Hans, that makes sense and is good news.
You are whittling down the pile!
JZ
toggle quoted message
Show quoted text
On Thu, Sep 21, 2023 at 4:15?PM Hans Summers <hans.summers@...> wrote:
Hello JZ
Failure of the modulator to operate properly shouldn't result in the death of BS170s, I don't think there's any indication of that. In fact it could well be the other way around, if a BS170 fails short. Though these AOD403 are pretty chunky.
FYI I discovered a related QMX bug today (and resolved it in the next firmware I'm working on). On power up, the power modulator is at full voltage. Not zero as it should be. It isn't until you first key the transmitter that the modulator starts its proper operation. This is a harmless bug but could have led people to conclude the modulator wasn't working correctly when it fact it's just a firmware bug.
I put the modulator output (PA supply voltage) now permanently displayed on the diagnostics page, not just when transmit is pressed. And I made it red if it is less than 85% supply during transmit, or more than 1V during receive. Green otherwise. Just to put some (generous) bounds to indicate proper operation. Showing it during receive was how I discovered the bug that it's on full after power-up!
I also put SWR on the diagnostics screen and was pleased to see 1.1 on my dummy load and > 9.9 to open load (I set a max display limit of 9.9). So the next firmware should have SWR working.
73 Hans G0UPL
On Thu, Sep 21, 2023, 10:53 PM John Z <jdzbrozek@...> wrote:
Ryuji,
I hope you avoid that experience. You are correct about the intrinsic
protection incorporated in QMX,, and so your risk is low, but not
zero.
I have seen on this forum at least one example of the modulator
failing to operate properly, maybe two. That is not a lot, but it did
result in the failure of BS170 transistors.
JZ
On Thu, Sep 21, 2023 at 3:46?PM Ryuji Suzuki AB1WX <ab1wx@...> wrote:
JZ, that is true, but QMX doesn't key abruptly anyway.
As you see on the spec sheet AFM907NT1 can withstand VSWR of 65 or greater at a reasonably harsh condition. The abs max DC ratings and the device's RF robustness are dealt with separately. That is different from general purpose switching MOSFETs where there is no special consideration for RF amp uses.
If you look at the device structure, LDMOS has guards against high voltage at drain. That structure does not exist in regular MOS power switching devices.
In case of AFM907NT1 in QMX, one way to buy more margin is to increase the output transformer's impedance ratio and use lower drain supply voltage.
Anyway, if I ever have a great pleasure of having to replace AFM907NT1 due to death by any of those spikes, I will be sure to report back.
