MeanWell Question

[SkiFletch]

Hey guys, I was over at Saltyscotsman's house today trying to help him troubleshoot his failing LED array. I've built my own LED fixture and have always built my own drivers, so working with meanwells is a new thing for me. He's got a system custom built by someone with the the 48v dimmable drivers controlled with a reefkeeper lite. 3 drivers, one for white, two for blues and one of the blues is gone. When I arrived, found a short on the string, fixed it, but the array still wouldn't come on after plugging it back in. So I removed the string from the meanwell and just tested the output. Got zilch with the multimeter. So my question for those with more experience with these supplies than I, will the meanwells show voltage on their output even with no load connected? It wouldn't surprise me if the short damaged the driver, but I just want to make sure before Norm goes ordering a replacement if the problem lies elsewhere.

 

[blanden.adam]

I would doubt they would. They are constant current drivers, so they modulate their voltage based on resistance. Higher resistance, higher voltage within the range of the driver. Conversly, no resistance would mean no voltage (at least in theory by math, I've never physically tested it myself). On top of all that, because the drivers are controlled by the RKL, the current should drop to 0 without a potential applied to the input wires (0-10V I believe, with 0 being the minimum and10V being the max.)

 

[stephen77]

meanwells

meanwells

If remember correctly it will show 0 with no load . Easy check to find out what's going on. 1 hook up suspect driver to known working string and see if it works or 2 take one of the known working drivers and hook up to the suspect string. good luck !

 

[SkiFletch]

Yeah that's easier said than done i this setup... Norm's said this publicly and I agree that this setup is not of the highest quality. Nothing's labeled, no connectors, taped solder joints, sloppy soldering, cheap heatsink, just a mess. I detected one shorted solder joint to the heatsink (sloppy soldering), repaired that and verified no other shorts with it, but it still wont light. All the diodes check out fine with the multimeter, so the only other thing I can think of is a bad power supply. There's definitely control from the reefkeeper since both LED strings are controlled from the same output. Norm's ordering a new power supply, and if that doesn't work, I guess the best course of action would be to replace their cobbled together taped rats nest with real wire and crimp connectors with these special things called labels :lol:

 

[stephen77]

Wish you guys were closer. I'm sure we could sort it out in no time . Good luck

 

[der_wille_zur_macht]

If remember correctly it will show 0 with no load .

I don't not believe you, but that's the opposite from how most drivers function - essentially a driver is monitoring current, and if the current is lower than the nominal value it'll increase voltage in an attempt to hit the nominal target, up to the point where it hits it's own voltage limit. So on an open circuit (no current) you typically expect to see the nominal max voltage. Meanwell's documentation for this driver shows it operating in a constant current mode until it hits max voltage, at which point it behaves like a constant voltage supply at that voltage.

That said I've never checked for this behavior on this very driver, personally, so that's all just speculation on my part. I wouldn't be shocked if a driver was burnt out given the description of the fixture. I would be eager to try to test the LED string end to end, maybe with a few batteries or some other very low known-good high impedance voltage source.

Good luck getting this fixed.

 

[stephen77]

I wouldn't believe me either!!! I just pulled my cover off and pulled fuse from string and still had 42v. So your right should still have voltage as long as it has power and a dimming/on signal.

 

[SkiFletch]

So I'm not crazy?

 

[der_wille_zur_macht]

Well I'm not sure I would go THAT far.

At this point it sounds like a new driver would be reasonable. The catch is, have you actually fixed the problem that killed the driver in the first place?