Bad Voltage Regulator on a Rockola 1000

Q&A about all types of jukeboxes: Wurlitzer, Seeburg, Rock-Ola, AMI, and more.



light-o-matic
Regular Member
Posts: 41
Joined: Tue Jan 05, 2016 5:32 pm
Location: Prospect, CT, USA

Re: Bad Voltage Regulator on a Rockola 1000

by light-o-matic » Mon Mar 21, 2016 4:45 pm

The relay you are referring to in the power supply, on the same board as the fuses, also turns the LM317 on and off. if you look at the schematic, that relay shorts out a zener diode on the regulator leg of the LM317. When the service switch is off, the LM317 will put out very little voltage. When you switch the service switch to SCAN, the relay clicks and the output of the LM317 will jump to 28V. If moving the service switch from OFF to SCAN to OFF again changes the output of the LM317 from 1v (about) to +28v to 1v, then the relay and the LM317 are working. If the output of the LM317 isn't +28v when the service switch is in SCAN, then you still have a problem with that+28v power supply circuit. Make sure you have +40 going into the input lead of the LM317.

You can also put a voltmeter across the zener diode and you should see the voltage change when the relay is on. I seem to recall that the zener was difficult to get access to. it was very close to another component. I wish I had the schematic with me today so I could quote you the component numbers. Sorry about that.

It sounds like you checked for cold solder joints. I had to remove the PC board to check them. The bad joints were not on the wires, but where the connector was soldered to the underside of the circuit board. Just making sure.

As I stated earlier, Rockola puts +40v into that LM317 and then regulates it down to +28v. But when the relay is off, the output of the LM317 goes to a very low voltage (around 1v). This means that there is 40v across the input to output leads on that LM317. Absolute maximum pin-to-pin voltage is 35v (37v maybe? can't remember) according to the spec sheet. It seems to work, but not a good design. As I said before, i put about 6 or 8 forward biased diodes in series with the input to bring the +40v down to around +35v. It's ugly, but it may account for all those bad LM317 chips on these Rock-Olas. This does not seem to be the problem at hand, however, so let's fix the known problem first.

Mark.

Who is online

Users browsing this forum: No registered users and 8 guests

It is currently Thu Oct 06, 2016 11:10 am