FT-920

Some good suggestion to solve problems with the internal ATU-Unit

Nearly a year ago I traded for a Yaesu FT-920 in excellent condition, except for the antenna tuner. I went to each band and started the matching process by holding the “TUNER” button in, and the tuner arrived at a good match on all the bands my antenna covered (40-6 meters). Then, the next day, the match was gone on some bands. Trying to tune those bands ended in a match failure, even with a dummy load! I first suspected a relay or some mechanical failure, so I opened the transceiver and tried to determine the problem. There seemed to be no mechanical problems; everything was secure. But to be sure I went through the Antenna Tuner Adjustment procedure in the Service Manual. Everything seemed to work for a couple of days, and then the problem started again. I dug into it once again and tapped on all the relays as it was tuning, hoping to find one sticking or making erratic contact-None found! I decided to watch the tuner as I went through all the bands, and noticed that one of the stepper motors had turned its Variable Cap beyond the fully unmeshed position. Normally the home position for these guys is fully meshed. There is a mechanical stop at this position. Looking from the front of the radio, the stepper motors turn the caps clock-wise, to, at most, the fully unmeshed position. There is no stop at this position, and there is no feedback to the controller to indicate whether this position, or any for that matter, has been reached. This time one was beyond the unmeshed position and I thought that one of the stepper motors had a problem and replaced it. I should have watched it more closely, because that wasn’t the case. The problem returned, and with it, continued slipping of my sanity. I had about decided that this was a return to Yaesu problem, and connected a manual tuner. It was driving me crazy, however, knowing that the auto-tuner was there, but not working. I opened it up again, started the tuner on each band and noticed this time that BOTH of the stepper motors had gone beyond the unmeshed position! Normally when you first turn the transceiver on, the Caps go to the home position (fully meshed) and then go to the saved position for that particular freq. But after one or both of the caps have gone beyond this unmeshed position, it seems the home position reference is lost. Then when you move to another band, the controller reads the stepper motor positions and relay sequences from memory, and acts accordingly. However, the motor position information is referenced from the home position, which has now changed, and since there is no positional feedback, the cap settings are completely wrong. And if you try to tune again, the controller thinks the caps are somewhere within that 180 degree arc from meshed to unmeshed, but instead one or both of them is beyond 180 degrees. Therefore the caps will be turning in an opposite direction. That’s why no match could be achieved, even on a dummy load. If you turn the radio off and back on, the motors will turn for the length of time they are programmed to turn, but will not go fully home, if they’re beyond 180 degrees. If you do this several times, they will eventually get there. Then the positional information for most bands is correct again. You go to one of those bands and everything works again. I eventually discovered that doing the warm reset (pressing GEN and ENT on the keypad while turn the radio on) would clear up the problem for a while, but it kept returning, usually every few days. I knew I shouldn’t have to reset it that often. Then one afternoon it was doing this wonderful thing again. Everything was fine on all bands until I went to 6 meters. The match was fine there, but when I went back to 10 meters, I heard a dip in the signal level and then it came back up a little. I knew the caps where in the wrong place. I opened it up, and that was indeed the case. I kept it on 10; turned the radio off and on enough to get the home position correct, and started the tuner on 10. It did its thing and a match was achieved. I went to 6 meters, the caps moved to the new location. Then I went to 10. Both caps went beyond 180 degrees. I went through the off-on-tune thing several times, and each time, when I went to 6 and back to 10, the caps went passed 180 degrees. The only way to clear it was to do a warm reset, and retune on each band. It then dawned on me what the problem could be. The tuner uses a serial eeprom to store values. This eeprom, as well as the micro controller itself are powered from 5 vdc coming from a regulator on the Tuner Control Board. This 5v line also goes to the Main Tuner Board where the relays, coils, caps, etc are. Having had some experience with these devices, I surmised that noise/rf could be getting into them via this 5v line and causing erroneous values to be written into memory. There are a couple of 1 mf electrolytics and .01 mf disc caps along this line, but with as much rf as could be running around in there, I thought it needed more. I took the board out, fired up the soldering iron, and added a couple of parts. With these additions, I’ve had NO problems with the tuner at all over a 3-week period. As a matter of fact, it seems to tune faster and smoother than before. Here’s what I added: 1. 330 mf 16V Electrolytic. This could probably be much smaller and work as well, but this is what I had on hand. 2. .047 mf ceramic disc. I wanted a .1 but this is also what I had on hand. A. Very carefully unplug all cables from the Tuner Controlled board. Don’t try to jerk them, you’ll pull a wire out or break it. Gently rock them from side to side till they come loose. B. There’s a white (at least on mine) flat cable that comes up from underneath the transceiver and slides into a connector on the Tuner Control Board. Don’t force this one out. There are small clips on each side that you have to pull up to release the pressure on the cable before it will come out correctly. C. Remove the 4 screws that hold the board in place. D. There is a screened position on the board marked “C5547”. Nothing is in that position on the board (mine), nor is there a part on the schematic with this designation. This is where I placed the 330 mf Electrolytic. Please note that the Negative connection is marked with a dot on the board, at least on mine, please verify that on yours! E. The .047 mf cap needs to be soldered as close to pin 8 on JP5004 as possible and ground (underneath the board). I traced the 5v line and found a spot on the board where I was able to get to a ground connection that already had solder on it. I also dress-up the leads going to this board some. Esp. the leads going to the stepper motors. They are at least 3 times longewr than necessary. I bundled them with a wire tie. That’s it! Put the board back in place and replace all the connectors. Be sure to get the right connector to the right socket for each stepper motor. I know this is fairly lengthy, but I wanted to give you some of the history in case you have similar problems. I hope this is of some help. 73 ----- Anthony NT4ZO ===== NT4ZO