Amps, Watts, Hours - Assistance required please.

[gav6280]

I am intending to direct drive 4 Led's with 4 18650's, so please can you just check i have this correct....

4 18650 cells gives a maximum voltage of 4.2 x 4 = 16.8v Total?

16.8v / 4 XRE = 4.2, So a forward voltage of 4.2 volts per Led?

Each Led is rated at 3 watts so, 4.2v / 3 watts = 1400ma drive current?

1400ma, i know! but please don't worry my heat sinking is 1st class :naughty: (Water cooling)...

So a battery capacity of 2600mah would last 1.8 hours ( 2600 / 14000)?

Or is that further dived by 4 because there are 4 Led's?

Thanks.

 

[mdocod]

The Vf is not something you determine, the LED does not automatically just run at 3W.
4.2V at 1.4A would actually be closer to 6 watts (volt x amp = watt), but that's just to point out that your math there is going backwards, it would be 3W/4.2V = 714.3mA if you were wanting to calculate current based on known wattage and voltage, it's beside the point as the Vf of the LED to run at design spec (~1A) is probably lower than 4.2V, probably much closer to closer to 3.8V, the only way to know for sure is to test your LEDs with an adjustable bench power supply (or variable resistor and multi-meters rigged up in series/parallel), the bin of the LED will give you the possible range that the Vf resides in as I understand, check with cree's binning and labeling PDF for more details on this.

You are very likely going to need to use a resistor in the circuit to make this work to prevent from over-driving the LEDs too badly. I would highly suggest considering putting a buck regulator in the mix somewhere, there are easy ways to do this with cheap lower voltage rated regulators, they can just be placed in-between the LEDs that are in series, so as to have some LEDs cutting down on the effective voltage that the regulator "sees." You'd want to find a regulator with a low voltage over-head requirement for the project.

Eric

 

[Lumencraft (Matt)]

Hi Gav6280,

I am no expert on the subject, but I do have a couple of direct drive light on hand. I have found that the emitters (WC Q5 in my case) if I am measuring correctly actually draw closer to 1100-1200 ma. I think doing this regardless of heat sinking is going to wear out the emitters somewhat prematurely. I cant say how long they last for sure because I only built those recently.

I would like to hear more about this water cooling.

 

[gav6280]

Now i am just more confused.

I need to keep this design as simple as possible with as minimal components, and points of failure as possible.

I have seen lots of single 18650 lights that are direct drive and they are fine.

So i assumed i could direct drive a 4 xre in series with 4 18650's in series? And with the maximum voltage from each cell being 4.2v, each led could only ever draw that or less, which would mean each was getting 1400ma?

In your direct drive lights, using a single 18650 and a single LED what is you run times?

 

[mdocod]

Often times, there is confusion about what is and what isn't a direct drive light.

Many P7 and MCE lights ARE truly direct drive, with no resistor or anything in place. This works fine because the P7 and MCE are capable of operating at 2-3 amps, and when you draw this much amperage from an 18650, the effective voltage output of the cell sags as a result of natural internal resistance. Taking that same 18650, and driving a single die LED without any form of resistance in the circuit (via a simple resistor, or the effects of a buck regulator), will very likely severely over-drive the LED.

I think what may be confusing you, is that we throw the term "direct drive" around very loosely, often when describing flashlights that have that typical "diminishing output" through a discharge caused by a lack of regulation when used with a single 18650, for example, a Fenix TK11, or Olight M20, etc. But remember, that the buck-regulators in these lights (that make higher voltage inputs possible), have some inherent resistive effect on the circuit whether it's operating in regulation or not, this has the effect of raising the required voltage for "full output" slightly above the actual Vf of the LED, and so, while the discharge charge of some of these lights has that steady slope characteristic of a direct-drive configuration, it is not 100% direct drive, there is SOME resistance at work there.

