MaxFlex Problem

[dmcalpin]

Anybody had a problem with a MaxFlex ?

I've got a MaxFlex in the last month from Cutter which doesn't work. I've never got any light comming out of the LED when it's attached.

I'm using a 6V battery from a Marwi Nightpro. Voltage from the VIN+ to GND is 6.6 volts while voltage from the + and - on the triple Cree XR-E LED PCB is 5.1 volts. Something is happening to the voltage but if the Maxflex is boosting shouldn't it be around the forward voltage of the LEDS ie 10 volts or more ? The LED is fine as it works with a Luxdrive 3021DE I have connected to a 14.4V battery.

I've refered this to Mark at Cutter who sent the problem on to George at Taskled. Taskled suggest that I have damaged the MaxFlex by "having 'soft wired' the board, you have most likely blown a component".

I soft wired(ie tight wire loop around the connection holes without soldering) and taped the connections with gaffer before connection to the power source so I can guarantee that there was no damage by me.


In the last year I've built several single and triple CREE XR-E lights with Luxdrive models 2008B, 2009SHO, 3021DE and never had a problem. I've also been building Halogen lights for the past 20 years so I think it's krap that the explanation is that I've damaged the board.

Anybody else have problems with a MaxFlex or further suggestions.


Cheers
Dave

 

[nightrider]

My nFlex just died as well... not sure why, but I think on side of the switch might have shorted to ground. One big issue with the MaxFlex (and most boost type devices) if I understand correctly, is if you fire them up while not being connected to a load is bad news... so yeah, if your connector to the LEDs was intermittent, that could surely fry it.

 

[georges80]

Well, given that I test ALL drivers I ship. If they don't work on initial power up when you receive them that kind makes it tough to assume they were shipped defective...

I suppose it's possible they were damaged in transit - but that's out of my hands since when I ship they are wrapped in multiple layers of antistatic bubble and well protected.

maxFlex is open circuit protected, but having an intermittent on the load is never a good thing since it will cause high current spikes when the load is applied/removed.

These drivers (as are all of them that are sold as bare circuit boards) require careful construction technique and attention to detail. I can certainly understand that it is possible for a driver to fail in the initial short term use due to what is called "infant mortality" - i.e. component failure or assembly failure that occurs - but that is pretty rare.

I have seen some 'amazing' construction techniques including folk that have drilled out the thruholes to allow large wires to be soldered in (of course destroying the connections that are thru the pads). I've seen folk that claim boards failed and later admit - "oh yeah, a wire came loose and there was a spark". I've seen more than a few bad solder jobs - burnt off traces, components that have been desoldered accidentally due to large soldering tips etc etc.

All I can say, is I test all boards, often several times prior to shipping them out which is more than most drivers suppliers do.

I'm sorry to hear you're unhappy with the answer that I gave (regarding what you call the "soft wiring") - but in my 20+ years as an engineer I would rather solder a connection and then inspect it rather than tie wires through a hole.

And I'll leave you with a couple of pictures to look at and then you tell me what happens if you tie a wire through the hole and the wire rubs through the soldermask to the underlying copper.

Also on the 2nd picture where a wire strand looks like it shorting against the inductor.

Did those defects in the above pics possibly damage the driver? I don't know, but hopefully you can see how I would question the construction technique.

I do stand behind my drivers and have repaired more than a few that folk have sent back. Some are beyond repair, especially the ones where the soldering iron was used as a medieval torture device

cheers,
george.

 

[nightrider]

George. Didn't mean to sound if I was chiming in to knock your drivers. I think they are great and have sent many light builders your way! I just happen to break my nFlex yesterday! I think I messed up my old nFlex by shorting a screw (that was grounded to the body and negative power) against one side of the function switch. Now the light is just on full all the time. It had worked fine for well over a year until I hosed it. Not your fault.

Yeah I can see how the wiring above would not be good thing.

 

[georges80]

George. Didn't mean to sound if I was chiming in to knock your drivers. I think they are great and have sent many light builders your way! I just happen to break my nFlex yesterday! I think I messed up my old nFlex by shorting a screw (that was grounded to the body and negative power) against one side of the function switch. Now the light is just on full all the time. It had worked fine for well over a year until I hosed it. Not your fault.

Yeah I can see how the wiring above would not be good thing.

No worries - I don't have any issues with the OP or your comment and did see you figured your nFlex died due to a short in the housing.

It's a reality of life that the drivers can be broken - as long as we understand how they can be broken and what not to do next time, we all learn.

maxFlex did originally have a design bug in some battery/led combinations and when I realized it was my fault I pulled it from sale and and folk that asked for refunds I took care of them. I eventually figured out what was wrong and reworked boards that folk sent back to me and also reworked the remaining boards I had on hand. The bug was documented on CPF and basically involved removing a capacitor and reworking a cap (the blue on in the pics) to stop the uC from crashing. The switcher core has been solid and I've yet to see someone pop the switcher chip (yet...).

cheers,
george.

 

[chris_m]

folk that have drilled out the thruholes to allow large wires to be soldered in (of course destroying the connections that are thru the pads).

That is indeed a silly thing to do, but I can kind of understand why people end up doing that, given the holes are too small for the wires I'd choose to use for transmitting power (16 0.2), so I end up having to remove strands from my wires instead. I'd certainly prefer the power connection holes to be a little larger.