EB5AGV's Workbench: Kenwood TS-480HX repair

Fixing a supposedly unfixable rig

AGVradio Lab

Even if fixing radio gear nowadays is my main work, I enjoy doing it. And I specially enjoy what I call challenges. A challenge is for me to fix a rig which has been deemed unfixable by other technicians. Yes, I know it may sound like a bad idea... but the satisfaction you get when the unit comes back to life is just too good to spare. So, when I saw on an eBay listing a nice Kenwood TS-480HX with a failure defined as needing control circuit replacement, I knew that was a good one for me :-)!

Unit as received

I knew from the eBay listing that the unit came in a nice briefcase, along all the accessories:

Once I got it out of the briefcase, it seemed like a good unit, with just some scratches on its case:

A good thing is that it came with a couple optional filters installed:

And, as described, it powered up and received perfectly. The fault was that it didn't transmit:

The technician said that the control unit doesn't recognize the power amp, which sounded to me like an imaginative way to say that he had no clue of what happened ;-)

Finding the fault

The TS-480HX is kind of special, regarding its power amplifier (PA). It is made of two identical 100W units, even with a sepparate power cable each, whose outputs are added up to get the full 200W. This is the PA:

What I found is that the RF TX signal was OK up to the PA section. Then, it died...

Checking the voltages at the first pre-driver Q1 (2SK2596) I found that there was no DC at drain on TX. This was clearly the problem for the signal loss on that stage. I checked the schematics and then on the second pre-driver Q2 (2SK2595) I found the same problem. Both voltages came from a switching transistor Q806 (SI7445DP). I measured it and it had a correct voltage on its drain (13.8Vdc), but there was no correct control signal on its gate, which should be 4.37V on TX and was close to 13.8Vdc.

That signal is managed by Q807 (DTC114EKA), which is a digital transistor. As it was located on the bottom side of the PCB and I didn't want to dismantle it, if it was not needed, I followed the control signal to Q807 on the schematic. And, sure enough, it went directly to the main CPU IC204. It is very bad news when the problem is on a faulty CPU port!. So I checked that signal right on the CPU pin (labelled PSC1 or PCS1, depending on the schematic sheet, a bug which took some time to be found)... and it was dead!. It should go HIGH (3.3V) on TX, but it kept LOW at 0V. Not good!

This is the CPU. Not a chip you want to replace at home, right?

Fixing the transceiver

OK, now that I knew where the problem was, I needed to find a solution.

I simulated the PSC1 signal and, when externally forced to 3.3V, the rig worked fine on TX. So I was confident the problem was that but... what could I do?. Checking again the schematics, I thought of a solution: there is another signal, labelled TXC, which goes high when the rig is on TX. I know that probably there is a delay among those signals but, well, could I use it to replace the damaged PSC1?. Well, there was little to lose... so I cut the trace from the faulty CPU line and bridged together both lines to the TXC output:

With some hesitation, I powered up the rig, and checked the now shared TXC control signal. It kept nicely high on TX. And, yes, there was power output :-)!. I tried on several bands, keeping the power low (as the power supply on hand was not able to manage the 40 Amps needed for full power). Sure enough, power was OK at all bands and it seemed there were no side-effects for using the TXC signal as TXC + PSC1. Good!

Then I found an small glitch: indication on the rig was OK for SWR and ALC, but not for output power. I looked at the schematics and found that the VSF voltage, which is sampled from the RF output, is OK. And that it arrives to the CPU A/D input. So I think that the problem is probably a bad calibration, due to someone messing with the Service Menu while trying to fix the rig. I will look at that later.

Well, as I was happy enough with the results, I did some small cosmetic works. The front panel glass was a bit scratched:

So I polished it. The result is pretty good:

I checked also the serial port of the rig, and it works fine :-)

Work pending to do

I need to try to calibrate the power meter, but will wait until I arrange a proper dual power supply system (30 + 30 Amps), so I can use the rig at its 200W setting.

So far I am a pretty happy camper :-)!