I'm not going to say that a true direct drive is impossible, only that you should do some proper testing of the LEDs, and see if they are going to handle it or not, if you get some LEDs with Vfs on the higher end of the spectrum, and your initial current on a fresh cell is under ~1.5A, and you have ample heat-sinking, then you should be fine,

Eric

 

[gav6280]

So putting something like the dx 1400ma liner regulator in, really would be a better solution overall. Even with my need to keep it a really simple build...

Unless there is a much better solution??

My requirments:

4 xre driven at 1400ma - I know that is over spec, but even 500 hours life is ok for this project.
Maxmum space 4 18650's

 

[mdocod]

I just ran a little test to confirm my suspicions:

I have a Seoul U-bin star mounted loose LED sitting here. I just tried to "direct drive" it, from a protected AW RCR123. The protected RCR123 I ran the test with, is literally YEARS old, and was last charged about a month or more ago. Open circuit voltage before test was ~4.06V resting.

With the resistance of the leads on the multi-meter, and my hand-held connections for the test, the current measured 1.31A starting, and started rising as it heated up, it was at 1.35A within a few seconds. Obviously, without proper heat sinking, leaving it hooked up like this would lead to thermal runaway, but this rise only took a few seconds, so it may have been just reaching a point to stabilize at had this been a longer test with more heat sinking.

This was just an old worn out RCR123 sitting well below the 4.20V charge you could expect to achieve on new cells, I wouldn't dare repeat the test with a fully charged brand new 18650, it would certainly fry that LED in mere moments.

Eric

 

[gav6280]

~Oh that is very interesting. Thanks for that.

There are people here though on CPF that are running there XRE led's at 1400mah, and have been doing so for months but that is with excellent heat sinking.

 

[Mr Happy]

There are people here though on CPF that are running there XRE led's at 1400mah, and have been doing so for months but that is with excellent heat sinking.

There are? I'm not sure what the point is though. If you run at 1400 mA, you don't get twice the light of 700 mA. Far from it. You get a little more light, far less run time, and a soon-to-be-fried LED.

[Edit: The maximum drive current for a Cree XR-E is specified as 700-1000 mA. If you go above 1000 mA you are just wasting electrons and money.]

 

[gav6280]

[Edit: The maximum drive current for a Cree XR-E is specified as 700-1000 mA. If you go above 1000 mA you are just wasting electrons and money.]

Not if you have designed it to be that way, and also accept that he LED is going to only last around 500 - 1000 hours even with great heat sinking.

For me that would still be more than 10 - 15 years use...

 

[Lumencraft (Matt)]

Hi guys,

I was just wandering if some one could confirm whether or not my method for measuring amperage/voltage is correct? First when I measure voltage I set up the power supply ie the light with cells and hook up the bulb with jumper wires kick on the power and then touch the leads of my dmm to the clips on the jumpers. Example is my 5761 v bulb is 7.2v with this method. But for amperage I do not use jumpers I just touch the neg lead to the neg batter end and then the pos lead to the light body to complete the circuit? I am not trying to hijack but I wanted to make sure I am doing this correctly before I give more input on the subject.

 

[gav6280]

Sorry i don't really understand what it is that you are trying to do?

Amps are usually measured in series and volts are measured across 2 pionts to measure the potential difference, I think it may have been better to start your own new thread? This thread is dealing with direct driving multiple LED's using 18650's.

 

[mdocod]

I'm hoping someone could step in who is more familiar with the various LED drivers on the market than I, and make a suggestion. Perhaps consider changing your thread title (if you still can) to reflect that you are looking for assistance picking out a proper current regulator for your configuration.

I'm tempted to say use a boost/buck style circuit (like one rated ~2.7-4.5V) or something like that, but run it in with 3 LEDs "behind" it and one "in-front" of it in a series combo, but I'm not sure if this would work or not. There was a GREAT thread about taking lower voltage regulators, and using them in high voltage multi-LED arrays by installing them in series with the LED string, such that the regulator itself was still operating within it's range, and controlled current across the entire series string of LEDs.

 

[Mr Happy]

To be honest, I don't have a lot of interest in this thread since gav6280 seems to be completely underwater here (pun intended). When people are out of their depth it is very hard to help them swim.