BLF A6 FET+7135 Light Troubleshooting and Mod thread

You could check resistance right on the star. If you make polarity right you should get a little light from the led. If nothing happens you may need to reflow it.

After the dome is off, the bond wires are extremely easy to break. They basically should not be touched at all, ever, or there is a large risk of breaking them. They’re hair-thin and made of soft metal.

Chances are that’s what happened, and it won’t work again without a new LED.

Finally got my new head today, most noticeable difference is the longer contact spring. So as usual did a wash down of the light with new head installed, for some reason before fully submerging it (standard personal water proof test) I took a look inside the reflector and there was water in it. Immediately took it out and dried it. Glass was fogging up already.

So yeah, learned how to open the head, tape on head and body, lay it real thick and nice then use hand powwah!...Left light on turbo to dry out the inner parts with the reflector off.

I tried seating the O-ring in front of the glass, however when I tighten the head the oring gets forced out as there is no groove holding the O-ring under pressure, just a flat shelf.

In the end I just applied some superlube to the lens shelf, oring and the portion of the reflector where the o-ring sits on. Seems pretty watertight for now but I do not recommend submerging for prolonged periods.

Despite above issues, this is still my favourite EDC currently and my brightest single cell light so far. Needless to say it gets the ooohs and ahhhs from the uninitiated around here

Lucky you!

The reflector is needed to press the ledboard against the body, which gives essential heatsinking. With the reflector removed the board is not pressed down any more and can come loose. With the light on Turbo in this situation there is a significant chance the board will overheat and the led will fry. This has not happened in your light luckily :-)

Wow it didn't occur to me at all as the LED board appears pretty solidly and accurately fixed to the pill though...I can even see traces of some sort of thermal compound under the cut sections of the board. In fact, I tried wiggling the board with tweezers and it didn't move one bit.

@xzel87,

I am surprised you did not damage the led bond wires with all your tinkering with the reflector. The reflector hole at bottom holds a small plastic piece that “centers” the led. That’s liable to damage the led bond wires by the rotation of reflector when you screw in the head.

[edit]
See the TK’s post above.

If you want the o-ring to sit on the shelf you need a slighter bigger diameter o-ring.

@blightsam

I am aware of the dedoming risk thanks to the OP of this thread. However, it was never an issue with my defective head and new head. I took out the plastic centering piece and placed it over the led board before screwing the head back on. Didn't feel like it snagged on anything, let alone dedoming.


@Pöbel


I don't have an assortment of orings at my disposal, in addition, as a buyer I shouldn't have to be worried about that issue in the first place as it really seems to be a QC and maybe waterproofing design issue.

I’m fairly sure the O-ring size too small problem is already on the list, it goes way back.

@hank

I was not aware of the oring issue, if that is true, then apparently it still wasn't solved by the manufacturer despite being made known to them? Regardless, it's not really a problem because I won't be dunking my light in real use anytime soon...if i have to there are more suitable ones.

Is the water sealing issue a problem using the light in the rain? Or is it more an issue when it gets dunked? For example dropping it into a puddle.

BLF-A6 driver repair.

In the process of installing a bleeder resistor for a lighted tailcap mod I managed (or it managed itself, dunno how it happened) to damage the driver of one of my BLF-A6 lights: in the 7 modes setting without memory, the first three modes became the same, quite bright, it looked like the 7135-chip did not respond to the MCU anymore and was always open. Today I spend some time (hours actually, I'm glad this is a hobby) repairing it. I took the driver out and clamped it in the vice, with the ledboard re-connected loose in the air for testing with a 18650 battery. First, I tapped all the solder joints on the driver with the soldering tip to make sure no cold solder joints were leftover. Did not help. Checked every component for shorts underneath, no luck.

Then I swapped the 7135 with another one from the spare parts box. Now something did happen: very weird behaviour: very dim light from the die (like: 1/1000th moonlight), and after tapping the battery connection (=switching on/off) a number of times suddenly full blast, then extremely dim again. I tried another 7135 from the box. Now an almost complete user interface, just the moon was much lower than stock. Then I tried a few other ones and mostly got this version of the UI with extremly low moon, and in the meantime I began to watch the lettering on the chips: 35A,35B,35C,35D,35E,38A,38N, what is the difference?. There are more 7135 versions than I already thought:

(I have a few more types actually, but could not find them back for the picture). Then I soldered the stock one (the one in the bottom right corner on the picture) back for a check, and it was indeed still broken: all 7135-modes were still max. I already decided to settle for the 35D because it had the most usable way too low moon, when I thought of digging into the complete drivers box for more variations of the 7135 and sure enough I found the same type of the stock BLF-A6 driver on an ancient 2x7135 1-mode driver from dx. Tried that one and presto: the UI is back to normal, the BLF-A6 apparently needs this specific chip.

