Is it possible to run two XML2’s in parallel from this driver http://www.fasttech.com/products/1612/10003372/1250101-v10-2745v-xml-t6-2-group-35-mode-led-driver-board?
It’s possible but I am not sure what you want to achieve?
Two under driven XMLs is what you will probably get…
these drivers are really not testing out so well, to drive a pair of xm-l’s in parallel at maximum rating, you need to supply 6a, I dont believe anyone has seen 5a out of one of these drivers, even when connected to a power supply, I assume your planning on a parallel battery set up to give a decent run time but unless space is an issue a master slave nanjg set up would give better results.
I also love the original versions of these from ios, but the 5a versions don’t seem to deliver unfortunately.
I had planned to make a headlamp for caving. I wanted to use 2 leds one flood and one spot. Driving the leds to their max output isn’t at the top of my list so that shouldn’t be a big issue. This was my thinking with the driver. Please correct me if I’m wrong as I’m new to all of this. If that driver actually puts out 4000mA thats 2000mA per led on high. If I remember correctly the very useful chart done by Match showed an XML2 puts out over 700 lumen at 2000mA. With both spot and flood leds running at the same time that’s a lot of light.
I have one of those drivers, and it isn’t that great. I had a difficult time getting up to 5A. If you are looking for around 2A or so per emitter, then it may work fine. I also was not a fan of the mode arrangement.
I am in the process of testing the new LightMalls linear driver which has relatively simple modes ( moon, L, M, H ) and should put out a similar current. it’s available for order, but you might want to wait until I at least power it up.
it should do the job, if it gets up to four amp, that’s the thing really, on an xm-l I got 3.7a from the 4a ios driver, I got only 3.3a from the same driver with an xm-l2.
I used a great 5a driver in my mt-g2 build but it does mean using a 2s cell set up as its 8.4v plus, it does however give a wicked low low, good medium and stupid high. you also get good colour rendering and acceptable throw from the one die. tir’s do appear to be available for mt-g2 but it needs a little google fu to find them, there are also 35mm ledil reflectors available for the mt-g2 in a few different beam angles which I think would fit the easy to led housings. let me know if you want any links I can provide
It’ll be a while before I make the final decision on all the parts I plan to get. So other options for the driver would be great.
here’s the driver I used.
bare in mind the bottom board is purely a contact board.
here are the mt-g lenses and cutter also supply mt-g2’s mounted on 20mm stars, ios also supplies them with copper stars if you wish.
http://www.cutter.com.au/products.php?cat=MTG+Single+Optics
you’d have to google the easy to led housings sorry.
Thanks for the help guys.
no worries :bigsmile:
I’ve never tried driving LEDs in parallel from the one driver - I always assumed it wouldn’t work properly due to variations in the forward voltage. How will you ensure that each emitter gets approximately half the current output by the driver?
magic lol
I’ve seen a couple other projects similar to mine where forard voltage didn’t seem to be an issue. I guess if one driver does’t work I will try the master slave thing although I’m new to this and that looks kinda hard.
I’ll be interested to see how it turns out. I would have thought that the negative temperature coefficient of the forward voltage will end up causing one emitter to hog more and more of the current as it gets hotter - until it burns out. Good luck!
As long as they are mounted on the same heat sink, the thermal runaway effect is mitigated somewhat.
I’ve driven up to five XM-L in parallel at close to 3A each with no short term issues. I’ll admit, I haven’t extensively used the light in question, but I have run it until it got quite hot and everything is fine so far.
As for this 2xXM-L2, I think driving at 2A each will be OK, one may bet more than the other, but it should not get overly imbalanced to the point where one LED is even close to 3A.
Negative temperature coefficient, runaway effect, I may be in over my head lol. I’m amazed at how much I’ve learned from different posts on here but clearly there is still a lot I don’t understand.
Just to chime in about the housing (s ) that gords was mentioning: http://easy2led.com/ — they’ve got 2 sizes. I’ve built headlamps out of the 20mm ones, and they’re nice and light/compact; the only downside is that there’s not much thermal mass/exterior area for higher draw applications. For really rough caving, I suspect they probably wouldn’t be waterproof enough, but I could be wrong. There was a flurry of builds with this housing over on mtbr.com in the lights/diy section a little while ago, so you could draw some inspiration there.
I’ve checked out some of the stuff on mtbr. There are some very cool builds there.
I’ve posted my initial test results of the custom linear driver from Lightmalls.
As I mentioned, it is available, however there are a couple of issues I’d like fixed before recommending it for purchase.
the only time I’ve ever heard of a real (as opposed to theoretical) problem running LEDs in parallel was when someone paired two LEDs (XP-E and XP-G2?) with very different Vfs, which resulted in the LED with the lower Vf getting far more current than the others, but still within spec. I’ve even built a mixed parallel XP-E2/XP-G2 triple star for a friend and by all accounts that’s been working great in heavy use (skiing, snowshoeing and kayaking) in an Easy2LED housing for the last 6mths or so.
Personally, I’d go with a mobydrv driver from DrJones so you can program the light levels you want for your application and have a lot more freedom in the switch you choose. If you’re set on 2A for each LED (personally I think 1.5A per LED would be more than enough) then he’ll most likely be happy to stack some extra 7135 chips for you if you ask him nicely. Going with a 3.7V battery will give you a whole lot more flexibility with regards to runtime vs. weight (especially if you have the pack on your helmet), plus it will be easy to carry a spare cell as a backup.