Can you direct drive 12V XHP70 with 16.8V input?

20 posts / 0 new
Last post
Bigchelis
Offline
Last seen: 2 months 1 week ago
Joined: 09/13/2011 - 22:53
Posts: 259
Location: San Jose, CA
Can you direct drive 12V XHP70 with 16.8V input?

I know the XHP70 12V can take 3 4.2V cells with ease but I want monster lumens

I have FET drivers and just want to know if I use a 4D Mag with 16.8V input will it kill it or give insane brightness

DB Custom
DB Custom's picture
Offline
Last seen: 2 hours 16 min ago
Joined: 01/13/2013 - 22:28
Posts: 20809
Location: Heart of Texas

Since the 12V variation is merely running 4 XM-L2’s in series, I would think pushing that much current through them would fry em nearly instantly. But since I’m not an electronics guru, I can’t say that for certain.

The other side of the coin is can the FET driver handle that much overhead? I would think something in the mix would go up in smoke, and that would be a fairly expensive experiment.

Monster lumens comes with multiple XHP-70’s. Try building a potent triple, or a quad. That should do the trick.

Good luck! I’ll keep an eye open for what you end up with…..

djozz
djozz's picture
Offline
Last seen: 1 week 7 hours ago
Joined: 09/07/2012 - 17:04
Posts: 18319
Location: Amsterdam

A XHP70 at 12V is basically 4 XM-L2 dies in series. If one XM-L2 (on copper) can survive direct drive on one li-ion cell, I see no principal reason why 4 in series would not run on 4 cells in series, I would expect a similar current. Only one thing I can think of is that the heatsinking of a XHP70 is worse than of 4 separate XM-L2's.

Bigchelis
Offline
Last seen: 2 months 1 week ago
Joined: 09/13/2011 - 22:53
Posts: 259
Location: San Jose, CA

I just built a XHP70 6V variant in Direct Drive for a family member.

With 2 IMR 26650’s (8.4V input) the performance is amazing in DD with 6.5A at tailcap. Actually focuses good with slight yellow in center with Mag SMO LED reflector.

So, my thinking was if the 6V XHP70 took the 8.4V perfectly why cant the 12V take the 16.8V perfectly?

Thanks,
bigC

DB Custom
DB Custom's picture
Offline
Last seen: 2 hours 16 min ago
Joined: 01/13/2013 - 22:28
Posts: 20809
Location: Heart of Texas

But for the driver, this may be true. In 12V direct drive it probably would run on 4 cells, but you’re protecting a 5V MCU with a Zener diode, and the overhead here is probably too much for it. Give it a shot and see.

My first 6V XHP-70 on a FET Zener modded driver hit 12.15A for 6200 lumens. I have one in an X6 now making 5500 lumens. And I have 3 in a BTU Shocker making 14,455 lumens.

The XHP-50 works brilliantly in a Convoy C8 with the orange peel reflector. Right at 3000 lumens. No dark spot in the middle. And the Minnie M from Ledil does a very nice job of handling the XHP-70 in an X6 light as well.

Give it a shot, see what happens…. :bigsmile:

Bigchelis
Offline
Last seen: 2 months 1 week ago
Joined: 09/13/2011 - 22:53
Posts: 259
Location: San Jose, CA

DBCstm wrote:
But for the driver, this may be true. In 12V direct drive it probably would run on 4 cells, but you’re protecting a 5V MCU with a Zener diode, and the overhead here is probably too much for it. Give it a shot and see.

My first 6V XHP-70 on a FET Zener modded driver hit 12.15A for 6200 lumens. I have one in an X6 now making 5500 lumens. And I have 3 in a BTU Shocker making 14,455 lumens.

The XHP-50 works brilliantly in a Convoy C8 with the orange peel reflector. Right at 3000 lumens. No dark spot in the middle. And the Minnie M from Ledil does a very nice job of handling the XHP-70 in an X6 light as well.

Give it a shot, see what happens…. :bigsmile:

Wow, that’s a lot of valuable information.

