Actually read a little further in that thread and came across this:
haha just went and found that link and you beat me to it.
So if the throw is about the same or even a little less than a dedomed XM class emitter, why are the big-name dedicated throwers using them? Just for the lumens bump? I don’t remember stock lights with throw ratings like the K70 until the 35 arrived
My opinion is that the XHP35 is a good emitter to use for a thrower because it is a good compromise between output and throw. Though it is a lot of trouble to drive it correctly because it’s 12V, and you can’t just use a FET driver even if you have 4s cells because it will die from the high current.
Big manufacturers are using it partly just because it’s new and will sell because it’s the latest thing.
good question, been wanting to know this as well. it would seem a dedomed xml2 and xpl should throw further than the xhp35 HI.
i still think dedomed xpg2 throws the best and is almost as bright as the dedomed xpl’s and xml2’s.
you will have a bigger hotspot for sure with the xhp 35’s i believe, and will be noticably brighter.
the throw maybe slightly less but all in all, i think it would be a good upgrade.
How about this question:
We know 4 cells in series with a FET will pop an XHP35, but what about 3 emitters in parallel? Or 4 emitters?
I did an estimation of this recently. It looks like even with 3 or 4 XHP35s in parallel, there is still probably too much current to each emitter. This is based on a rough extrapolation of the forward voltage from cree’s graph on their datapage and considering the voltage drop from the cell IRs and other circuit resistances. I estimate that an additional ~0.2 Ohms of resistance (like adding a coil of wire somewhere) would bring the current to around 2.5A to each emitter.
So, thin wires then? 22 or 24awg? Would that give enough resistance?
How else can we power 3 or 4 in parallel? A buck driver won’t give us enough current.
I agree adding some resistance, although at first glance not the most elegant solution, is a good and easy solution. A linear driver is another possible solution. In either case the driver or extra resistance would have to burn off a lot of power, so it is not super efficient. With a triple XHP35, (2.5A(3))^2(0.2Ohms)=11.25W dissipated in the extra resistance or linear driver. But, this is not huge compared to the ~100W going to the LEDs.
0.2 Ohms is actually a lot of resistance and one would need to think about how best to add it to the circuit. For example, 24awg wire has 84 mOhm per meter so it would take over 2m of this wire to add 0.2 Ohms.
If you are going to do this project I can do a more precise estimate of the amount of resistance you have to add. I have an XHP35 that I can measure the forward voltage of.
I am just trying to determine how feasible it is. I don’t want to drop a bunch of money on a fancy driver.
Sounds like I need to add 2meters of 24awg wire and use the worst batteries I can find 
4 (li-ion) batteries in series would be ~16.8V, and theoretically, that would be regardless of whether you put “x” emitters that are in parallel. That’s in theory though and doesn’t account for battery (voltage) sagging and any resistance from wires (v=ri, so if big resistance, the voltage drop across the wire would be higher. Then again, also, with ‘x’ emitters in parallel, the resistance on each path might also be slightly different.
There’re a lot of variables, so answering exactly what “would” happen vs. what “could” happen is very difficult.
I’m with you. A linear driver would probably burn up anyway unless it was very well heatsinked.
This sounds like a fun project; I will think a bit about the best way to add that resistance. Something like a homemade wire wound resistor. Like a bunch of thin wire encapsulated in some thermally conductive epoxy or something.
I suppose I could always just PWM a FET below max output, that’d be the cheapest. Don’t like FET-only drivers though, too much variance based on voltage. How much heat do you think a 7135 can handle with a big copper heatsink?
Yeah, good idea about just PWMing down a FET driver.
I imagine a 7135 could drop a lot of voltage if it was heatsinked well. But heatsinking 20 7135 chips (for 7.5A) seems difficult. You could also combine the two approaches; add some resistance somewhere else to lessen the load on the 7135 chips.
I may remember someone telling that 7135’s have an internal thermal stepdown…
I believe they do, but I imagine at some point they probably fry.
EasyB, I wouldn’t use that many 7135’s, basically a FET+4-6 set up. Low modes on 7135’s, and FET “turbo” that is limited by a bit of pwm trickery to keep the amps in range.
The hosts I would be looking at should have plenty of room for copper attached to the 7135’s for sinking.
So maybe this is the way to go? I could also just not charge the batteries all the way if all else fails.
Theoretically, 700mW is about max for a 7135 to dissapate but it needs up to 1000 mm^2 of connected trace area per chip to do that or a sink large enough to soak it up. I don’t think it’s feasible for 4s cells since max overhead is ~ 4V and 3s cells drops out right away. The problem is that the more cells you put in series the greater the voltage difference between charged and depleted. Best available option is probably the H6flex with 1,5A per led.
I too was wondering how a MTN MAX buck driver @ 2.5A would make out running three XHP in parallel with four Series cells.
Maybe a couple of CN5611 ?
They can do CC from 40 - 800 mA.
With these kinds of voltages their voltage drop is not a real problem i.m.o.