Keep frying my p7's.....why?

[Amandrew]

I recently did a p7 mod on my mag 3D. I cut the access plastic off of the mag switch and soldered two wires onto the switch, one pos and one neg of course. I bought a p7 heat sink and Alumina Adhesive from 4 sevens and a p7 (CSXOI) from kai. I put all of this together and it ran great!!! The only problem was the first p7 just stopped working. I thought that I might have short circuited the p7 with the aluminum reflector. So I bought another p7 same one. I installed it again, and again it worked great!!! I put the plastic cut down reflector back in just to be safe. After a few hours of running it, it fried again. It just stopped working. I have tenergy 10,000mah d cell in it, 3 of them to be exact. What am I doing wrong?

 

[Benson]

I recently did a p7 mod on my mag 3D. I cut the access plastic off of the mag switch and soldered two wires onto the switch, one pos and one neg of course. I bought a p7 heat sink and Alumina Adhesive from 4 sevens and a p7 (CSXOI) from kai. I put all of this together and it ran great!!! The only problem was the first p7 just stopped working. I thought that I might have short circuited the p7 with the aluminum reflector. So I bought another p7 same one. I installed it again, and again it worked great!!! I put the plastic cut down reflector back in just to be safe. After a few hours of running it, it fried again. It just stopped working. I have tenergy 10,000mah d cell in it, 3 of them to be exact. What am I doing wrong?

Just guessing, I'd say you're running too high a current. Did you measure the current when you first turned it on? With DD, the current depends strongly on the batteries and the I-V characteristic of the particular LED used, so there's no substitute for an actual current measurement when you get it assembled.

And there's no way shorting an LED should hurt it -- you could blow up a driver (in lights that have them) or harm your batteries with excessive current draw, but the LED doesn't care, and as soon as you take the reflector back out, the LED should light up again (if the batteries aren't dead)

 

[steve6690]

http://www.candlepowerforums.com/vb/showthread.php?t=227323

I'd try a DSX0J. I could be wrong, but I think 3 x D cells is not going to "sag" enough to keep the current within limits

hth

steve

 

[Amandrew]

I thought that might be it with 10,000 mah batteries having to much current. Is the DSX0J p7 the highest rated as far as current? If im still over driving what are some other options? I read of direct drive p7 mags all the time here. I must have over read the too much current part:thinking:.
Thanks for the help guys.

 

[Alan B]

I've put together a few P7 direct drive lights and the "J" bins seem to work well with 3 cell alkaline/nimh or 1 cell li-ion power. The current I've measured is in another thread here but around 2 amps, and it will be a bit more with the meter probes out of the circuit. With direct drive you don't want to get too close to max current because as the LED heats the current will go up even further. This is especially important with large NiMH cells since they can really put out a lot of current if the LED forward voltage is too low.

The other thing that will kill the P7 is heat. Make sure the thermal epoxy layer to the heatsink is very thin, and that the LED is electrically isolated from the sink. Also insure that the heatsink is conducting heat to the flashlight body well, and don't run it too long on full power as the heatsinking is possibly marginal for long term ops. Consider putting in a PWM dimmer to allow lower levels and less heat when that is adequate. The d2flex works nicely.

 

[steve6690]

As Alan says...what I was getting at is that you need to consider the voltage bin of the led. To work at 2.8 amps, an "I" bin needs 3.25 - 3.5 volts and a "J" bin 3.5 - 3.75. When you direct drive from 3 x Ni-Mh "D" cells you'll be getting around 4.2 volts off load but when you apply the power to the led the voltage will "sag" a little. I think the problem with your setup might be that the voltage is not sagging to a low enough level quickly enough to stop your P7 from burning out. I may be out with the actual values above, but I think the point I'm making is accurate..

 

[Amandrew]

I know heat wasn't the problem, I put a thin layer of adhesive like you said. That I never ran it for more then 1 min at a time. Now as far as current, does the DSXOJ handle the most amount of current or is there a resistor I can use to keep the current down? If heat and short circuit weren't the problem what is the sollution for a direct drive to not fry another p7?

Also the cells were sitting for 2 weeks waiting on china shipping my p7 before I used these batts with it. Is the voltage still an issue?

 

[steve6690]

They both handle the same amount of current. Since you aren't using a constant current driver you are NOT feeding the led with a set current. You are feeding it with a voltage which will reduce as the batteries discharge. 3 freshly charged d cell ni-mh will produce around 4.2 volts. A J bin p7 needs 3.5 volts to 3.75 volts to consume 2.8 A. An I bin p7 needs 3.25 to 3.5 volts to run at 2.8 amps. The higher the voltage you feed an led the more current it will draw. For example : if you feed an I bin P7 4.2 volts it will draw more current than if you feed the same voltage to a J bin P7 because an I bin needs less volts than a J bin to draw the same current. You are feeding 4.2 volts to an led that only possibly requires 3.25 volts to draw 2.8 amps. A J bin might need 3.75v to draw 2.8 amps therefore by feeding 4.2 volts to a J bin you have less chance of burning it out by running it at too high a current, than an I bin.
There are a few solutions to this. The easiest is probably to use a current limiting resistor, plus use a D***J bin led. As regards the batteries, you need to measure what voltage they are at. If you are frying led's when the batteries are not even fully charged you will have serious issues when you have just charged them as the voltage will be higher still.

