MaxFlex shifting from high to low on it´s self

[Jopi]

Hello,

I just finished building to bicycle lights. Both use MaxFlex as drivers, and both have 4 XPEs. Batteries are twin cell li-ions.

First one works as it should, but the second one doesn´t. The second one shifts inconstantly from "high drive level" to low. The driver is assembled per the manual: the input ground and led ground soldered to one ground "spot" in MaxFlex, the STAT led ground and switch ground to the other. In the first "working as it should" lamp I didn´t use STAT led and, if I remember correctly, soldered switch and battery ground to the other and the led ground to the other "ground spot" in MaxFlex.

Have I understood the manual incorrectly? The switch has longer cable than suggested (about 10 inch) but it´s the same in both lamps. Could that cause the problem.

I´m little desperate about this as the assembly consisted a vast amount of epoxy and taking the lamp apart wont be easy.

Any thoughts appreciated.

 

[VidPro]

it is not a thermal problem, as UI would crash completly, or mosfet would be toast, which you wouldnt get anything.
and he specifically states that if the switch isnt grounded and connected correctally that is will cause that specific problem.

the only other thing i could think of is if the settings arent set right?
Previous level
force Level
max drive level
or something. (see below)

did yours come with OS10 or is it the windows mobile version the sofisitcation of his drivers are a brain fry, hang in there.

when it "shifts" is it the SAME kind of shift that would occur if the battery came undone, because i wouldnt want to leave out the possibility of the drive current triggering any (over current) protection on the battery (momentarily). so check that , cause its easyiest. a quick burp from the battery everytime it goes to full high.

 

[rizky_p]

you batter ask George @taskled about this.

 

[45/70]

I'm probably not going to be much help here. I've never dealt with a MaxFlex, only bbFlex's (MR Bulk version of bFlex). I do know that they are finicky about having a proper ground.

The other thing I can think of, is I remember George telling me that it was important to run the switch wires well away from the LED leads, on a hipFlex project that I'm working on. The signals can get mixed up if you don't. He said it was actually possible to run wire lengths of up to 12 inches, if you kept the switch wires away from the rest.

As I'm going to be using the status LED on my hipFlex project, I hope that isn't the problem! If wired correctly with the resistor and all, it shouldn't be, I hope.

Dave

 

[Jopi]

Thanks for the replys, guys!

VidPro: It doesnt shut down, it just changes to lower drive level. For eg when set to UIB it changes the drive level from L5 to L2 without any user input. If you change it back to L5 it changes back to L2 sometimes after 1 sec, other times after 20 sec.

45/70: The switch wires run away from the led wires so that cant be the problem.

What just drives me nuts is that before I glued the heatsink to the housing (and before I glued everythint together...) I tried the maxflex and it didnt do this.

Oh well, maybe it´s time to pick up the hammer.