Test/Review of 1* AA/AAA 1-Mode 550mA Linear Booster LED Flashlight Driver Circuit (Nanjg 102)

Flashlight Modding and DIY Parts
1

1* AA/AAA 1-Mode 550mA Linear Booster LED Flashlight Driver Circuit (Nanjg 102)




Driver is from fasttech.com.

Official specifications

  • Constant Current: No
  • Driver Type: Boost driver
  • Input Voltage: 0.9~1.5 -volt
  • Output Current: 550 mA
  • PCB Height (w/o Components): 1.5 mm
  • Mode Count: 1
  • Diameter: 13 mm
  • Product Weight: 0.6 g




Measurements

Tested with: Cree XP-G2 led
The driver is a boost driver
Height: 4.5mm
Diameter: 12mm


This driver is always pulsing the led, spoiling any power and efficiency calculations.

High



The driver does not have stabilized brightness, but increases brightness with voltage. It uses a lot of current if the battery can deliver.







Increasing the voltage from zero, shows that the driver turns on at 0.8 volt.



Here the pulsing can be seen, it is very fast (way above being visible) and I expect the frequency to vary with battery voltage.



Conclusion

This driver does what it is supposed to do, but be very careful with lithium batteries, they may overload the driver.



Notes

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

2

Is it just me, or is this little driver incredible?

0,75A to the emitter with 1,2v input. Close/around 95% efficient! 1,5$.

As far as I know. Typical cheap AA/AAA boost drivers often have 40-75% efficiency and around 0,3-0,5A output.

What I dont understand is that the previous tested driver from FT, that looks similar and have same SKU, was quite crappy/typical.

Previous review.

Latest/this review.

What am I missing?

3

Thanks HKJ.
2 diodes, 1 transistor, and I guess the fourth component is an oscillator to drive the transistor?
Maybe that you could put capacitors at the input/output to be able to measure the efficiency? Because that’s one important data that we don’t have here.

4

Hi,

I know that you said that you were testing with an XP-G2, but I was wondering what the maximum emitter voltage this driver can output? I’ve noticed that boost driver specs don’t usually specify that… is that because there’s an implicit assumption that you’re trying to drive a certain type of emitter?

Thanks,
Jim

5

A boost driver wont have any problem driving different LEDs with different Vf. Heck, you could probably even run 2 LEDs in series and that would still work! But with this kind of unregulated driver, the higher the Vf, the lower the (average) current will be.
A perfect boost driver would supply a constant current no matter the input or output voltage (There are always limits though), but that’s too much to ask for a driver that small I guess.

6

Does anyone know a cheap source of Energizer lithium AAA batteries? I have this driver on order.

7

Probably this sentence: This driver is always pulsing the led, spoiling any power and efficiency calculations.

Adding capacitors would change the working condition. I could do the calculations with my oscilloscope, but that would require a lot of manual work.

Depending on how the driver is made, there might not even be a limit on the maximum voltage, except when the driver destroys itself (Notice how some drivers says they may not be run without load).

To keep my testing time reasonable I only test at the specified parameters for the driver.

8

analog meters :wink:

9

Wont help.

To get power you need "average(voltage*current)", not "average(voltage)*average(current)".

As long as voltage and current are fairly stable, either of the two works, but when both voltage and current bounces up and down, they must be multiplied, before doing the average.

10

What I was wondering is if this driver could potentially drive a 6V MT-G2, even/albeit at, say, 1000 mA?

11

A wild guess is that it will drive that 6V LED but at a lower current.
If you try let us know the outcome! :slight_smile:

12

Except if you had an analogue power meter, like electric companies use or did use, it would do the multiplication in analogue before the analogue averaging. I think the one I saw used two coils attracting each other, one with the current in it and one with a current that depended on the voltage (resister limited maybe). The mass and damping of the rotating part did the averaging.

13

That meter is designed for 50/60Hz AC, not for fast pulsing DC, it would get the average wrong.

Even my electronic power AC/DC meter will have problems, it is not fast enough. It has to be an oscilloscope.

14

Nice review as always HKJ!

I got a couple of these…tied one into a XT-E I got on a whim as a test on my bench…whoah bright with a AA (but aren’t they all when you are staring right at the emitter when you fire em up :stuck_out_tongue: )

15

Different components, but also a small booster driver (I don't have a NANJG 102): I hooked an old sk68 driver on a AA-NiMh battery to a MT-G2, it only delivers 2mA to the led:

16

Yes the electric company ones must have used iron core magnets. If you use only wire coils with no core, like the one we used in a college physics class probably had, it will work up into the microwave range.