 

[blackdragonx1186]

thats...odd. what type of heatsink are you using specifically? try a HAIII one. thats what im using in my DD P7 mag. i have no resistor of any kind, and ive left mine running for a couple hours straight. the head area got quite warm, as i expected it too. im also using titainium 11K D cells.

 

[Amandrew]

The heatsink i am using is the 4sevens.com heatsink. I dont think heat was an issue though because I never ran it over a min. What p7 are you running in yours?

 

[DM51]

You are getting some good replies from knowledgeable members here, which is good, however LED Flashlights is not the right sub-forum for it. I'm moving it to the Homemade & Modified section, where it belongs.

 

[Mike Painter]

Unless the P7 (all of them) is a lot different from other LEDS you are probably suffering from to much current. A LED will try to draw as much as it can.
Moving from Alkaline batteries to NiCAD taught this lesson to a lot of people in the early days and not having at least a current resistor quickly became common knowledge.
I suspect there are a lot of P7's with the smoke let out of them out there because of a "good" battery combination

 

[spencer]

I would get a driver for it. Thats just me though.

 

[Alan B]

NiMH cells under load are 1.2 volts, so you'll get 3.6 volts from three. This is about right for a "J" bin but too much for an "I" bin.

I put together another direct drive "J" bin with three C alkaline cells today, and we saw about 2.3 amps with fresh cells. So the "J" bin does not seem to need a resistor. With an "I" bin you should use a resistor to reduce the current. Approximately 0.1 ohms should work, you can measure and tailor it to your parts.

Adding a driver with the P7's is not so easy as there aren't very many set up for the high current. Those that are often require more voltage than you'll have with 3 NiMH cells. If you want to go to a driver you can use a HipCC from Taskled, and a 6xAA cell adapter in 2D to get the voltage up. If you want levels you can add the d2flex.

 

[Greg G]

Another thing is that the LED Vf can change. That appears to be what just happened to my DD Mag with an I bin P7 and Accupower LSD's. I have been using this light for the better part of a year now with great success. I recharged the batteries the other day, and it now has two dies that turn angry blue with about 8 seconds of use.

I'm adding a resistor before running it again. The tint on this P7 is really nice, it looks like an SVO, and I don't want to lose this LED.

Update: while taking the light apart to add a resistor I pulled up lightly on the P7 and it moved. The Arctica Alumina had come unstuck. This light gets dropped occasionally onto concrete as it's a worklight. Reglued it and it's ok now.

 

[blackdragonx1186]

The heatsink i am using is the 4sevens.com heatsink. I dont think heat was an issue though because I never ran it over a min. What p7 are you running in yours?

im using the DSXOJ from kai. http://www.kaidomain.com/ProductDetails.aspx?ProductId=5485

i measure around 2 amps of draw i believe.

 

[steve6690]

NiMH cells under load are 1.2 volts, so you'll get 3.6 volts from three. This is about right for a "J" bin but too much for an "I" bin....

True, but with a high capacity battery such as 10,000mah D cell it can take several minutes for the voltage to sag to 1.2 volts. More than enough time to fry a P7.

 

[Alan B]

True, but with a high capacity battery such as 10,000mah D cell it can take several minutes for the voltage to sag to 1.2 volts. More than enough time to fry a P7.

Have we seen this failure? At 3 amps how long does it take to drain the surface charge from a NiMH?

 

[steve6690]

Have we seen this failure? At 3 amps how long does it take to drain the surface charge from a NiMH?

Without hooking it up to a meter and measuring the voltage/current under load we will never know for sure. Worst case scenario...he could have a "I" bin P7 that draws 2.8 amps at 3.25 volts.
One thing about a direct-drive setup is there's not much to go wrong. His P7 is frying because it's getting too hot. The question is why is it getting too hot ? Is it drawing far too much current or is there inadequate heatsinking, or is it a combination of both factors ?

 

[Alan B]

Without hooking it up to a meter and measuring the voltage/current under load we will never know for sure. Worst case scenario...he could have a "I" bin P7 that draws 2.8 amps at 3.25 volts.
One thing about a direct-drive setup is there's not much to go wrong. His P7 is frying because it's getting too hot. The question is why is it getting too hot ? Is it drawing far too much current or is there inadequate heatsinking, or is it a combination of both factors ?

I would expect that if the heatsinking was marginal it would show angry blue, and possibly be damaged rather than so quickly destroyed. Since it goes completely out rather quickly perhaps the current is waay too high, and it isn't a heatsink issue. The OP didn't mention heat, if the light was used for short periods then overcurrent is more likely. If for long periods then heat buildup is more suspect.

In either case the "J" bin will reduce the current and likely solve the problem. One graph I looked at showed current doubling for 0.2V of change (with zero circuit resistance).

Here's what I'd do:

Install a J bin. Make sure the thermal epoxy is thin, and double check the electrical isolation.

Measure the current at the tailcap. If over 2.8A add some resistance (probably won't be needed).