I bought a TF X9 as one of my first lights after finding this place. I also ordered a 2.8A Nanjg 3/5 mode driver https://www.fasttech.com/p/1186301 As I had read that the X9 was both direct-drive and under-driven. Today the tips for the soldering iron came in and I set to work. However, upon screwing down the cap and clicking the light on, there was not much light output, there was obvious PWM flickering and whine, and it was stuck in SOS mode and would not even finish a full cycle before resetting or cutting off. Sometimes it will cycle into a low mode. When installing the driver I:
-Unsoldered the existing leads so as to re-use the original ones. I made dead sure the solder did not go anywhere it was not already when I pulled the driver from it’s bubble-wrap packet. The solder for the + lead is a little close to the ring around the edge of the board but there is a definite visible strip of green separating the two.
-I did have to file down the outside edge of the driver to make it fit in the recession on the underside of the X9’s pill, but none of the chips or components are damaged.
-The ring edging the bottom of the driver has been soldered to the edge of the pill.
Is it possible I put too much heat into the driver at some point and damaged it? Or have I fumbled a short into the system at some point?
Thank you for the time, and I apologize if this is a stereotypically-inexperienced problem.
a common problem that cause such reaction is the reflector touching the emitter solder points on top of the pcb.
You might also have a ground issue. Double check that the ground is solid between the driver and the pill, and even that the pill has solid connection to the light itself.
Other than that, I too am at the mercy of the more experienced.
Ah, good point DCM. He did indicate he changed the wires. So the emitter connections would have been worked on. Even the plastic reflectors can create shorts due to the metalized finish also covers the back of the reflector.
I don’t have any unprotected cells My most recent tests have been with no tailcap, just jumping the edge of the body with the neg pole of the battery. And yes, in retrospect it is but it’s never been a problem with any other forum before.
I tested the light without the reflector/head first, just the pill screwed into the battery tube and the naked emitter. There’s no short. And I tried resoldering the bottom edge to the pill, no-go. I even moved the + lead slightly, thinking it might be shorting against the lip the driver rests on when pressed in. It still does what I said it does.
have you tested only driver and led connected, outside of the pill?
P.S. Try using a macro mode on your camera it will allow you to get very close to the object…
I'm stumped. The driver looks good as far as I can see in the pictures. Looks like you have it well grounded. I can't really see behind the red wire. The left leg of the amc7135 chip should not have any solder touching the LED positive wire or that diode that is adjacent to it.
Could the red wire going to the LED from the driver be touching inside the pill? I like to put a sleeve of shrink wrap on it, push it down firm against the solder connection (even split the end a bit so it’ll flare over the solder) and use the hot iron to shrink it to the wire. This way it cannot touch inside.
Don’t see anything else. Exasperating, but you could remove the driver completely and try it with the star while outside the light.
Next time, try testing the driver with leads soldered to the emitter PCB outside the light first, as suggested earlier, to make sure that’s working all right before putting things together in the pill.
Using leads with alligator clips, and magnets on each end of a battery, connect one lead from the battery + to the spring of the driver, then with the other lead connected to the battery -, just touch the grounding/negative ring on the spring side of the driver board. The emitter should light up (don’t be looking at it directly!).
Will do. I’ll also not bother removing the preexisting leads on the driver, I’ll just solder the original wires to the new leads, there’s plenty of room in the pill for it.
Did I understand what you said, that you’re planning to solder the leads that come with the driver to the leads that are already connected to the emitter? If so, that’s probably not the best idea, as you’d be doubling the lengths of the driver-to-emitter leads.
Also, if you do that, make sure you cover the point where you join the leads with something non-conductive, like kapton tape or electrical tape.
Preferably, just solder the leads that come attached to the driver directly to the emitter, I think.
The leads that come soldered to the Nanjg 17mm drivers are cut as if they were going to be installed in a Convoy M1/M2 host, they’re about 40% the length they actually need to be to reach the star in an X9.
I’ve got some 22ga Silicone covered wire and always replace the wiring with this. I put the driver in with the wires longer than they need to be, then use Arctic Silver 5 or Arctic Silver Thermal Adhesive to put the star on the pill and cut the wires to fit, no slack, no twisted up mess of wires inside the pill. If I need to remove the driver at any point, I do so by de-soldering the emitter. In this way, I have the shortest wires possible and know without a doubt the solder points are solid and all is well.
Even my Solarforce S2200 has these wires, as does the K3, and every other light I have almost without exception.
Have you verified that the red wire is indeed connect to the led positive? I've received lights that had the red being used for negative and black for positive. I even had a light that had the emitter it base backwards. So the negative sign on the base was actually connected to the positive of the emitter.
Your comment did give me pause, as assuming that the emitter would work both ways is something that I would do (I know that it isn’t how they work, a diode is a diode is a diode). I did go back and recheck, but the + lead on the star is to the + location on the driver and vice-versa for the - lead. I think that I must have nicked one of the chips flanking the - lead on the driver and it just fizzled out after playing with it enough times. That or cooked one or more chips. I’m using a Hecacon 1002, set to 700 as I though quickly melting the solder would be better than slowly flowing more heat into the board. Either way, this thing is dead. I still have a P60 host for my EDC and Monday I should be getting a Small Sun T620 in the mail, so I’ll still have a throw-ey toy for when I walk the dog at night until the new driver comes.
You would have to ruin all 8 chips to make the driver not work. Each one contributes 350mA of current and they are connected in parallel. If you wanted to by pass the MCU, you could touch positive to the right leg (your right looking at the chip) of any one of the chips and all will turn on.
The emitter could be on the base back words. You could connect a cell directly to it to see if it works. Just use a partially charged cell and just touch the pad long enough to see it works.