How to build a linear LED driver that is dimmable?

[PeterF]

Of course i have been searching the websites of the usual suspects for a bit but could not find anything that i particularly liked.
There are a few things that are set:

Battery, single cell LiIo, with 3V to 4.2V.
LED, standard white LED with around 2.5 to 3.5V forward voltage, depending on the current.
Around 700mA to 1A current.
No PWM, analog dimming only, yes, i really mean that, no sense in telling me “you can not see it anyway”, i hate PWM.
Linear, no switch mode.
Around 5 to as many steps possible.

Now, there is the problem as you can see, no PWM, i would like to have a adjustable linear current source.
There are plenty of constant current switch mode power sources, but no linear ones.
A voltage controlled MosFET current sink seems to be the best idea.

Any recommendations on how to build something like this?

[EasyB]

It sounds like continuously adjusting the gate voltage of a MOSFET would get you what you want. I’ve done this a couple times using a voltage divider and an adjustable resistor, and it works well. It is very simple to DIY as far as circuitry.

mod: continuous brightness adjustment via sliding switch

I have been more interested in optimizing the user interfaces of my flashlights. Ramping FW like Narsil and from Toykeeper have the features I want and are intuitive to use. Maybe an even more intuitive UI is a sliding switch controlling the brightness from off to full. The way I chose to implement this was to continuously vary the gate voltage to a MOSFET via a voltage divider with a sliding variable resistor. The light I put it on is my 7xXPL modded Convoy L2. I had previously drilled a small…

Mod: Convoy S2 with rotary tail switch

This mod combines aspects of two of my recent mods, the e-switch in the tailcap mod and the sliding resistor switch mod. It uses a voltage divider with an adjustable resistor to continuously control the gate voltage to a MOSFET. In the tailcap, the MOSFET needs a separate power source to turn on. In my tail e-switch mod I used a Li ion coin cell. I would prefer to use a supercapacitor instead of a coin cell, but the attiny13A MCU requires 2mA while on, which is too much for a supercapacitor of …

If you want a more traditional clicky switch the LD-3 driver from led4power is a linear driver with no-PWM modes.

Or there are triple channel drivers like the texas avenger. You could put a different number of 7135 chips on each channel. This would only actually give you 3 pure linear without PWM modes, though.

[Wildcat]

Or look at EagleEye X2R driver. 1*7136 chip, 3 levels, no PWM.

[PeterF]

Thanks for the replys!

@EasyB

It sounds like continuously adjusting the gate voltage of a MOSFET would get you what you want. I’ve done this a couple times using a voltage divider and an adjustable resistor, and it works well. It is very simple to DIY as far as circuitry.
mod: continuous brightness adjustment via sliding switch
Mod: Convoy S2 with rotary tail switch

Yep, that was pretty much my idea with the only difference that i would replace the potentiometer with a microcontroller and a few passive components.

You could put a different number of 7135 chips on each channel. This would only actually give you 3 pure linear without PWM modes, though.

@Wildcat

Or look at EagleEye X2R driver. 1*7136 chip, 3 levels, no PWM.

I thought about a solution with multiple 7135 but that gives me a resolution in “only” 350mA steps, that is was too much in the low mode.
but the 7136 sounds pretty good, sadly i can not find it from a reputable source.

[DEL]

You might want to try TA’s Commander driver:

Texas Commander "TC" Constant current Opamp driver without PWM

Ok, I don’t have time for my normal long post explaining every little detail of this driver. Plus with all the advancements coming down the line in driver tech I see bigger and better things coming out this year anyways. Still this is a good concept and works great so it is worth posting it up. Basically while this was not the goal, this is an open source LD-3 driver. The difference being that the LD-3 has a better 4 layer PCB and can handle a bit more heat then this driver. The features and b…

Not in use in any light as far as I know, but it tested very well. It just tends to overheat in the mid-range amps (2-3 A). Your 1 A requirement sounds like a perfect fit. It has true linear regulation in 255 steps. It even works with most of the firmware around here if you do not need the DD mode or moon mode.

[PeterF]

@DEL

You might want to try TA’s Commander driver:

Texas Commander "TC" Constant current Opamp driver without PWM

Awesome!
Thank you, that is pretty much what i had in mind to build myself,
That would spare me a lot of work if i ever get around to converting the project, would just have to write my own firmware.

[Rufusbduck]

The LD-3 is a ready made driver that uses a voltage controlled FET, no pwm.

[PeterF]

@Rufusbduck

The LD-3 is a ready made driver that uses a voltage controlled FET, no pwm.

Thanks for the reply, i recognize your username,
have a look at mine and maybe you will notice why i can not use a pre-made driver

[Rufusbduck]

I had read your other thread but thought this was a separate deal. In any case the LD-1/2/3 use a voltage controlled gate FET w/sense resistor which seems to be what your looking for so between it and the Taskled linear driver you should be able to get some ideas.

[PeterF]

@Rufusbduck

I had read your other thread but thought this was a separate deal.

Sorry, i did not want to sound like an ass

It is just an idea for the far away future that i could not shake and needed to know how i might make it work.

In any case the LD-1/2/3 use a voltage controlled gate FET w/sense resistor which seems to be what your looking for so between it and the Taskled linear driver you should be able to get some ideas.

Yep, that seems to be the way to go, i think i should even be able to ditch the amp and use the differential amp on the the AVR after a bit of calibration.