Since I have FET zenor DD drivers at my disposal worse case is I go to low mode

foolioGrimz
foolioGrimz's picture
Offline
Last seen: 6 years 8 months ago
Joined: 10/09/2015 - 05:01
Posts: 209
Location: Texas

I know this is an old thread but.. what about true direct drive from the cells?

dchomak
Offline
Last seen: 4 years 7 months ago
Joined: 03/17/2012 - 20:10
Posts: 4122
Location: Connecticut

This is what I understand.
A single XM-L on copper can survive DD from a single 18650 because there is always some resistance in the circuit (wiring, springs, switch and a very small loss thru the FET) The sum of of resistances limits the actual maximum current that is allowed in the circuit to the point that the XM-L is just running on the ragged edge. In fact in some DD lights one might actually purposely introduce a little resistance to limit the absolute maximum current so as to allow for a margin of error.

It’s true that an XHP-70 is 4 XM-L’s on the same die, BUT when configured in 12V the “effective” resistance in the path to the LED (wiring, springs, switch and a very small loss thru the FET) is 1/4th of what it is in a single XM-L.
What I mean by that is percentage wise, the path resistance is lower in a 16.8V system than a 4.2V system.

BTW, that is why AC electricity is transmitted at very high voltages over long distances and stepped down to lower voltages closer to the point of useage. Less line loss.

Andrew2007
Andrew2007's picture
Offline
Last seen: 6 years 4 weeks ago
Joined: 11/04/2015 - 16:07
Posts: 590
Location: Northumberland, England

Am I thinking straight?

4 xml’s, when running just one you may draw for example 2.5amps depending on the supply.

But if you have 4 in series it could be 4 × 2.5 = 10amps and all of this 10 amps will be going through each xml. Shocked

Are the xhp70’s beefed up to some extent so that they have no problem drawing 10 amps.

Is that why when connecting multiple emitters they’re normally connected in parallel and not series so you’ll have a modest 2.5amps through each one and not 10amps?

But the overall current will depend on how good the cells are?

dchomak
Offline
Last seen: 4 years 7 months ago
Joined: 03/17/2012 - 20:10
Posts: 4122
Location: Connecticut

Andrew2007 wrote:
Am I thinking straight?

4 xml’s, when running just one you may draw for example 2.5amps depending on the supply.

But if you have 4 in series it could be 4 × 2.5 = 10amps and all of this 10 amps will be going through each xml. Shocked

Are the xhp70’s beefed up to some extent so that they have no problem drawing 10 amps.

Is that why when connecting multiple emitters they’re normally connected in parallel and not series so you’ll have a modest 2.5amps through each one and not 10amps?

But the overall current will depend on how good the cells are?

An XHP-70 configured for 12V is like 4 XM-L in series each LED would see the same 2.5A each for a total of 2.5A.
Running 4 XM-L’s in parallel, each seeing 2.5A would draw a total of 10A for the 4.

foolioGrimz
foolioGrimz's picture
Offline
Last seen: 6 years 8 months ago
Joined: 10/09/2015 - 05:01
Posts: 209
Location: Texas

You just hurt my head. Lol. But in other words the same amount of heat is being generated since regardless, the same amount of energy is being expended? Is there a difference performance wise 12 vs 6 volt configuration if the total watts is the same?

gamezawy
gamezawy's picture
Offline
Last seen: 7 months 1 week ago
Joined: 10/08/2013 - 07:34
Posts: 505
Location: Egypt

I have a light project and i want to direct drive a xhp70 with 4 cells with 16.8 input no driver with normal cells not IMR, also same question about the XHP70.2 So how both leds will behave sure they will be on DTP MCPCB

texas shooter
Offline
Last seen: 51 min 25 sec ago
Joined: 08/26/2012 - 02:14
Posts: 2041
Location: Texas
gamezawy wrote:
I have a light project and i want to direct drive a xhp70 with 4 cells with 16.8 input no driver with normal cells not IMR, also same question about the XHP70.2 So how both leds will behave sure they will be on DTP MCPCB

1st on these two LEDS the XHP-70 runs a little higher LED Vf than the XHP-70.2. So the XHP-70.2 will allow more current on identical set ups. Your real issue will be the quality of the cell and how much voltage sag vs current through the LED.

rngwn
rngwn's picture
Offline
Last seen: 2 months 1 week ago
Joined: 05/17/2018 - 03:44
Posts: 441
Location: 暖武里,Таиланд

Actually, you can go ahead and look at the LED datasheet and see how these part behave at cetain voltages.

