Astron 12 volt linear power supply

 "I never wonder to see men wicked, but I often wonder to see them not ashamed." - Jonathan Swift

NOTE:  This was part of yesterday's post, and in that post, I mentioned that I'd update it.  The section has become so long now that it needed to be a stand-alone post.

The base station radio at work stopped working a while back, and it turned out to be due to the 12 volt power supply.  I asked the boss if I could take it home and attempt a repair.  He was cool with that, and told me to do whatever I wanted with the old one.  I'm sure he has forgotten about it by now.

The power supply is a lower-end Astron unit.  A low-end Astron is still better quality than most of the solid state linear power supplies out there.  

 Nothing really seemed wrong with it.

I'm going to be quite honest and admit that I didn't do much electrical troubleshooting on this thing.  I sniffed the transformer for the odor of burnt insulation.  Then I inspected the circuit board on both sides, and saw nothing that looked unusual.  One test that I did electrically was check the output voltage, loaded and unloaded.  Unloaded it put out 13.8 volts, as rated.  When I tried to load it with a light bulb, the voltage dropped to zero.  This good-quality 12 volt power supply could not light up a 12 volt car bulb. 

So I figured the voltage control worked, but it just wasn't putting out enough juice - which made me think the output transistors had failed.  These things are screwed in and intended to be replaced from time to time, and they have a big heat sink, but eventually they fail. 

A while back I scoured the internet for an electrical schematic for the Astron SL-10MR.  I could not find that, but I found one for an Astron-11, close enough to figure out a few things.

Below:  Arrow showing the output transistors in the circuit.

The way these things work is that the rest of the power supply creates 18 volts (the line coming over to the pair of transistors from the left).  The power transistors work as a pair to regulate that 18 volts down to 13.8 volts.  A sensing and control circuit monitors the output voltage and adjusts the voltage bias on the base of both transistors.  The transistors conduct more or less current from left to right as needed to hold the output voltage stable on the right output side.  The beauty of this circuit is that the output voltage doesn't drop very much as you load the power supply.  

However these transistors have to pass a lot of current, so they get hot.  Thus the heat sink, the failure rate, and the fact that they aren't soldered into the circuit - they are plug-in transistors.  Below:  The old ones out on the bench.

I bought a box of one dozen 2N3055 Power Transistors for about $12.  The alternative was to get two of them for $12.  Amazon has some bizarre pricing.  I only needed two, but why not get 10 more for the same cost?

,

I also had to buy some thermal paste to conduct heat (but not electricity) from the transistors to the heat sink.  The thermal paste added a few dollars more.

On to the update part of this post.

Each power transistor has a thin mica insulating sheet that sits between the heat sink and the transistor.  I'm not sure why this is necessary, as the mounting screw connects the heat sink electrically to the body of the transistor.  I'm sure there's a valid reason, so I cleaned the mica sheets up and re-used them.

Below left:  Mica insulating sheets with old thermal paste on them.

 

 I was about to install a new pair of transistors out of the box, but then wondered whether I had the equipment to test the old ones vs. the new ones. 

 

And thanks to the new shop WiFi, I was able to locate a YouTube video explaining how to test these things.  It was a pretty easy test, even though I was about to replace them at that point anyway.  Putting the cart before the horse, as usual, LOL.

I used my own old Radio Shack multimeter, just like this guy, and used the exact same color jumpers to perform the exact same test.  Both the old transistors were dead.  3000 Kohms of resistance when they should be conducting slightly.

The new ones showed better results using the same test.

The second transistor that I picked out of the box and tested had a weird reading, so I threw it in the trash.

Below:  About to goop up the newly tested transistors with thermal paste.

New transistors mounted to the heat sink, with the thermal paste cleaned up as best I could.

The moment of truth:  Will it now be able to pass enough power to light a bulb?  Yes!  And back to work it goes as a spare.  You know, it doesn't take much to make me happy.  This really made my day :)

I elected to leave the plastic caps off the output transistors.  They get hot, so why block air flow over them?  In this case, the power supply is not located where it could cause harm by burning equipment or fingers.

And yes, I brought it back to work after spending my own money on it.  Don't really need a fixed voltage power supply.  It's in the file cabinet with the rest of the radio junk.