soooooo...why is Anduril mostly seen on FET+1 drivers?

LED Flashlights – General Info
1

Why is it that Anduril is seen mostly on FET drivers? When Anduril does so terribly at sustainable output, why is it all the craze?

The Wurkkos FC13 SFT40 gets a rather terrible sustainable output of 187 lumens at 25°C ambient and default thermal limit.

Post #3000 :smiley:

2

Not sure what you mean. Manufacturers want maximum brightness, thus FET. To get lower lows the use another channel with a 7135. There you have your FET+1. This has nothing to do with Anduril. Many other inexpensive flashlights are using a FET plus a resistor or smaller regulator as well.

The problem with the FC13 is not Anduril’s fault, just bad design of the driver/LED/host combination.

3

FET DD ‘drivers’ only have that high output on a freshly charged cell.

4

Anduril its just UI it works on any driver type, but it should be tuned to specific hardware. Since its open source any “Krappos” can use it the way they like . Anduril code have good thermal control, but without proper hardware it wont work.

5

I’ve seen this low output on every Anduril based model though, but that would be limited to Sofirn, Mateminco/Axtrolux, and Wurkkos, maybe they’re all just cheaply designed.

Okay, so, Anduril isn’t limited to FET drivers then, that clears that up. Have you seen any models with a Boost driver using Anduril? I do recall Fireflies uses a FET+Buck?

6

Pretty much all the Emisar lights have a boost driver option now (at added cost). The comparisons are easy to see there since the leds and UI are the same. The boost driver lights have a lower maximum lumen value, but they last longer and can sustain higher output than their FET counterparts iirc. Per Grizzly’s testing HERE the boost driver heats much slower and sustains ~40% more lumens and also lasts ~30% longer when set to approximately the same output.

7

I appreciate the responses, thank you.

I’ve been making YouTube reviews for 2 years now, but my knowledge is fairly limited when it comes to how drivers work. I always thought that Anduril must need PWM to adjust the brightness, most likely a comment I read somewhere.

8

Most drivers have a microcontroller outputing a PWM signal to control the brightness :

  • in the case of a FET+7135 driver, the PWM signal turns ON and OFF the FET or the 7135 linear driver ICs, the lower the duty cycle (the longer it’s OFF) the lower the brightness.
  • in the case of a PWM dimming DC-DC driver, like the Sofirn SP10, the input voltage is boosted to a voltage that is around the max output desired, then the PWM signal controls a FET, turning ON and OFF the LED.
  • in the case of analog dimming driver (Linear or DC-DC), the PWM signal is filtered with a low pass filter, (crudely) transforming it into a analog voltage that serves as a reference, compared to the voltage across a current sense resistor to achieve a constant current output. The lower the PWM duty cycle, the lower this reference voltage is, the lower the voltage across the sense resistor, the lower the output.
  • Other drivers like mine or the new Fireflies ones developed by Loneoceans use a DAC (digital to analog converter) instead of PWM, directly outputing an analog signal, thanks to Gchart Anduril can do this with the Attiny 1 series now.
  • A few other drivers use digital regulation like YLP Unicorn/Gecko, the firmware is much more tailored for the driver in this case.

So when people say that Anduril has nothing to do with the type of drivers found in Anduril lights, that’s technically true, but in practice not exactly true for a few reasons :

  • Anduril is historically used with FET+7135 circuits because they are very simple to make for community members without electronic knowledge.
  • reference circuit for community firmware (not just Anduril) are FET+7135 (DEL, Texas Ace)
  • developing a DC-DC driver requires time and investment, even if it’s just slapping an Anduril compatible MCU on a previously non Anduril driver, it still need to be tested. Tweaking the Anduril parameters require some knowledge of the Anduril code and manufacturers have shown they don’t, for FET+7135 driver they just take a pre compiled one from the list usually.
    Also there may be some issues that actually need a firmware code change to fix.
  • FET driver allows high output with very cheap components and small drivers, the shift in popularity towards more efficient drivers is only very recent.
    So when a low cost manufacturer like Sofirn wants to make an Anduril flashlight, the target audience wants high output, the low driver cost and simplicity allow cheaper investment and less risk.
2 Thanks
9

Are you sure FET constan current drivers use smoothed out PWM ?

10

A FET constant current driver ? A linear driver with a FET to drop the voltage, like Convoy and Emisar linear drivers ? Yes they use a filtered PWM signal.

11

Yes, linear drivers.
I thought they did it the same way as a 7135 current limiter.
Didn’t know they use smoothed out PWM.

12

I think it is/was/was designed to be all the craze because of the amazing amount of enthusiast features contained in it and the ability to customize so many functions of the light. It built upon previous efforts and took it to the next level (or next three levels). Getting manufacturers to shift to different hardware is somewhat a separate issue and until very recently the “nice” drivers were mostly found in much more expensive lights. Throughout the development process it was clear that the richness of the firmware was going to turn some people off, and sure enough we’ve seen a lot of people knocking it (often unfairly) because it’s too complex for them to feel comfortable with. The odd irony is that if you can get through the complexity for initial setup, the firmware was designed to allow a light to be uber simple if you didn’t want all the bells and whistles.

13

Noctigon 5A driver.

At 100% PWM duty cycle, the R3-R4 voltage divider, smoothed by C3 gives 50mV at IN(pin1) of the Op-Amp (U3), the Op-Amp output control the gate of the FET so that IN=IN-(pin3)=50mV so you also get 50mV across R10, which means 5A, at 50% duty cycle it would be 25mV : 2.5A …etc.

So it doesn’t matter what microcontroller/firmware is used, it just need to turn on the regulator (the Op-amp here), and output a PMW signal.

Most driver with DC-DC converters are pretty similar, the output of the opamp is wired to the feedback pin of the converter IC, IN+ and IN- are swapped because the converter increases the output voltage when FB is too low, which is the opposite of a NFET and its gate-source voltage.

1 Thank
14

I can’t prove it but I have noticed this 100s of times. Lights with andruil and electronic switches ramp down faster than lights that have a mechanical switch and non andruil interface.

15

Regarding the Emisar Boost driver upgrade, the moonlight mode is not as low as the Linear +Fet drivers.

16

Oh, so maybe you can actually see something in the lowest mode with the boost driver? :+1:
:wink:

17

Actually, yes :slight_smile: So far, my only light with boost driver is the DW4 which I use as my headlamp. For that, I actually like the little extra brightness on the low mode.

18

I’ve proved it to myself lots of times in my light tube —- most of them do their little dance where they ramp down and then slowly ramp back up ( no matter what I have the Ceiling and Temp set at ) Only a few work properly and stay close to the set top of the ramp