But I’d predict that the LED would be overdriven by a factor of two (or more) well before 16.8v (unless the 16.8v source is current-limited (or have significant internal resistance) that the voltage will drop to reasonable level when plugged into a big load).

At that point your LED will cook very nicely, and if they don’t (or not immediately) you might be looking at 100w+ of heat that you need to dissipate with your flashlight body. That’s also a burning/scalding risk.

gamezawy
gamezawy's picture
Offline
Last seen: 7 months 1 week ago
Joined: 10/08/2013 - 07:34
Posts: 505
Location: Egypt

Ok thank you for your inputs but allow me to ask the question in other way

Have any one tried to direct a XHP70 AND XHP70.2 on DTP MCBCP With normal 18650 cells not IMR 4 in series that gives 16.8V, And how they behaved in good heatsinking, i mean a actual experience

rngwn
rngwn's picture
Offline
Last seen: 2 months 1 week ago
Joined: 05/17/2018 - 03:44
Posts: 441
Location: 暖武里,Таиланд

I have never done this, but you might have a better luck with 18650 LiFePo4’s which will produce 12.8v at 4s.

The overcurrent should be more survivable for your LEDs.

Lexel
Lexel's picture
Offline
Last seen: 2 years 2 months ago
Joined: 11/01/2016 - 08:00
Posts: 5895
Location: Germany

both XHP70 and 70.2 are fine on DTP especially with non high drain cells

Sirstinky
Offline
Last seen: 6 hours 57 min ago
Joined: 11/02/2018 - 17:07
Posts: 1790
Location: Vancouver, WA

I’ve build a few xhp70.2 lights. One 12V, one 6V. The 70.2 has been tested on this forum to easily take 9A. Look at Texas Avengers test and you can see how it behaves at the various loads. That will give you an idea of how m8ch power you will be dissipating at those currents/voltages. He ran one up to 20A. Figure if the fv of the xhp70.2 12v (according to Cree) is 11.2v, at the max drive current 2.4A that’s around 30W. Remember, as the voltage increases the power goes up rapidly! If you are able to DD on 16.8V, you’re still overvolting the led by quite a lot like 4.5V, and small increases in voltage means very large increases in current draw. At 14V you’ll be pulling close to 10A if your batteries can handle it (Samsung 25R5’s can), that’s 140W. In a small passive cooled light, you’ll be burning things in short order. T junction of the xhp70.2 is 105C and that has to go somewhere. It’s doable with a big bodied light with lots of thermal mass or maybe a small fan and heatsink. Imalent has been doing gonzo power xhp70.2 builds and that is what they do…big lights with fans inside. They are also $500

texas shooter
Offline
Last seen: 51 min 25 sec ago
Joined: 08/26/2012 - 02:14
Posts: 2041
Location: Texas

gamezawy wrote:
Ok thank you for your inputs but allow me to ask the question in other way

Have any one tried to direct a XHP70 AND XHP70.2 on DTP MCBCP With normal 18650 cells not IMR 4 in series that gives 16.8V, And how they behaved in good heatsinking, i mean a actual experience

You’ll need more than good heat sinking. You’ll need great heat sinking along with lots of mass to eat that 140+ watts. An excellent idea from rngwn is to use 18650 LiFePo4’s. Between the closer to usable voltage and the cell nature voltage sag under load. You’ll probable make a more practical and usable light.

Lexel
Lexel's picture
Offline
Last seen: 2 years 2 months ago
Joined: 11/01/2016 - 08:00
Posts: 5895
Location: Germany

Another option if you can’t heat sink properly like that one Mag lite modder with a solid piece of copper made on a lathe is use a buck driver that limits the current

I have buck drivers, the question is how good your heat sink is to handle a specific current

If you don’t have proper cooling even 15W can be too much