Hi all, a brief update on everyone's favorite FD issue...
Power relay MOSFET failure update
As most fo you know, MOSFET failures continue to affect some FD owners, and I'm very eager to find both the true cause(s) of the issue and a solution, if possible.
To better understand the issue I've been reading up on common MOSFET failure causes, which, let me tell you, is a *great* way to cure insomnia!
Of the several causes I've read about, the likeliest so far seems to be
inductive voltage spikes, which are large swings of energy that occur when an inductive process is stopped suddenly. All of that energy needs to go somewhere, and unless one gives it a specific path in which to drain, it can slam back into the MOSFET and cause damage to its structure.
Sometimes the returning energy spike is strong enough to kill the MOSFET, but many times nothing bad happens, at least externally. This matches the particulars of our MOSFET failures as I understand them, so this sounds like a promising cause.
This writer had a good description of the issue, as well as some potential fixes:
https://www.daycounter.com/Articles/Inductive-Voltage-Spike.phtml
Anyway, that's my current hypothesis for the likeliest cause of these failures, that or something similar - there are several energy-spiking phenomena, and I think it may be one of these.
More testing ahead, hopefully resulting in a fix
Now that I have a viable hypothesis, as well as some potential fixes to explore if that hypothesis proves true, I would like to take some readings and confirm (or disprove) that hypothesis. Science, but more Sherlock-like (
) than mad-scientist-like (
).
This issue should be relatively straightforward to see with a decent oscilloscope, and all but impossible without one. My current oscilloscope is barely an oscilloscope (
), and not up to this task at all, so I've purcha$ed a better oscilloscope. It should be here Friday or Monday, and then* I can get a better look at what is going on inside the heater as it operates.
*Once I get it set up and calibrated. And climb a short but probably steep learning curve. And probably several other tricky things about which I am so far blissfully unaware. Sigh...
If the issue does turn out to be a voltage spike, there are several relatively easy ways of addressing that with diodes or other small components (discussed in the article linked above), including a few that might be easy to retrofit.
That's where things stand with this issue at the moment. I hope to know more within a week or so.
I am going to continue to build and ship heaters while I'm continuing to research this issue, as it seems premature to stop production at this point, but I'm aware that some of the heaters I am building now may be affected by this issue. I don't know if this is the right call - there are arguments for both sides... 
I am hopeful, though, that this new oscilloscope will help me identify the root cause and come up with a fix for the issue ASAP.
That's your mid-week Fluxer update.
Cheers,
