I need some help troubleshooting this light....

I started with a KD 4x7135 V2 LED Driver 1520mA, no jumpers soldered, 9-mode, and a cree XR-E led, in a P-60 pill. This is in a Ultrafire WF-501a host.

This is the second light built using this driver. First one works great.

The new light flashes full brightness for a moment and then settles into the chosen mode, (5mA, 35, 100,) etc.

This drop-in preforms the same in other hosts i have. Other drop-ins perform correctly in this WF-501a host, so I’m leaning for the problem being in the driver.

I tried re-soldering all of the connections, that didn’t change anything, so what do you think?

If it’s not the driver, it’s probably over my head.

Anybody have an idea as to a fix?

Thanks everyone !


Parker, I'm not quite sure what to say about this one. How long is the flash and settling period?

Very quick, but it definitely goes to full brightness, perhaps 1/4 second. Almost like a camera flash.

I'd clean that board as good as you can then look for any small pieces of wire or carbon that could be causing some light bridging.

I've had a couple of 2800 mAh drivers from KD do the same thing. I never found an answer.

If everything looks clean, I'm not sure where else to look besides replacing parts.

Didn't some of the 4Sevens Quark lights suffer from what users called "pre-flash". This sounds similar to the issue you are descriging.

As far as I know, it was something to do with the driver but I don't know if it was "fixed" in later models.

A search for Quark pre-flash may net some results if you don't do any good in this thread.

Unintentionally I have just stumbled on this problem again with a KD 7135 driver. This time it is on a new light build that was working fine until I started playing around with the switch. It has started to flash when turning the light on but starts with the mode stored in memory. It changes modes fine once on. I have a gut feeling that my problem at least here is something to do with how fast the switch is clicking when operated from of somehow causing the driver to think it is changing mode but isnt, only being switched on if any of that makes sense.

Jumped the gun. I have just removed the pill from the light and powered it of my power supply to negate the switch and battery. Earth was hooked directly to the driver and powered up with the positive to the driver spring. The led flashed everytime it was powered up and then ran in the mode that it was last used on.

This driver was functioning fine up until today.

If R1 weren’t soldered correctly would that cause the cap to retain any voltage spikes?

It could be, especially since we know that the voltage can boost a little bit on those drivers.

Thanks guys. I'll have to investigate this R1 thingy.