DoTransistorsWear Out?
several times and so why bother to give a report at all. I mentioned getting crowded out earlier. I have on many occasions tried to operate on both 20 and 40m when there has been a contest on but find that shortly after calling CQ a contest station will come onto a very adjacent frequency and it seems that if I persevere they gradually move closer and closer until they are on my frequency and I have to go QRT.
We in the hobby are constantly bemoaning the lack of people, especially young and middle-aged people, coming into the hobby. If they would like to operate and have to work during the week, then the weekend is often the only time they have but on most weekends they cannot get on any of the bands except the three ‘new’ bands but often at least two of these are not open. I have noticed that now there are so many contests at weekends they are also starting to become regular mid-week activities.
As mentioned earlier I am not anti contests but I do think the time is rapidly approaching when there should be either some limit to the number of weekends when they can take place or a strict regulation to just how much of any band can be used for contest working. At the moment the whole of the CW section is used with no consideration to the QRP or QRS sections. I am sure that there could be portions of each band that are free of all contest stations and the organisers of the contest arrange monitoring stations to police this and disqualify those who don’t abide by the rules. Is it not possible to combine several contests onto the same weekend?
I am sure that there will be quite a few who disagree with me as well as those who agree but I do believe that if we are serious about getting people into the hobby we should give them the opportunity to get on the bands and actually learn their operating skills.
John Pullen G4TGE
Barton upon Humber that’s only a partial solution. The good news is that most contests, even at weekends, are of limited duration and don’t attract a large number of participants, so there is usually free space on the bands. Thanks for your kind comments about the magazine!)
Dear Don,
In response to Jim Carter’s letter, Do Transistors Wear Out? (Letters, September) some thoughts. Firstly, everything wears out eventually (except eternity?). The original belief that unlike valves, they would last has proved untrue. There are several failure modes known to me. Note, I am not a specialist in this matter but I have repaired many equipments over many years.
Firstly, as they suffer thermal cycling, this causes strain both within the chip, and between it and the connections. One may assume there is both thermal and magnetic vibration also due to the signal current. In high power RF transistors failure was caused by aluminium atoms being able to wander off, weakening the wire bond to the chip. The answer to that was to use gold. It must be remembered the chip is exceedingly small so the current density can be very large.
Poisoning can occur due to imperfect seals of the case; hence why the Military insist on ceramic not plastic packages in some cases, or metal not plastic.
I learned when still at school, repairing a friend’s HiFi, that transistors can go intermittent. It appeared a driver had gone, but on ringing the maker, an engineer told me that the likely cause was an intermittent open circuit in the output device, that when I measured it, the ohm-meter’s voltage had temporarily cured it, but it would fail again, and advised its replacement, and I didn’t need to order the Japanese original, but TIP41/2C would do fine (a transistor I have found uses for long after then). I had a duff transistor in my Marconi spectrum analyser: but when I checked on my PeakAtlas semiconductor analyser, it was fine. I tacked it back in, and it worked: for a couple of minutes. Another case was a C-Audio RA 3000 amplifier: 625 watts/channel.
It had a short on one side of the output complementary pairs but as I checked for shorts with each output transistor, they all seemed OK, so I had to laboriously remove and check. Again, I used the very useful Peak-Atlas and as I went along these N-channel enhancement MOSFETs, one came back as depletion-mode. How it managed to change I cannot imagine, but