FW3A Troubleshooting / FAQ

Its possible. i had a closer look this morning with a magnifying glass and it looks like that trace is now melted & stuck to the spring.

EDIT-UPDATE: ok after closer inspection, there is no doubt the small trace had melted to the tail spring, why it happened after a small fall off the night table is a mystery.

Probably hit the floor tail first with the impact causing the spring to pierce the mask, switch on, ramp up then the trace acted like a fuse with the current overload. Interesting.

That was my first thought as well. Surely the energy would not travel both directions along that trace. It seems only half the trace should be burnt, probably the outer half from the spring. I suspect there was a short in the inner tube to the outer tube which caused a lot of amperage to flow all the way across that little trace. The fact that the urnt trace got stuck to the spring seems almost like and after effect than it is a cause. IDK, what do you guys think?

To me it looks like high current flowed through the inner tube when it normally would only see very low current.

Extra length of the button top probably didn’t help either

Neal responded quickly and sending a replacement fortunately. What i will do however on the replacement is remove the spring, and resolder it rotated a little, so that the end of the bottom spring creates a open gap over that trace to prevent the spring from breaking the coating and shorting to that trace if that is what happened then it hit the floor, (which is what probably happened as CRX mentioned, the impact of landing tail cap down probably caused the short. its an easy fix to prevent it from happening in the future, and will show a photo of the mod/fix when i receive it.

So, is this a general design flaw of the light? Is it likely to happen with a drop that hits tail-first? Or is this a case where that light had a manufacturing flaw, and it’s not likely to happen in other lights?

I think its a matter of everyone checking the tail cap board spring, to see if theirs is soldered in the position where the spring is touching that trace or of its positioned with the spring gap over the trace. (the one i had the spring was soldered with its bottom edge directly against that small trace that crosses under it, and the spring end-gap was to the right of the trace.

Looking at the image of my version the gap in the tail spring coil is direclty over that switch trace.

Actually, it should help. The extra battery length means there’s less room for the battery to gain momentum which reduces the impact forces.

Are you saying the sharp end of the spring was directly over that trace? I assume it was a flat coil.

I’m not sure you understand. All versions have the trace go under the spring. That is not an issue. If the sharp (cut end) of the spring is directly over the trace and it’s soldered in such a way that it’s pressing directly against the trace, then a heavy blow might push it through the solder mask and make contact.

Your light in the picture looks fine. I see no chance of the spring ever touching the trace.

Den, I’m sorry to hear about your light’s failure. I’m glad you’re okay and that Neal is taking care of a replacement.

My spring is resting on the board in the area of the trace. A piece of paper won’t fit between the spring and PCB.

With the trace and FW3A label positioned at 6 o’clock, the end of my spring is soldered to the pad directly to the right, between the 4 and 5 o’clock position.

This places the ‘gap’ between the 1 and 2 o’clock position.

I plan to rotate the spring when I get back to my soldering iron.

I assume (perhaps incorrectly) that springs tend to be installed so that the ‘end’ is on a pad and connected with solder.

Yes that was my point, the gap…
How it should be.

The bigger question is if the trace burned first due to some other reason and the spring was nearby so they got melted together. Or was there current from the spring to the trace?

Negative current from the battery is already flowing up to the driver at all times (same for positive current) so I’m having trouble unstanding why the trace burned.

This trace basically connects a very weak positive signal from the MCU down the inner tube and across this trace. When you push the button this weak positive signal gets grounded through the vias and to the spring.

So if the spring shorted the trace it should be no different than holding down the button. I’m pretty sure the MCU limits the signal current to like milliamps.

I can’t think of any reason that trace would carry high current, high enough to burn it. The spring touching it should not cause a high current flow.

In these 2 pic’s it looks like the spring is only secured/soldered on one of the 3 pads? The other 2 pads look untouched?

Looking at the 5 I have, all of them have an air gap over the signal tube trace with the spring sitting on the 3 solder pads.

Just an observation…

That appears to be the fault in the one i had short, while the edge of the spring was not directly on the trace, as seen in the photo the spring edge was roughly 110 degrees at roughly 2 o’clock on the board, so the spring was directly resting on the trace. Also i did notice as seen in that photo the spring is only soldered to one of the three solder points, and i could lift the spring up off the board a little as it only had one weak solder point. As seen in the second photo, you can see where the two other solder points were not soldered to the spring at all.

lol i had posted that exact thing just as you did. indeed, it was only soldered in one small spot on one of the pads, the other two were not.

Resistance, the spring only held or making contact on one solder pad. Maybe the srping got hot enough to burn thru the damaged masking after the fall, and the heat from resistance burn the trace? Kind of strange really, unless the driver shifted and made some kind of short? It is strange indeed…

So the bottom end of your spring is soldered at the pad nearest the ‘F’ in the FW3A label? That allows for the maximum air gap.

Upon closer inspection, my spring has solder on all three pads, but only one is making what I’d consider a ‘good’ connection. The other two look a bit cold and could use some more solder. I’ll touch everything up when I rotate my spring.

Yours looks like the position of mine. If they had placed the spring cut-end on the 7 o’clock solder pad, then there would be a good 1/2 mm or more gap over that narrow trace. Its an easy fix to rotate the spring as fortunately the switch board is easy to remove from the tail cap.