söndag 18 januari 2015

Rock Ola 473


I have been working with the amplifier during the day and the first thing to check was the fuses. The fuse for the left channel driver board was blown and the main fuse for the left channel had been substituted with a 20A fuse instead if the 3A that should have been there. It’s obvious that somebody had tried to repair it before, without success. The service solution used was instead to reconnect the woofers and play the sound at the right woofer and left mid/high range speakers. I suppose it was done to not make it to obvious that one channel was quiet. Since the rated current of main fuse was way wrong the fuse for the driver board blow instead. Luckily nothing was harmed by the wrong fuse, if I swap the right driver boards to the left channel I have sound in the left channel, so the power transistors are still OK. When measuring through the faulty driver board I found that the transistor Q6LR was shorted from the collector to the emitter and everything else seems to be OK. Unfortunately I didn’t have a replacement for it in stock, but it is on order from my supplier. I will replace it by a TIP42C and since it is not the original transistor I also will replace the Q5LR with a TIP41C to have a reasonably matching pair.


 
Since I couldn’t finish the repair of the amplifier I instead took a look at the display problem. It’s a quite interesting construction where the display segments that should not be lit are shorted through a matrix of diodes and the selection indicator PC board. When measuring through the diodes I found five that was shorted and five that was leaking in the back direction. Once the diodes were replaced the display gave the proper record numbers.
 




torsdag 15 januari 2015

Atlantis

The fire damaged Atlantis is still standing in the workshop and I couldn’t resist buying a translite that popped up at Ebay. It’s not in top condition but far better than the one that was in the game. I still don’t know if it is possible to restore the game, but it will not be a proper pinball without the translite, so if it will be restored, the translite is essential.

 
I helped a friend to repair a Bally 6803 CPU board for a Lady luck a while ago, the same CPU board as in the Atlantis. To repair the Lady luck CPU I arranged a PSU to be able to run it at the workbench using Leon Borre’s test program, which is a great tool. I then also tested the Atlantis CPU, and it was completely dead. I tested all socketed IC’s in the repaired CPU and they were all OK and since it didn’t boot using the test program it might be a faulty IC socket or a problem in the address decoding. But that’s a problem for another day.

söndag 11 januari 2015

Rock Ola 473


I have been working with the other problems and also found a few more. For the credit registration it turned out that the coin switches was connected in the wrong way. Two switches had to be closed at the same time to register a credit, that’s why I sometimes managed to get a credit while fiddling with the switches. Once they were properly connected it worked as it should. Also the coin acceptor had problems to accept the coins and this was due to old oil and grease. I have cleaned the mech and now it accepts the coin most of the time. I need to take it apart and clean it again to see if I can get it to work better.

The selector problem was a bit tricky. The key press sequence worked as it should, the lights for the first and second number lighted up and the third key was locked in the pressed position by the lock coil, but the write in motor didn’t start to spin. The cause of this problem was that the slider of the 0 button had derailed and didn’t make contact as it should.

 
The ground has to pass through the 0 switch and through the rest of the buttons in numeric order to reach the pressed one. Then the write in relay is energized starting the write in motor. Once the proper selection is reached the hold solenoid is turned off and the ground is disconnected from the write in relay due to the opening of the latch switch and the write in motor is stopped. So in my case the ground path, marked in red in the schematics, wasn’t properly closed due to the derailed 0 switch. Once the slider was put in place the selection system worked as it should.

 
The switch board had been damaged earlier and was repaired by glue and a stick of wood. I think that’s the cause for the derailed slider, the board isn’t as rigid as it was from the beginning. I did my best to secure the board to the mechanism and cross my fingers.
 

I also cleaned the locking mechanism while having it at the workbench. It was very greasy and the solenoid made a loud buzzing noise while energized. I degreased all parts and then run them through the glass bead blaster. Once assembled and lubricated where it should be lubricated, not all over it, it worked without any annoying buzz from the solenoid.



 
The window above the selector mechanism had some residue from a label on it. I cleaned it with alcohol and polished with Novus but then I realized that there must be something missing here. I’m quite sure this is the inner glass and that there should be also an outer glass and the picture entered should be sandwiched between the two. I made a new outer glass of Lexan, now I just have to figure out what picture to put in there. 
 
 
I also found two new problems while working on the machine. The second digit of the record display does display a number 1 instead of number 7 and the amplifier is quiet at the right channel. But that will be for another day.

tisdag 6 januari 2015

Rock Ola 473

Next up is a Rock Ola 473 jukebox with several problems:
1. The coin mechanism does not accept any coins.
2. It’s not possible to enter any credits manually with the coin mech switches.
3. When there are credits in the machine it’s not possible to select a record, nothing happens when the third key is pressed.
4. The gripper does not manage to the return the record to the magazine.



I started with the fourth problem. The gripper mechanism needed to be taken apart, cleaned and greased. The old grease has become more like gum and causes the gripper to be a bit too slow. It didn’t manage to close until the arm had already been lifted from the record and the record remained at the turn table. I found a youtube video showing exactly the same problem: https://www.youtube.com/watch?v=b9zx8y2WU74

Here you can see what the gripper mechanism looked like before I started the rebuild. There are dried grease and oil residue on the outside. The same at the inside, I had to use a knife to get rid of the dried grease at the shaft inside the gripper spider.


Here are the parts after de greasing and cleaning. Before I assembled the gripper again I run the gripper arm and gripper spider in the bead blaster to clean of the oxidation. Once assembled and properly greased the gripper worked as it should again and problem number four could be ticked of the list.

 

 
The gripper is a clever piece of mechanics that performs a complicated sequence using surprisingly few parts. There is probably a lot of thinking and testing invested in the construction and once it was working it was used in the jukebox designs for a lot of years, with just minor changes during the time. The most common problem with the gripper is that it is more or less ceased by old solidified grease and grime. To properly solve this problem the gripper mechanism needs to be disassembled, cleaned, greased and re assembled. This job isn’t too complicated, it takes some time and it is important to keep track of the order of the parts and the gears.
First make sure to park the gripper at a location where there is no records in the magazine. Then disengage the motor from the gripper mechanism in order to be able to rotate it manually through gear #15. It is possible to rotate the gripper mechanism using the small tab at the bottom of the motor. But it is a tedious job to do that since the rotations goes through the motors gearbox and you need to make a lot of turns to complete a record loading sequence.
Next step is to carefully knock out the lock pin #25. Once it is out it is possible to separate the inner gripper (#23) from the gripper reversing assembly (#21). There is a square hole at the bottom of part #21 where you can knock out the square shaft of part #23 in case it is ceased. At this point you can also remove the gripper arm (#22). In some cases it might be enough to clean and grease these parts and assemble the gripper again. But I think it is well spent time to disassemble the rest of it and clean and grease also these parts to ensure smooth operation. For next step rotate the gripper assembly to point straight up, this is to be able to slide out gear #12 and # 13 from the gear at the gripper shaft assembly below (#14). Make a note (and take pictures) of the location of each of the three gears to make sure you put them back in the same position, else the gripper will not work. First remove the trunnion shaft button (#11) using a screw driver, it’s just to pull it straight out. Then remove the screw that locks gear #13 to the trunnion shaft. This screw might also hold the needle brush mechanism. Remove the gear #12 and #13 from the shaft (#10) and then slide out the shaft. The spider (#8) is now also free to be taken out of the mechanism. Clean the parts from old grease and grime and lubricate them with new fresh grease and put everything back in the reverse order. Make sure to engage the gears in the same way as they were before you took the unit apart. Rotate the mechanism manually a couple of times to check that it works properly before reconnecting the motor.