disaster999 Posted May 27, 2015 Share Posted May 27, 2015 So Ive decided to upgrade my previous build by adding 2 Lumia 5.2, swapping all the drivers to Meanwell LDD 1000H drivers with the 5up boards, and switched out the Arduino controller to the Coralux Storm controller I got all the supplies from LEDgroupbuy and Coralux and tested out the lumia. I hooked up all the channels and turned the light on. The light lights up but I further tested out the light by individually testing each channel. Since some channels only support 700mah, and I have all 1000H drivers, I though I can limit the output of those channels from 255 (100%) to 180 (~70%). I hooked each channel individually and tested the light again. To my surprise, those channels with 700mah limit would barely light up. Their output from 0 to 180 stayed the same. Its when I start cranking up the output past 180, the LED would begin to brighten up all the way to 255. I tested the other 2 channels that can support 1500mah and they didnt have the same problem as the 700mah channels, but ramping them from 0-255 didnt feel smooth (smooth as in the brightness doesnt increase linearly) Im not sure if my thinking is correct, limiting the PWM signal to the driver to ~70%. or Should I of bought the appropriate drivers for each string? My light is powered by a Meanwell 48V 9.4A power supply, so I think i have plenty of juice to power the lights. Link to comment
Horerczy Posted May 27, 2015 Share Posted May 27, 2015 Don't use 1000h drivers on 700mA max channels. The high current pulse will fry the LEDs. Link to comment
disaster999 Posted May 28, 2015 Author Share Posted May 28, 2015 Weird I dont see any information on the data sheet that suggests high current pulses. The only thing they show are the ripple noise which is 350mV for the 1000H. Link to comment
Horerczy Posted May 28, 2015 Share Posted May 28, 2015 Ldd don't shift their current to dim. They change the amount of time they're on during their duty cycle. So no matter what a LDD-1000H will be pulsing 1000mA just the duration of the pulse changes. Link to comment
disaster999 Posted May 28, 2015 Author Share Posted May 28, 2015 WELP, seems like most of my LDD 1000H drivers are a waste and have to get new ones. Link to comment
jedimasterben Posted May 28, 2015 Share Posted May 28, 2015 Weird I dont see any information on the data sheet that suggests high current pulses. The only thing they show are the ripple noise which is 350mV for the 1000H. That is what PWM is all about Link to comment
disaster999 Posted May 29, 2015 Author Share Posted May 29, 2015 Just ordered new LDD 700H drivers, I have 5 1000H spare so I guess ill save them for some other projects or sell it really cheap. Link to comment
disaster999 Posted May 31, 2015 Author Share Posted May 31, 2015 Another question/problem ive encountered...I just finish upgrading my light and hooked it up. Im only running the whites and blue channel since the 700mah drivers isnt here yet. I have 3 LDD drivers driving 14 blue LED each and 2 LDD drivers driving 13 and 14 white LED. I have channel 1 for the whites and channel 2 for the blue. what I noticed is, when I have the light running for a while, one of the 3 blue string will turn itself and will not turn back on. I tried manually turning off the lights though the controller and back on again with no success. The only thing it will turn back on is manually power cycle the light, but even then, that blue string will turn itself of again. Im not sure whats wrong. I tried swapping out the drivers, didnt work. I tried changing socket location for that string, didnt work. I made sure each string is getting the same signal through the controller. I could try changing the channel to channel 3 and see if that works. Or, swapping the location of that one blue string with another. Getting tired of this thing ahha Link to comment
Benny314 Posted May 31, 2015 Share Posted May 31, 2015 As ever lets not over complicate things before we start. So the very basics of swapping out an LDD yields the same result so this rules out a duff LDD. Next thing to check will be your wiring (solder joints, terminal block connections etc) on the string that is going out. Sounds like it could be a bad connection/solder joint that is causing high resistance and getting hot, when it gets hot enough the resistance gets high enough to stop the flow of electricity turning the sting off. You then fiddle about changing drivers over and scratching your head long enough for it to cool down and then repeat the same overheat. It's unlikely to be a bad connection to the controller PWM feed as it sounds to consistent. If it was a signal problem it would either not work, or go out when the wire was knocked. 2 other channels working fine prove the signal is good and it's not a controller fault. Good luck. Link to comment
