Acom 1500: Intermittent Faults & Other Problems

Intermittent faults are the worst problems to find and repair. Unless there is a clear set of steps to reproduce the problem it can remain elusive. It is equally true of hardware and software. There is a lot of both in a modern ham shack.

Several months ago my 4-year old Acom 1500 amplifier began to periodically trip the protection circuits. There was no unique cause or consistent pattern in the protection code data. In a few case there was the soft "pop" sound near the front of the amplifier cabinet. A reset and all was well again. Then the faults became more frequent, whether transmitting or not, and whether online or in standby. It was perplexing.

A cursory inspection inside showed nothing obvious, no carbon tracks or other physical signs of distress. With close to 3000 volts in the B+ supply there are many potential paths to failure, whether a weak capacitor, dust or loose connection. Typical suspects such as the tube (4CX1000) and vacuum relay did not appear to be to blame.

Since I have a spare amplifier (3 in total) I set the 1500 aside until I could decide how to proceed. There are no factory authorized Acom service centres in Canada and it is inconvenient, to say the least, to ship heavy and fragile equipment to the US or overseas. Inquiries among friends generated no useful insights. It was a puzzle and concerning but there was also no urgency. I fretted having an expensive piece of equipment forcibly idled.

When I returned to the amp some time later I got my first clue to what might be happening. At times during normal operation the plate current (Ip) would read 0 ma. That's a very odd reading when there's a kilowatt of RF being transmitted. The same would happen when idling or transmitting without input. But again, only sporadically. There was no pattern, and a fault still occurred from time to time. I had no good ideas on how to proceed so I set it aside again until about two weeks ago.

This is a recent picture taken after I appeared to have resolved the problem. Once I narrowed the cause to measurement and protection, specifically Ip, (or one of the causes) and did my best to deal with it, I installed a physical meter in the Ip measurement circuit. Since the digital display on the amp cannot show instantaneous deviations of Ip, I hoped to spy any instantaneous anomalies by putting the meter where I could see it while operating. The microammeter only lightly loads the circuit and was calibrated with a series trimmer pot.

Unfortunately this circuit assumes that the current pickup in the B+ is working. There is no good or really safe way to splice in my own. Ip is indirectly measured by detecting the voltage across a precision low value resistor that is in series with the B+ line. The resistor is indicated by the arrow in the picture of the mains PCB and rectifier/filter assembly; The resistor is on the latter. Only low voltage DC is presented to the Mains PCB.

There are two stages of processing of the detected voltage. There is one op amp on the Mains PCB and another on the Control PCB. A 3-pin cable runs from the detector to the Mains PCB and a 5-pin cable runs from there to the Control PCB. All are circled. The physical meter taps into the Ip pin on the 5-pin cable, which is between the Mains and Control PCBs.

One immediate suspect was the connectors. It is not uncommon for oxidation or other flaws to creep in that degrade continuity. That can result in a 0 reading or a spike when it makes and breaks connection. I pulled and cleaned every connector of this type throughout the amplifier, and there are a lot of them. It has to be done carefully since moderate and evenly applied force across the connector is needed. It is easy to cause unintended damage on those with many pins. The pins are approximately the same dimensions as Dupont connectors, both in pin size and 0.1" spacing.

There is no good way to know whether the connector work made a difference. All one can do is run the amp and see if the intermittent fault returns. I did that but I then also took another step. 

I opened, inspected and vacuumed dust of all compartments in the amplifier. I then blew high pressure air through the inaccessible areas to remove more dust. It's a useful step though not without risks. These include contact damage, air pressure damage and static discharge. The latter is a greater risk this time of year when the relative humidity is low.

Nothing was damaged by the cleaning. That was a relief despite my confidence in proceeding. I was as careful as I could be, especially around microprocessor circuits. I also disassembled the front of the amp and exposed the Control PCB. With care, and magnifying glasses, I inspected the measurement components and connectors. No flaws were found. All the PCBs are now very clean.

I went through the RF deck with lights and mirrors to see under and behind components. I did the same for the tube compartment, top and bottom. There was dust inside the chimney so I cleaned it and the air filter. The antenna selector board was also inspected and subjected to air pressure to blow out any hidden dust.

When all was done and the amplifier reassembled, it worked. That is, there have been no faults since, other than a power on fault that detected a connector I had forgotten to push back on. Oops. Since then I've worked many stations, SSB and CW. CW can be a better test due to the repeated on-off "spikes" of a high power carrier. This weekend I used it and the Acom 1200S to make over 1000 QSOs during the LZ DX contest. So far so good. 

Is it truly fixed? Of course I can't know that without identifying the ultimate cause. Perhaps after 4 years it was simply time to clean the inside of the amp, any amp. All that I'd done until now was to periodically clean the air filter.

Repair work came to many hours. I doubt it would have been more economical to ship it to another country since an intermittent problem takes time to find, and they of course charge by the hour. If they recommend board or component replacement, assuming they also fail at finding the fault, it would be even more expensive. Since working around the high voltages inside a tube amplifier requires care and skill, it is not recommended for the inexperienced. Every ham has to make the judgment call for themselves. 

Solid state amps only appear to be safer since there are no high voltages beyond the mains supply. However the power supply and circuits deal with the same power level. No kilowatt-class amplifier is a safe space. Be careful if you choose to do as I have done. I'm no expert but I do have experience working on amplifiers.

I hope that if nothing else you enjoyed pictures of the inside of the amplifier. As I understand it, the Acom 1500 is no longer being manufactured. The industry migration to solid state continues. A well designed and maintained tube amplifier can last a lifetime. With few new ones on the market, they have to.