ianryeng Posted February 21, 2010 Share Posted February 21, 2010 After a quick search I verified that no one had mentioned this previously. My question: Why not run two of the cheap DX 320mA current sources in parallel to make a 640mA driver? Is there any reason not to? Superposition suggests that the led's would receive the output of both. Maybe a diode on the output of each before the node to prevent one from charging the other. My concern is that I have overlooked something and will destroy my led's... Why do I want to do this? I have the whites for my array and have a few of these 320mA current sources but am waiting for an indeterminate amount of time before my meanwells and royal blues get here. Just wanted something to run over the tank in the mean time. Any feedback would be appreciated. Thanks, Ian EDIT: updated values Link to comment
doctaq Posted February 23, 2010 Share Posted February 23, 2010 ive seen them do something similar to use smaller drivers to power like a p7 or mc-e, but that was for flashlights and was on a much smaller scale, id be interested in the answer too, you should move this or repost to the lighting forum to get more views. the only problem i can see here would be the lack of dimming. Link to comment
ianryeng Posted February 24, 2010 Author Share Posted February 24, 2010 Thanks for the response doctaq. I don't see why it shouldn't be a scalable solution and theoretically I saw no reason why it shouldn't work. I may just have to try and see what happens. The only thing I don't have is additional diodes to put on the output before they meet to prevent one from charging the other which is the only problem I could foresee. If I get a chance I will try it out tonight and post back. It is a temporary solution until I get my meanwell's (and currently don't know how long the lead time is...) Link to comment
phoenix Posted February 24, 2010 Share Posted February 24, 2010 id be interested in seeing this too actually, wonder if some of the more LED savvy people will comment? Link to comment
evilc66 Posted February 25, 2010 Share Posted February 25, 2010 They can be run in parallel for ~640mA (320mA each). Make sure you keep the number of LEDs within the voltage limit. You would want to check the output from the drivers, as they aren't always the most accurate. Link to comment
ianryeng Posted February 25, 2010 Author Share Posted February 25, 2010 Thanks evil From my understanding these drivers are typically over-rated anyway but I will be sure to check the current if/when I run them in parallel. Link to comment
ianryeng Posted February 26, 2010 Author Share Posted February 26, 2010 So, I wired two of these current sources together last night and just hooked them up. The array they are connected to currently is 8 XR-E CW (bought 20 and using 12 when I get my RB's for the actual light). That said it is technically under the rated range (# of led's) but the current sources seem happy. The two in parallel are putting out 557mA. I never measured the current out of just one source but I would be amazed if it was actually 320mA. Anyway, for the short run until I get the rest of my led's and drivers this definitely an adequate solution. I may even run three in parallel (I have them anyway) but the output is visibly brighter and will leave it alone for now. If I feel like it/ get around to it I may put three in a box with switches to enable them providing a discrete stepped dimming I will post pics shortly Link to comment
evilc66 Posted February 26, 2010 Share Posted February 26, 2010 Careful with the stepped dimming idea. You will need to turn off the AC side of the circuit, not the DC side. Switching off the DC side could cause damage as the inductor unloads. Link to comment
ianryeng Posted February 26, 2010 Author Share Posted February 26, 2010 Careful with the stepped dimming idea. You will need to turn off the AC side of the circuit, not the DC side. Switching off the DC side could cause damage as the inductor unloads. I would not have done that but I am glad that you mentioned it. I first tested the drivers with no load to verify that the silkscreen was correct before connecting the led's and you could tell the drivers were not happy. Link to comment
doctaq Posted March 10, 2010 Share Posted March 10, 2010 how important are the diodes here? would the drivers really charge each other? i am interested in copying your idea for a fw setup Link to comment
ianryeng Posted March 10, 2010 Author Share Posted March 10, 2010 how important are the diodes here? would the drivers really charge each other? i am interested in copying your idea for a fw setup Not as important as I initially expected. I do not experience any flickering and the drivers are staying cool. I am really happy with the setup (at least until I get my RB's and meanwells). I am considering putting it over my fw setup when I am done with it. It seems really bright considering its only ~16W. If you do implement it I hope you will document it I would like to see -Ian Link to comment
doctaq Posted March 10, 2010 Share Posted March 10, 2010 so youve tried it without diodes? because if you have i will be ordering soon Link to comment
ianryeng Posted March 10, 2010 Author Share Posted March 10, 2010 Yes, the pics above are two in parallel without any isolation driving 8 cool whites. It has been running since that post with no issues. I cannot say if there will be any ill effects in the long term but the drivers seem happy operating in this configuration providing a very constant current supply of 557mA and are cool to the touch. (You will get shocked if you touch them though lol) Link to comment
doctaq Posted March 18, 2010 Share Posted March 18, 2010 sry to beat this to death but how has it worked thus far? have you measured effeciency of this guy yet? Link to comment
ianryeng Posted March 18, 2010 Author Share Posted March 18, 2010 sry to beat this to death but how has it worked thus far? have you measured effeciency of this guy yet? They are still working great (even after I shorted it by accident one day... dont ask ) I have not measured efficiency, if I find the time tonight I will try to test it (I don't have a kill-a-watt or anything convenient to do it unfortunately, just a multimeter). When I do get a chance I will be sure to post the results. Link to comment
ianryeng Posted April 3, 2010 Author Share Posted April 3, 2010 Sorry for the extended delay, the low current range on my meter needed a new fuse... The details: Input (121.3V)(0.127A) = 15.41W Output (26.13V)(0.564A) = 14.73W Efficiency 14.73/15.41 = 0.9559 So, about 95.6% efficiency running two of these current sources in parallel... much better than I expected. Link to comment
Cheetos Posted April 3, 2010 Share Posted April 3, 2010 Sweet! I love threads like this. Link to comment
coolwaters Posted April 3, 2010 Share Posted April 3, 2010 that was my idea it was thinking of doing this about 2 years ago for better color chooses and cheapness. didnt come out cheap at all... but its great if u want to drive those generic LEDs that need over 1A of constant current. Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.