FW3A Troubleshooting / FAQ

He already did. It’s head related.

A lot of people say this is normal at very low output levels. They should all get back to normal as the power level goes back up.

Yes, correct, it is definitely the head, as it does the same thing on all three tubes/switches. I pretty much figured after checking everything out it was more then likely a driver issue, I did not think of the of the 7135/s as the culprit until you mentioned it. It does make sense

Here is my troubleshooting thought process, if anyone is interested.

We know Dels driver design is good because we have hundreds (thousands?) of working lights, but 0.5% seem to have strange driver issues (not inner contact tube stuff). This leads me to believe it might be a component issue. Genuine components can be hard to get in China, but they usually work just fine. There may be cloned Attiny85 (according to Lexel), FETs and 7135 chips. The other components are resistors, capacitors and diodes and those don’t usually have issues. Sometimes an SMD capacitor can short internally, but that will manifest itself very clearly. Some of the FETs don’t start having issues until you get into the 150+ watt range, plus that is not an issue in the FW3A. So we have the 7135 chips and Atiny85 chips to look at. 7135 current regulators get cloned hugely. There are clones of clones of clones, etc… and we have definitely seen some go bad on many different models of lights. Usually just a tiny percentage of those lights as well. It’s like the electrical tolerances are not as good as the genuine ones from ADDtek, etc… If you get one of those outlier chips it might cause this type of issue.

I can’t really remember what kind of trouble the cloned Atiny85 chips caused (Lexel thinks there may be clones because they sometimes won’t reflash properly, but it might just be that the company that originally flashed it used the wrong fuses or that they encrypted it somehow). I don’t think there were any functionality problems. Since we are only getting a flicker and all the other more complicated functions work, I’m leaning towards a bad 7135 chip.

The only other factor might be the soldering itself. I remember seeing one leg of the MCU not fully soldered on some other model light and that caused weird problems. If you are good with an iron, you could go over all the joints to make sure they are 100%. It’s not something the average person can do, but it doesnt hurt to try.

Sorry I must of missed it.

Unfortunately I missed the warnings about the tail and opened it a few times. I’ve lost my nubbin now. Considering most flashlights open at the tail, having a tiny loose part like this is poor design imo.

The design has been changed. Now the tail caps have retaining rings. They originally wanted to glue the tails, but we forced them not to.

How? It’s all people talk about in any FWXX thread and also there is a note in every box. It’s not a poor design, it’s a modders/enthusiast/flashaholic dream imo.

Gave that a try with a bit of hot air no change unfortunately. I appreciate your taking the time to respond!

I was just talking to Texas_Ace and he agrees that the defect rate on those 7135 chips seems to be getting higher and higher. He has run across so many bad ones that he’s changing his driver designs to either FET plus single 7135 chip or else dual FET with resistor bank. Reliability is way better.

If the defect rate starts to get too bad, BLF may need to start designing similar drivers with 1 or no 7135 chips at all.

My problem is with Neal’s. I had a defective unit a month ago and he said he’d send a new one. I been messaging him for weeks and he finally gets back to me and just says “hello” in the message and ignores my questions on where is the replacement. He finally says today oh sorry I’m too busy, I’ll send tomorrow. checked his FB page and others have the same complaint. Does anyone know is this normal for Neals? Can I get anyone else to warranty this?

Same here, eventually got refunded after PayPal intervened.

OK, so:

I have had 2 of these for a while now. No mentionable issues.

Today I chance to one of them the 2-way clip.

In the process I noticed a clear O-ring under original clip, it cut in pieces while taking it off.

Now, with 2-way clip if I tighten the tailcap gently: flashlight will give me only 1% power momentary strobish-thing. Nothing else.

If I tighten the tailcap FIRMLY, the switch “click” sound disappears and switch feel goes from “clicky” to smushy or something. No feel at all BUT thats not the worst it also likes to Fire up on full beam on the table all by itself.

Threads and surfaces are cleaned, no help, but I think this has something to do with the click itself.

From the rubber plate between SS-tailbutton and clickswitch the plastic “nub” was loose. I took it away -> bad feedback from clicky, put it back -> no help to original issue.

Yep.
Assembled/disassembled a few times more.

I don´t get it…. when I place all the parts inside and press them in tailcap for example with a battery, the “clicky-feel” is correct.
When I assemble the light, it just doesn´t work.

My purple FW3A just ran one of my 18650s down to 1v while turned off. Anybody had this happen? Is there a way to fix it?

Sometimes this happens. :weary:
AFAIK, Neal is still a one man store. He takes all the orders, gets everything that is in stock shipped out, orders new inventory, tries to answer all the PM from here, Facebook, emails from his website, etc…, but he gets overloaded and backed up. I don’t know why he didn’t send a replacement a month ago, it might have slipped through the cracks. He’s an honest guy, though so he will get it sent to you. Even if it goes out today, it might take a week or so for him to get the tracking number. This is the busy season in China and the mail system also gets overloaded.

Warranty is only through the company you bought it from. No one else will do it as it’s basically them sending a new light/part for free, plus paying the shipping costs. This takes a big bite out of the small profit margins.

Ok, cleaned it up, again.

I cut away 50% of the hard plastic stud inside the rubber. Not helping.
I cut away the outer rim from rubber and it helped. I think the rubber could not sit again in the tailcap groove once you take it out one time.

On the other hand, the clicky now does not have much feel and the metal cap tries to stick once and awhile.

Oh wel… at least I can use it now, sort of :expressionless:

Cutting up the rubber was a bad idea seeing as that is very difficult to undo. Did you have solder blobs on the underside of your switch board? If so, I would have tried removing/flattening those out if you have a soldering iron. Would I do that now though? Probably not

Can you tell me about this clip?, I’m not up to date, but I don’t recall there being a 2 way clip for the FW3A.

By “1% power momentary strobish-thing” you mean the double blink you get when power is first connected? That is normal. The threads are not anodized so it easily gets main power when screwing it together.

Your problem is tricky. It sounds like when you tighten the tail cap firmly the nub was already pressing down on the metal dome of the switch, but just barely short of making contact. Then the switch might actually touch and that’s why it ramped up.

Have you tried removing the 2 way clip and seeing if that fixed the problem or if it’s still the same?

Something is going on with your switch. Try assembling it like this. Hold the tail cap threads up, drop the metal button in, make sure it’s centered. Drop the rubber diaphragm in, make sure it’s centered. Make sure the plastic nubbin is held in it’s groove. Then drop the switch circuit board on top, make sure it’s centered then keep holding it threads up and screw on the battery tube. Verify the clicky feel.

Sometimes if the tail cap is turned sideways before threading the battery tube on, the rubber part might shift sideways a little and cause an issue.

You can also verify the switch working by using a DMM in continuity mode. Take the assembled battery tube/tail cap, put one probe on the battery tube threads and one on the end on the black inner tube. Clicking the switch should complete the circuit.

I have a video of how to do it. Switch check starts at 2:07.

If everything is assembled and lined up properly, it kind of sounds like the nub is too long. That seems really unlikely though.

Do your parts look like this?


.

Neal is selling a deep carry clip.