How to find which 7135 isn't working?

Flashlight Modding and DIY Parts
1

I built my first driver from scratch, a 15mm 5x7135 from MTN.

First - holy crap. I will gladly continue to pay for fully built drivers in the future.

Anyways, I got it together with lots of patience and no real issues. I tested it using very low input power (3.1v, 0.07A) and it made light, which was a huge relief/success.

When I powered it up with a battery though, I only got 1.38A on high.

4 x 0.35 = 1.4

Hmm, I think one of the 7135s isn’t contributing.

How do I find the slacker 7135 that isn’t doing his share? (yes, I realize it is almost certainly an installation error on my part, but from the top everything appears to be ok)

2

All the connections are in parallel so there’s no way to isolate one other than by disconnecting each one in turn. This could be done by removing them one at a time or by cutting the pwm trace to each one in turn and re touching with solder. It’s a pain no matter how you do it.

3

I’d start by retouching each and every connection with an iron. It wouldn’t be a solder bridge or none would work. There’s also the possibility that your meter wires are attenuating the measurement, changing to thicker, shorter probe leads or a clamp meter. It surprised me how much effect it has, easily enough to bury one chips worth.

4

Dang, that’s not what I was hoping to hear at all. :frowning: I was really hoping I could probe one of the pins to check voltage at each one for example.

I have both a modded HF meter with 6inch 12gauge leads and a UNI-T DC clamp meter, so in this case (unfortunately) it is not a measurement error.

5

Maybe IR photo or temp probe to see which one stays cool? Cycle to a medium mode to get highest wattage dissapated.

6

Maybe not the answer you were hoping for, but 4x0,345 =1,38
Isnt it Just caused by transition resistance?
Since the difference is almost negligible

7

It’s a 5x 7135 driver…

8

Just stack a sixt one and you are good

I had a driver where one was messing up the MCU signal resulting in no light at low modes and too few light in higher modes

Turned out that it was the 7. from 8 I had unsoldered and soldered again after checking the output, took me a lot time as its a pain to unsolder those little bastards

After that I would now unsolder all heating em up too much to get em off quick and throw em away and solder new ones

9

So I did some more thinking/investigating later last night:

I measured 1.38A, and 4x0.35=1.40A.
But the 7135’s that I used were supposed to be the higher binned 0.38A units from MTN.
And 4x0.38=1.52A, which is more than just a measurement error off from the 1.38A I was seeing.

I noticed the battery was:

  1. only at 3.7V
  2. a 5-year old Trustfire

So I charged up a few 4-year old Trustfire 14500’s to 4.22V. The Trustfires are all I have at the moment since I only had one 14500 light until recently. I ordered both some high-drain Windyfire 600mAh and high-cap Sanyo 840mAh 14500’s last week actually.

With a 4.22V 4-year old Trustfire, I got 1.60A. Not 1.9A, but progress!

1.60A is more than 1.52A, so I no longer suspected that 1 of the 7135’s wasn’t working. Now I think the old Trustfires just can’t handle that kind of current.

I wired up an NCR18650B that I have laying around at 3.8V… drumroll… 1.90A on the dot.

Mystery solved – it was just a weak/old battery. As I mentioned, newer and better cells are already in the mail.