I don’t think this is a stupid question… but we will see.
I bought one of the new model FWAA lights from Neil back in March. I have really liked the light, but it just stopped working properly. I had some issues with intermittent button function. I took it apart and cleaned all of the typical things. The ends of the inner tube, threads, contact rings in head and tail. Put a freshly charged cell into it and… it worked fine.
Until a day or so later I tried to use it and got nothing in response to any button presses. I took the cell out and put it on a meter. It came up as 0.2 volts. Obviously over-discharged. So I put a new cell in. A soon as I tightened the head the light cam on at full output. Nothing I could do with the switch made any difference.
What do you folks think? A shorted FET? MCU failed. I am just trying to get an idea of what happened and whether I might try to repair it. It is probably not worth it, but maybe it is a good challenge.
THX. No I didn’t do any disassembly further than taking the head and Tail cap off of the main tube. This time. Though a couple of months ago I did do the O-Ring mod for increasing the switch pressure. But it has worked for a couple of months since. Would this spring problem also cause the battery to over discharge?
Edit. I took a switch assemble (full tail cap) from w working FW1AA and tried it. I got the same results… full power. No ability to changle levels.
EDIT 2 I took the head off of the ailing FWAA and put it on the body of the Working FW1AA and got the same result.
If the O-Ring is too thick(CS) it can also cause eSwitch issues.
Each FW is slightly different, I use 1.5mm on some and 2mm on others. I’ve even used a razor to slice a hair thin layer off of one.
You want a small amount of play in the metal button.
If it’s too snug, it won’t allow the switch to break the connection.
Sometimes it’s intermittent and sometimes it constant like you have now.
OK.. I did check for clearance/play when I originally did the mod. Plus it worked fine for a few months. But, also, I tried a known good switch from my FW3AA and it did the same thing. I tried the malfunctioning head with the entire switch and body from a known working FW1AA. It still malfunctioned. I really think something died in the driver. I wish it was the switch, much easier to fix.
Consider me a newb, but I have been building a lot of drivers over the last 2 months. Every single time I’ve had a build go direct drive, and the emitter wires were NOT shorting to the reflector, the 7135 had failed, shorted open, or had a bad/cold solder joint.
That was one of my first thoughts. I can pull the head apart and make sure there are no shorts.. Though I think it would have shown up before now… But Maybe… At least that would be an easier fix. My hands are not as steady as they once were so SMD stuff it a bit intimidating.
The thing has come out of my pocket and fell a couple of feet… but mostly (I think) on carpet. I just looked at it and there are no dings on the bezel… or anywhere that I can see. So it hasn’t hit hard enough to show physical damage. Are you thinking the reflector may have cut into the leads for the LED because of impact?
I had thought of that.
Or a potential impact causing a borderline solder joint to have an issue.
I did have an earlier driver build that was seemingly working fine for a while.
Then started direct driving like you mentioned.
When I did the troubleshooting, I removed a few chips. The 7135 had major signs of a “cold weld” on one of the legs.
I soldered the exact same 7135 back in place, and it has worked fine since then.
It sucks, because it wasn’t obvious that was the issue until the chip was removed from the pcb.
I wouldn’t start desoldering stuff until you check the emitter wires, and hopefully someone more knowledgeable than myself chimes in on the stuff I’ve mentioned.
Yes. I should have thought of that . It is a TIR
So most likely a component failure.
I did send Neil a note. Maybe he can help me out with a driver or replacement light.
Meanwhile I will pull the head apart and see what I can see.