the various functions, everything was looking good until the sound check. All I got was a lot of white nosie and some faint tones mixed in but quite a few were definitely missing. This can really only be the sound chip (AY-3-8913) or the 68B21. Unfortunately I don't hold any of these and as they are usually soldered in I can't do a quick substitute from another board. This will have to wait till I get my hands on some. I pretty much suspect the sound chip though rather than the PIA but you never say never.

  • Like 2

Share this post


Link to post
Share on other sites

×