LD-3: LD-2 on vitaminS (2-12Amps CC with configurable UI)

[DavidEF]

Wow khas! Melting a plastic bag? Yeah, I’d say that’s a good level for ‘normal’ use!

[Agro]

@Flintrock

So the area where there becomes some difference is not technically the efficiency of the driver itself, but how effectively it manages the LED.

A linear driver that runs Vo nearly equal to Vi but reduces brightness with PWM will technically run more efficiently than a driver that reduces Vo with constant current (like this one does) BUT,

a) you don’t get current regulation then

b) the LED itself is run harder and becomes less efficient. It turns out the whole system efficiency is higher for the constant current setup.

So, because this thing can deliver constant current without PWM at any power level, the whole system efficiency should be as good as is possible for a linear driver. At the lowest modes, that’s probably about 65% driver efficiency. (2.7/4.2 for example)

Several more efficiency questions…

1. With a tripple XP-L, what would be more efficient, a single 7135 350mA PWMed down to the average 180mA or this driver doing 180mA?
2. Does power level affect efficiency? I can get 300 mA from 3A@20% or 6A@10%. Would efficiency be the same?
3. What’s the direct-drive efficiency?

[DB_Custom (Dale)]

Good work Khas!

With it being this easy, thanks to LED4POWER, anyone that’s never experienced 3000+ lumens in a light small enough to have a shirt pocket clip owes it to themselves to try this! Schoolgirl giggles all around.

[led4power]

1. 180mA is more efficient, but at those currents difference in efficiency at 180mA and 350mA is probably negligible. Single 7135 is more cost effective in this case.
2. Higher current means lower efficiency, especially at close to maximum currents,where LED efficiency can drop to ~50lm/W (3-4 times worse compared to low current efficiency). You can easily calculate eff. from current vs. lumen-out graphs measured by BLF members( search djozz LED tests). Efficiency is equal to Lumen/(Current*Voltage).

[Agro]

@led4power

1. 180mA is more efficient, but at those currents difference in efficiency at 180mA and 350mA is probably negligible. Single 7135 is more cost effective in this case.

Cost is a another (though also important) concern. Actually I meant 8-12x7135s that can drive the led with a single chip. That’s because when it comes to efficiency, 100-200 lumen modes are what matter to me the most. LED efficiency at below PCT is something that I don’t know how to gauge. Controller efficiency is unknown as well….I’ve seen HKT tests of LD-2 and several others, but they are not enough for me to be able to tell which will win in such case.
You say that difference in efficiency is negligible. Could you please clarify if you mean LED efficiency or LED+controller combo?

@led4power

Higher current means lower efficiency, especially at close to maximum currents,where LED efficiency can drop to ~50lm/W (3-4 times worse compared to low current efficiency). You can easily calculate eff. from current vs. lumen-out graphs measured by BLF members( search djozz LED tests). Efficiency is equal to Lumen/(Current*Voltage).

Yep. But question 2 was meant to tell me more about which LD-3 variant would be the best for me. 3A with DD is probably the best, but 6A is tempting. I’m gauging efficiency to be able to tell better.
I’m also considering DrJones, modified HEDBEBES 104A, and several others though.

[EasyB]

Agro, you might be interested in some relevant analysis I did recently: Power conservation analysis of linear and PWM brightness regulation methods

I think I can answer your question 2: with this linear driver the current of your highest mode would not affect the efficiency of lower modes. So there’s no negative to going with the 6A option, lower modes should be effectively equivalent to the 3A option. I guess the different versions of the driver (3A, 6A,…) have different sense resistors? Sorry led4power if this is not correct.

[led4power]

@Agro

@led4power

1. 180mA is more efficient, but at those currents difference in efficiency at 180mA and 350mA is probably negligible. Single 7135 is more cost effective in this case.
2. Higher current means lower efficiency, especially at close to maximum currents,where LED efficiency can drop to ~50lm/W (3-4 times worse compared to low current efficiency). You can easily calculate eff. from current vs. lumen-out graphs measured by BLF members( search djozz LED tests). Efficiency is equal to Lumen/(Current*Voltage).

Yep. But question 2 was meant to tell me more about which LD-3 variant would be the best for me. 3A with DD is probably the best, but 6A is tempting. I’m gauging efficiency to be able to tell better. I’m also considering DrJones, modified HEDBEBES 104A, and several others though.

1. 8-12x 7135s nanjg-style drivers use PWM to get lower modes(pulsing 3-4Amps),so efficiency is lower because today big LEDs has typ. ~90-110lm/W at those currents, and 160-200lm/W at currents lower than 350mA . Non-PWM linear driver itself actually has a little bit lower efficiency because of LED lower Vf (which means more voltage to burn),but that difference is usually <15% worst case (fresh battery, difference between low mode current and high mode current is large,but that's also worst case for PWM driver).One usually forgotten fact is that PWM driver has higher losses in current path resistances (spring,switch, wires) - to be precise I2R losses are 1/D times bigger,where D is duty cycle (3.5A with 10% duty cycle gives 1/0.1=10 times higher losses in circuit resistances). Overall lamp/flashlight efficiency is always better with non-PWM vs. PWM, how much it depends on difference between low mode current and high mode(100%) current.
2. I you want to run your LED >4A,then choose 6A version. There is no difference in overall efficiency betwwen 3A nad 6A at same currents.

[Agro]

Good drivers like the ones I mentioned have special 1x7135 channels to drive the leds at real 350-380 mA, optionally PWMed to even lower values. With this 350 mA driving 3 LEDs we’re talking about 200+lm/W minus controller losses. So I’m talking about 180 mA average PWMed from actual 350 mA compared to actual 180 mA from LD-3.

[led4power]

1st answer,post #85.

[Agro]

OK, thanks.

[The_Driver1 (The_Driver)]

Does the low voltage protection also work in moonlight mode?

[led4power]

Yes.

[AnhTran]

Hi led4power, can I use this driver without an electronic switch (convoy C8, xhp70 and 2 Li-ion)?
If yes so how will it work?
thanks

[The_Driver1 (The_Driver)]

Nope, only clicky.

[led4power]

@AnhTran

Hi led4power, can I use this driver without an electronic switch (convoy C8, xhp70 and 2 Li-ion)? If yes so how will it work? thanks

Convoy C8 has tailcap switch only?

[giorgoskok]

@led4power

@AnhTran

Hi led4power, can I use this driver without an electronic switch (convoy C8, xhp70 and 2 Li-ion)? If yes so how will it work? thanks

Convoy C8 has tailcap switch only?

Yes

[AnhTran]

Yes, so this driver have to be used in a light that have electronic switch at the body?

[steel_1024]

LD-3

only clicky(tailcap switch)

[Agro]

led4power, do you intend to send it to HKJ for review?

[led4power]

There is no need for that,since hardware part of driver is very similar to LD-2 (which HKJ did test).