The root cause of the short which blew the fuse was a shorted zener diode which is installed across the output of the supply. The purpose of that zener is to provide protection for the radio which is connected to the supply in the event of a failure within the supply regulator circuitry which could result in the output power increasing to a high voltage. The zener is a 15 VDC rating so it's purpose is to clamp any increase in voltage to that level. During normal operation the supply puts out +13.8 VDC so the zener is effectively invisible to the supply. Of course now it is shorted so it is anything but invisible.

First step in the troubleshooting and repair process was to remove the offending component and see if there was anything else wrong with the supply. I clipped out the shorted zener and confirmed that, yes, outside of the circuit it still indicated a dead short on the Ohmmeter, and no, the output terminals of the supply no longer indicated a short now that the zener was removed. Perhaps that was the only problem. The zener shorted out and other than that, the supply was functioning properly. That thought was bolstered by the fact that when I replaced the fuse and turned on the supply, the fuse did not blow again and the output voltage read +13.8 VDC.

Now it's time to do a quick check with the radio power leads connected. On the bench the output had read +13.8 VDC but that was without a load. That's not what happened when the radio was connected. Now the output read +21 VDC and the radio did not power on. Well that's not good. Was it really trying to supply my radio with 21 volts? I was not going to push the experiment too far. I don't need a costly radio repair or worse yet, have to buy a new one to replace a total loss.

In order to assure that the radio actually does work and that a part of the problem is not with it, I hooked up another power supply. One which puts out the proper voltage but is just a workbench supply and does not have adequate power to handle transmit with the radio. The radio powered on fine and worked just fine in receive mode. Voltage measured was +13.8 VDC.

The power supply worked - or I should say, it appeared to work - open circuited displaying the proper output voltage read on my Fluke DVM. Connected to the radio, the output voltage is too high. Obviously there is something more involved than just the zener.

Outside of the matter of troubleshooting the problem with the supply, one thing I wanted to do was to replace the fuse/fuse holder with a pushbutton circuit breaker. Perhaps it's just a matter of me becoming more lazy in recent years but I'd much rather push a button to reset a tripped breaker than to go searching for a proper sized fuse. Not that I am oblivious to the fact that without finding and fixing the reason the fuse or circuit breaker tripped in the first place is is just an exercise. While the supply was open for repair, it was a good time to do that. The replacement of the fuse holder with the circuit breaker was painless and it fits perfectly. The photos may suggest that it is a tight fit but that is not the case at all. There is plenty of clearance between it and surrounding objects.

Screw-in cap fuse holder

With the new circuit breaker installed.

I must comment at this point in defense of my failing to clean the dust layer off of the surface of the inner case walls before making these photographs but you will just have to forgive me that slight indiscretion. I had no excuse, really. Too focused on trying to resolve the problem with the supply.

I will jump ahead a bit, skipping some of the detailed steps I took during the troubleshooting process. The documentation for the supply stated that if the zener which is across the output shorts then one or more of the regulator pass transistors (mounted on the rear panel heat sink) will be shorted. They tested ok, but something was definitely wrong to cause the supply output to be 21 volts and not power the radio.

Now I had a new problem. Suddenly the fuse blew again (this occurred before I installed the circuit breaker) and I discovered that two diodes in a half-wave rectifier which is used to provide bias and pre-regulation had shorted. Completely by coincidence, the collector of one of the 2N3055 regulator pass transistors had shorted to the case ground. Though the transistor was not shorted internally. There was not short between emitter, base or collector of the transistor. The pass transistors are mounted on a large heat sink on the back panel. Electrically the transistor cases (the collector) is intended to be isolated from the the metal of the case. The screw that mounts the transistor to the circuit board on the inside passes through the metal of the heat sink and is insulated by having a sleeve of what looks like shrink tubing. When I found the short to ground and removed the screws of the affected transistor, I noted that on one of the screws the threads had apparently worn through the insulating sleeve causing the short. I replaced that sleeve with two layers of shrink tubing and after reinstalling the transistor, the short was gone.

I have no good explanation as to how the short came to be at that time. The blown diodes were not part of the original problem. Only the shorted zener. Of course there was a problem present since the supply output voltage was too high when connected to the radio. At that time the diodes were not being shorted out. The only explanation I have is that initially the "short" was not complete. Instead of reading zero ohms to ground, it may have read some higher resistance. Enough to cause the regulation to be off but not the later developed zero ohms which later in the troubleshooting process caused the diodes and yes, the primary line fuse, to blow. Perhaps during the course of manhandling the assembly, I managed to cause some movement which allowed the threads of the screw to cut all the way through and make hard metal to metal contact with the case.

The documentation stated that if the zener across the output terminals is shorted then one or more of the transistors in the pass regulator chain will be shorted. I never found that to be the case. They were not talking about a shorted mounting screw but rather an inter-electrode short such as a shorted collector/emitter. I guess I lucked out on that one. When I corrected the screw short, replaced the two shorted diodes and installed the new circuit breaker, the power supply worked fine. I looked at the output on the oscilloscope and was pleased to see no ripple on the DC line. Before I installed the new output zener, I hooked up the radio and it worked fine. Once that was verified, the final steps before closing up the case were to tidy up the wiring and solder in the new 15 volt zener across the output terminals. The power supply is now back in service and working fine.

Originally the circuit board was affixed to the floor of the case with rivets. I removed those to be replaced with screws and nuts. Makes access to the board easier. The two shorted diodes have been removed and the new ones not yet soldered in.	The replacement diodes are being soldered in. I used a heat sink on the leads of the diodes to minimize heat transfer to the diode possibly damaging it.

One feature which I really appreciate and respect about this power supply maker is that all of the components used are commonly available ones. Off the shelf items as opposed to oddball proprietary components often used. With such a supply as this, it's easy to obtain replacement parts. I bought everything I needed, including a few items which I thought I might need such as a couple of 2N3055 equivalent transistors, but did not need to use, at Fry's. I also have a schematic and a parts list from the maker. As I recall, I received that from them a few years ago when the main bridge rectifier failed and I ordered that from the maker. Of course now I know that I could obtain an equivalent replacement outside because I now have the documented specs for the component. The document I have actually lists voltages found throughout the regulator circuit.

Since it is a commercial supply, I feel it prudent to not publish the schematic or too much technical data regarding the supply on this page. I will be happy to answer specific questions regarding the supply or the troubleshooting project.

Written December 2010

Custom Search

Questions? Comments? Contact me