Review/Test of XM-L Multi-cell Circuit Board, 3A, 5.5-12.6V

Review/Test of XM-L Multi-cell Circuit Board, 3A, 5.5-12.6V





Driver is from Intl-outdoor.com, but he does not sell it anymore.

Official specifications

  • Suitable for 2*18650 to 3*18650 batteries or 26650 batteries
  • Suitable for 1 LED, or even 2-3 LEDs in series
  • Input Voltage: 5.5v-12.6V
  • Step-Down(Buck) driver
  • Current: 3A
  • Constant current
  • Modes: High-Medium-Low (100%-30%-5%)
  • PWM: 3900Hz
  • PCB diameter: 19mm
  • Contact board diameter: 21mm
  • Height: 9.5mm, 15mm with the contact board.
  • Leads: 24AWG
  • Tinned leads
  • Thermal management, overheating protection 55-60C
  • Low voltage protection
  • Memory function
  • Thermal pad included (5mm thick)



Protection functions:
Flashing 1 time every 5 seconds - High temperature
Flashing 2 times every 5 seconds - Low voltage
Flashing 3 times every 5 seconds - High temperature and Low voltage



Measurements

Tested with: 1, 2 and 3 Cree XM-L2 in series.
Diameter: 21mm and 19mm
Height: 10mm
The driver can be used with 2 or 3 LiIon batteries in series.
The driver has high, medium and low.
PWM is at 1kHz
Switcher frequency is about 460kHz
The driver has memory, the actual mode is stored when the light is off for a short time.
A fast off/on will select next mode.
Medium duty cycle 48% at 1 KHz
Low duty cycle 9% at 1 KHz



High 1xXM-L2



The driver has very good stabilization down to 5.5 volt, where it warns about low battery. This warning is only useable for two cells.
The efficiency is not that great with 3 cells, but gets better with two cells.



Up to 4 watt is lost in the driver.



The led as supplied with 13 watt and 3.6 ampere. This current is to high for the led and can be a problem.



Running the test in reverse, i.e. slowly turning up the voltage, shows that the driver starts at about 4.8 volt.








The current to the led is very smooth, only a small amount of the buck converters frequency can be seen.



Low voltage warning is two flashes, but it is only useable for two cells.



Medium 1xXM-L2



Medium is about half the current of high, in my opinion it is a very high medium level.




I do not measure the led voltage, but assumes it is the same as in high, this is necessary to do the power and efficiency calculations.
The led has about 6.3 watt in medium.




The driver uses pwm to set medium level and the duty cycle is 48%, this matches the 50% current very well.



Low 1xXM-L2



In low the current is 0.35A.




The power to the led is about 1.2 watt.




The driver uses pwm to set low level and the duty cycle is 9%, this matches the 10% current very well.



High 2xXM-L2

This driver can also drive two leds in series.



With two leds in series, the driver does some funny stuff, just before it goes out of regulation. As can be seen on the curve I measured a peak current of 4.25A.



The loss in the driver goes down with two leds.



And power to the leds is double up, i.e. 25 watt.



Medium 2xXM-L2



In medium the current peak can also be seen.
The over 100% efficiency is, of course, not correct. This is because the fixed led voltage is wrong, when the input voltage gets below about 7.5 volt.




The medium mode is about 12 watt with two leds.



Low 2xXM-L2



The current spike is also present in low mode.




The led power is about 2.3 watt.


High 3xXM-L2



This test did not work as expected. The driver delivers a bit more current than the leds is rated for and either that or an even higher spike killed one of the leds.
The led is rated for 3A, but the actual current was between 3.75A and 3.9A.





This terminates my test of this driver, I am not going to risk more leds



Conclusion

The driver has good stabilization for one led, but does supply to much current. The low battery warning can only be used for two batteries.
Using it with 2 or 3 leds, does increase the risk of blowing the leds, due to the current spike just before going out of regulation.



Notes

How do I test a led driver
List of all tested drivers

Thanks for the review, I just got such a driver last week.
I ordered this one which looks similiar
http://www.lightmalls.com/super-output-ssc-p7-led-driver-board-low-high-middle-5-5v-15v-2

And there is never too much current :wink:

Looks suitable for mtg use

Thanks for the review.

Seems like intl-outdoor have just stopped selling the 4,5A large torroid version too. They had it very recently..

Oh, well. This driver and its siblings are quite popular.Sold from many stores. Easy to resistor mod, and seems to do direct drive when using 3 cells in series combined with 3 LEDs in series (Something Tom E have done a lot with the 4,5A version from IOS).

There seems to be some inconsistency (from some stores) if the the drivers use a small or a large toroid. But several of the high output versions seems to have a large toroid, atleast sometimes..

3A

3-Mode (no contact board)

3-Mode (no contact board)

2pcs, 3 mode (no contact board)

5 pcs, 3 mode (no contact board)

3-mode (higher price, with contact board)

5 pcs, 3 mode

5 pcs, 5 mode

5-mode

5A

3-mode

5 pcs, 5-mode

5pcs, 3-mode

Just wondering, why is 3.5A too much current? I thought that is well within the envelope of a XM-L2 or a copper-sunk XM-L.

And thanks for the review! :slight_smile: I saw this driver on Fasttech and wondered about it. It looks like this driver does need a contact board since there is no plated edge on the main board. How hot does the driver itself get? Is the driver supposed to be mounted on something inside the pill, or does just the contact board need to be mounted and the main board can simply ‘float’? Or does one pot the driver to the contact board in a scenario as this?

The maximum rated current for XM-L2 is 3A and I do not expect a driver for XM-L2 to deliver more than that, without modifications.

That the driver increases the current, when the voltage is low, is a very bad design.

I do suspect that one of the leds in my setup was a bit weak (It had been tested at 3A).

As you can see in the photos, it does include a contact board.

I do not measure temperature, but have added a high speed fan to my test setup to get rid of heat. I.e. you only measure of heat is the driver curve with watts lost in driver. 4 watt is a lot of power and do require a small heatsink to get rid of.

When working with electronic I usual expect a TO220 package (Old style flat transistor or 78xx package) to handle up to 1 watt without a heatsink.

Thanks for the detailed testing HKJ. Very nice work. :wink:
The driver appears to be the same one I played around with here from lightmalls, but without the battery board. 2 Li-ion 3 amp Driver Super Output Current Mod "Now with Larger Toroid driving a MT-G2"

Is that the manufacturer’s recommended maximum? Because I see all the time on here people driving almost twice that through copper-sunk L2’s. Match tested the L2 on Al and Cu and on both he took them to at least 5 amps.

Yes, it is the manufacturer rating, usual a led can take considerable more current, if it has enough cooling.

I do not know if the driver had a huge current spike or if the led was a bit weak, but the led was not very hot.