disaster999 Posted May 31, 2015 Author Share Posted May 31, 2015 Double and triple checked my connections, went over each joint with the soldering iron to make sure theres not cold solder. Even went and double checked each connection to make sure its not causing a short with the headsink anywhere. Couldnt find anything that indicates any obvious problem with that particular string. I went on an additional investigative work and checked the input and output voltage of the drivers. When I got the power supply I made sure the output was 48V, but there wasnt any load on it. The voltage dropped a little to 47.3. I checked each string's voltage, the whites are at round 38V and 41V since theres 13 and 14 respectively on each string. The blues are at 46.2, 44.7, and 43.6. This was a bit odd since each string has the same amount of LEDs. The string that reads 44.7V are the lumias + 4 other CREE Royal Blues, The other 2 strings all consists of 14 CREE Royal Blues. I did notice the string that turns itself off was running a bit high at 46.2V, 3 volts higher than an identical string running the same LEDs and pretty close to the input voltage. The spec sheet says the output voltage is always 2 volts lower than the input so I though that contributed to the problem. I bumped the power supply voltage to 50V but that didnt work. I did notice that when the string turned itself off, the output of the driver isnt supplying any voltage to the LEDs. Other strings, when I remove the output wires, reads the input voltage from the power supply. This makes me think that there must be something wrong with that particular string causing the driver to "trip" and go into protection mode. Im not really sure what else could do. The only thing I could try is reduce the # of LED that driver is driving and see if thats enough to sustain that string and not cause turn off. Link to comment
jedimasterben Posted May 31, 2015 Share Posted May 31, 2015 Just to be sure - all of your DC grounds are tied together, right? Link to comment
disaster999 Posted May 31, 2015 Author Share Posted May 31, 2015 All the ground from the power supply and each string of LED goes to through the 5up board. I don't have a ground from one string directly touching another. Link to comment
jedimasterben Posted May 31, 2015 Share Posted May 31, 2015 All the ground from the power supply and each string of LED goes to through the 5up board. I don't have a ground from one string directly touching another. And the Storm is properly grounded to it, as well? The 5up boards' grounds are just one ground separated into three terminals. Link to comment
disaster999 Posted May 31, 2015 Author Share Posted May 31, 2015 The storm gets its own power supply totally separate from the rest of the system Link to comment
jedimasterben Posted May 31, 2015 Share Posted May 31, 2015 The storm gets its own power supply totally separate from the rest of the system Not power, the PWM outputs. Link to comment
disaster999 Posted May 31, 2015 Author Share Posted May 31, 2015 I actually dont have the PWM outputs gounded to the 5up boards actually. Lemme give that a try and see if that helps. Link to comment
jedimasterben Posted May 31, 2015 Share Posted May 31, 2015 I actually dont have the PWM outputs gounded to the 5up boards actually. Lemme give that a try and see if that helps That is your problem, then. All DC grounds (including PWM grounds) need to be tied together or wonky things like this happen Link to comment
disaster999 Posted May 31, 2015 Author Share Posted May 31, 2015 Thanks Ben, I hope that fixes it. Ill report back tomorrow as I dont want to turn on my lights (almost midnight on my end) and wait 10-15 mins and see if grounding it actually worked. Link to comment
disaster999 Posted June 1, 2015 Author Share Posted June 1, 2015 Well that didnt seem to fix it. I got my 700H drivers today too and I installed it. None of other string work as well. The 5up boards have 5 strings, lets call them 1, 2, 3, 4, 5. I have the whites on 1, blues on 2 and 5, RGB on 3 and UV on 4. The only strings that works are 1 and 5. 2, 3, 4 seems to have the same problem as before. They would turn on, but turn off itself after a while. The only thing I can think of doing is to swap the 5up boards since I have 2 on my light. See if its the 5up boards thats causing this. Link to comment
jedimasterben Posted June 1, 2015 Share Posted June 1, 2015 Well, it might be the board, it's not outside of the realm of possibility. I've had one bad board myself. Link to comment
disaster999 Posted June 1, 2015 Author Share Posted June 1, 2015 Maybe it is the connection... I took the light off for the billionth time, strip the 5up board clean of wires, and rewired the problematic strings all over again. Went over each connection with the multimeter and made sure no 2 strings are shorting each other out. Making sure the 5up board isnt defective. After I was done, I lite the light up and let it run for 15 mins. Nothing turned itself off which was a relieve. Everything stayed on throughout the ramping down sunset stage. The true test begins when the light comes back on tomorrow and put through its paces when the light stays on for 10 hours. Link to comment
Benny314 Posted June 2, 2015 Share Posted June 2, 2015 This does sound like there was a wire or 2 with a bad connection if it appears fixed now. Wiring faults are so hard to run down, but hopefully your strip and rebuild has resolved the issues for you. Glad to hear it and good luck with your testing tomorrow. Link to comment
disaster999 Posted June 2, 2015 Author Share Posted June 2, 2015 Came back home to the light at full power. Seems like its finally fixed. Thanks for all the help Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.