Morale: I don't know if this has already found out before, but there's clear variations in how differently marked 7135-chips react to (at least) the BLF-A6 firmware

Nice detective work djozz. Who would've guessed.

So do you think that us folks with a misbehaving driver should get a 7135 chip of the “1452” type and swap it on the board? Would that solve the power off issue on mode cycling?

BTW great work, indeed.

Yes, unfortunately. Different types of 7135 chips respond very differently at the low end, which makes moon nearly impossible to do consistently. I suspect this is part of why so few budget lights include a true moon mode.

We expect too much from these 7135 chips. They are pretty impressive for what they do, very convenient and cheap.
High frequency PWM is not their thing though.

Most firmwares use a PWM frequency of around 7-9 kHz for Moon mode, BLF A6 included.
At 8 kHz, a PWM duty cycle of 1/255 gives us an ‘on’ pulse that is only about 0.5 microseconds long. But the 7135s need about 2 microseconds before they start to turn on and another 4 microseconds before they are stable. In those 6 microseconds the response is like an under-damped control system. We have a transient ‘on’ peak, then it cuts back and then finally stabilizes where it is supposed to be. What we see with the 8 kHz PWM Moon mode is just the transient peak at 2 us. The timing and amplitude of this peak probably depends on the exact sample of 7135, cell voltage, the emitter, wire/track lengths, etc.

The ones I measured were groups of 6 or 8 from standard 105C/D drivers. Some types may be better, some may be worse and individual samples of the same type probably vary as well.

The only way I see to get somewhat repeatable output from different sets of hardware is to slow the PWM. A lot. The smallest pulse width needs to be at least a few microseconds more than the 6 microseconds stabilization time. Lets say 2 microseconds more. So we start with a minimum pulse width of 8 microseconds. (Conservative design would select an order of magnitude larger, we do not have that luxury here.)

If we want to get this 2us margin at a PWM duty cycle of 1/255, it translates to a PWM frequency of 500 Hz (1/(8us * 255)).
Or, we can decide to not use a duty cycle lower than 4/255, then we can get away with a PWM frequency of around 2 kHz (4/(8us * 255)).
If we apply this rule to 8 kHz PWM we are stuck with a minimum duty cycle of 16/255.

Most people will, however, object to the lower PWM frequencies. We also do not get a very low Moon mode.
In the end proper current regulation is really the way to go, not PWM.

Example 7.2 kHz PWM, 32/255 duty cycle. Even at 32/255 about a third of the ‘on’ time is wasted/unstable:

Now, if we cut back to 4/255 duty cycle, all we are left with is part of the unstable transient peak:

Agreed. At the 2kHz speed you described, it looks like a strobe light to me. And proper regulation is nice to have.

Proper regulation may be difficult, since it requires more hardware and might use an extra pin. On BLF-A6 there is an unused pin… but probably no room for another component or two. I wonder if we could make it power the LED directly from the MCU. It’d probably require a resistor to adjust the current to the correct level, but otherwise I think it might be feasible if it’s only used for moon. Then there would be three power channels — moon, 7135, and FET.

The 7135 abuse method seems to work well enough most of the time, if people aren’t too picky about the exact output level. Like on the A6, it seems to range from 0.26 lm to 0.55 lm, possibly more. Pretty wide range, but they’re all reasonable for “moon mode” use.

Has anyone seen this before? It looks like the light settles into a mode, notice the flicker and then it is stable? Please disregard the auto exposure, can’t change that on the phone video recorder. I am going to send a note to Neal, I am afraid it will quit altogether. I have used Keeppower 35A and 30Q batteries, verified the switch is tight and cannot disassemble the head. :frowning:

[Flicker](<a) data-flickr-embed=“true” href=“https://www.flickr.com/photos/98509702@N05/21981754936/in/dateposted-public/” title=“20151006_202105_001_001”>

Matt

To me it appears to be a crook earth somewhere.

Very hard to do with a camera that cannot be set to manual exposure! And it happens so soon after mode change that the exposure does not get time to settle in.