Thanks for the suggestion. Is there a driver like that in a 30mm size though?
Not that I know of.
If you know how to stack 7135s then you could get a smaller mountain electronics driver and add the chips to get the current you want.
If you type in google “graph 3.27*5-x**(.015*5.01.01) and 2**(.136*(x/1)+6.236)” (without the quotations) it will graph the functions. Then find the point where the two curves intersect.
Using 7135 chips is not really a solution here, IMO. The supply voltage and emitter voltages would still have to be matched fairly closely. If they aren’t matched the chips could burn up. If they are matched well then you just end up with “mediocre” output all the time.
Modding with direct drive setups has advantages. It is fun to plan, like the unusual setups proposed in this thread, it is electrically simple with fewer things to break, and it usually results in the highest performance.
OK, I got it now, Thanks.
And yeah I don’t think the other driver pointed out will work. I tried to build a 4xMT-G2 using that type of driver once and I don’t think it worked out right. Plus any way you go the massive heat is going to quickly reduce lumens no matter what you do. The point would be to have something that can blast out massive lumens if even just for a short while. It’s not like this is a “practical” light anyway. LOL Just horsing around to see how far I can take it.
The other option I was thinking of was a large 12x with 4x26650’s I know the wiring on that would work out fine, but I would have to buy that light and I already have this 5x32650. Right now it’s set up with XM-L2’s using 15 emitters and a FET driver. It’s pretty darn bright already. But it would be fun to see what it would do with 15xXHP-50’s. It’s a side switch light with no tailcap switch and I could get rid of the spring on the driver too. I’ve melted a few springs in my day making these crazy lights.
So you have 18 spots for XHP50s? A 5s Li ion with 2s9p 6v XHP50 would result in 47A from the cells, or 5.2A per XHP50. That current is doable for short times from the best 26650 cells. This will pretty much pull as much as the cells are willing to give, without overdriving the emitters. Or you could make it 2s8p or 2s7p XHP50s.
graph 4.2*5-x**(.026*5.01.01) and 2**(.136*(x/9)+6.236)
Tom E, I want to link to your 16x XHP50 mod because its the biggest one I’ve seen yet: What did you mod today? - #2517 by Tom_E
I think this has a 2s3p 18650 power source and 16p 6V XHP50 emitters. My tool predicts 47A will flow from the power source, or 2.9A per XHP50, which is consistent with the measured ~34,000 lumens.
graph 4.2*2-x**(.026*2/3.01.01) and 1**(.136*(x/16)+6.236)
No, the driver works with 2S and will run a 6v LED just fine.
Max output for most of the battery life.
Buck driver is also an option.
Ok, but we are talking about 5s here…
Yeah it’s 18x; 2s9p would be crazy and it looks like it would work. I kind of thought it would with that many emitters. There would almost have to be enough battery sag that it wouldn’t fry them. There is no way to run it long without something melting down but for short bursts it would be amazing.
Thanks for the help, that’s a cool tool.
Wow, that's cool! But how does it model the wiring? I would think the LED wiring is very critical, not to mention the FET, cell resistance, etc.
I'm expecting to break 40K though with VTC5A's and a UCLp lens 
.
True, I would go with a buck driver in this case…
The wire, FET, and cell resistances are in the first function (like .026.01.01). Of course, the numbers I have in there are estimates. The real value of the tool is being able to estimate how much of an increase you will get by changing things. For example if you change to VTC5As, which have, say, .02 instead of the 30Q’s .026 ohms, the total current will increase to about 52A from 47A.
I’m all for buck drivers, but it would need to pass 700+ watts to compete with the FET setup.
The problem with 7135’s(or any linear driver really) and series cells is the voltage change from charged to discharged is much too great for them to regulate well. With 3S cells the differential is 12.6V down to ~9V. If you pick LEDs with a maximum Vf sum of ~1V less than the max to protect the 7135’s then driver will be out of regulation from 11.6V down to ~9V. For series LEDs it seems to me that either boost or buck is the way to go.
Only realistic option is FET only. I'd use an SIR404DP, just a matter of voltage and amps using LIPO's - I got no experience with them, but thought high amp drones use LIPOs? Without buck or boost though, you are totally dependent on the performance of the cells.
I’ll try that FET I have some around.
The power density of RC LiPO/Li-ion packs can be much higher than the cylindrical cells we use. But the energy density is generally lower and they just don’t fit well in flashlights made for cylindrical cells.
If the light is going to use 700+ watts then it will reach a few hundred degrees C within seconds making it pretty much useless.
Almost 800W if it hits the 5.2A per goal. :smiling_imp:
Bwaaahaaaaahaaaa, must have more power!! :disappointed: