LED drivers and Accessories you want, but don’t exist

So 2 channels but only with 7135 banks. The perfect driver for the thing i had i mind would have a 7135 chip for low on both leds and a fet for full power for both leds. So 4 channels in total. Would it be possible to do something like this with the attiny chips we usually use? Light would be a e-switch light (with lighted switch controlled by the mcu).
I am not sure if the attiny chips we usually use have enough output pinns for something like this

It’s technically possible with ATtiny85, but not very good. Since the ATtiny85 has only three hardware PWM outputs, the fourth output (or all of them) needs to be PWM’d by the firmware. Better chips are available, though. And some of the BLF members are working with some of those better chips. ToyKeeper has been working on a ATtiny1634. I can’t be sure, but it seems somebody said that the BLF LT1 uses that. It has more pins, more hardware PWM channels, and other features compared to ATtiny85.

And is cheaper and takes less board space.
Pure upgrade though Narsil doesn’t support it yet.

Thanks for the infos DavidEF. I hope this new driver project will work out. It would make some new things possible. Nice to know that they work on new hardware!
And infos how far they are? Or if it is looking good so far?
I really love toykeepers work (and i am thankfull she invested so much time in it) but with the new low vf leds and more new things comming, the driver hardware designs we have now are reaching its limits sometimes.
Some new Anduril compatible drivers would be really nice.

Well, loneoceans is planning to build a new buck/boost or switching driver and is asking for suggestions HERE. I have already suggested that he make it compatible with Anduril or Narsil, and I think he will. But I also asked if he would use a more modern MCU and port one of those firmware choices to it, and he said it is outside the scope of his project. Still, having a good buck/boost driver compatible with BLF firmware will help with those low Vf emitters.

Affordable waiven/RLT collars


I checked the loneoceans project. It is really interesting. Narsil or Anduril would be a huge + . It is a pain to have to many lights with so different UIs.
To bad he will not go for a new mcu. Using a new mcu would open up way more option. But i take what i get :sunglasses:
But that is still not helping with the idea i have in mind. Will have to push it untill we have something new.

To change from the above discussion, I’ve been looking for a ramping, constant current driver with a maximum in the range of 700-1000mA for modding some small scale video lighting. This may exist somewhere but, if so, it’s proving difficult to find. There are ramping PWM drivers and there are constant current drivers with modes but they all seem to be for a 2+ amp minimum.

A 17mm buck for 1x18650 would be nice but mostly I’m just looking for something which makes it easy to step through outputs compared to soldering and desoldering 7135s. Something as simple as cycling through 100, 250, 400, 550, and 700mA modes would probably be plenty fine.

Anyone know of something like this or which could be reconfigured to do something close?

Mains (AC) powered 28-36v 50-75w led driver

Today I stumbled upon 2 linear regulator that look interesting to me…but I don’t understand them well enough to know whether they are really good. They are definitely more complex than 7135….

The first is Diodes Inc. DLD101.
It’s a 3x3 mm item that supports currents up to 1A. Reasonably priced.

The other is Analog Devices ADP8140ACPZ-2-R7.
4x4mm, 4 channels of 500 mA each. Extensive protections. $10 a piece.

That DLD101 looks like it could drive a string of LEDs (way higher supply voltage) via the FET, but the control NPN is limited by the resistor so that it won’t fry. So you could have a string of red LEDs for a car’s taillight running directly off 10V-15V and not have to worry about it. Vaguely recall the FET could stand up to 100V or so, too, so you could even zener-limit the NPN side. But make sure the FET doesn’t drop too much voltage at 1A, else it’ll fry itself, so LED voltage and supply voltage have to be pretty closely matched.

7135s are essentially just fixed gain NPN transistors that are preset to 350mA (more or less). Either they drop excess voltage when too high or saturate at 0.2V if too low. And you can fry the enable pin (don’t try connecting to 12V, kids!) if hit with too high a voltage.

The thing I don’t like about these types of regulators is that the overheating protection is to shut down. I’ve used CAT4104s which can sink 700mA in a small package, but with LEDs with low forward voltage they would overheated rather quickly. With multiple 4104s the light turned into a disco ball when getting hot. With small packages and high amps, these types of regulators overheat pretty fast.

That’s why I’ve settled on CN5710s, they throttle down output when overheating, and do it gradually so it’s not noticeable. No need to implement fancy temperature throttling firmware, they take care of it themselves.

Thank you for the input. :slight_smile:
It looks that ADP8140 does it right, regulates current to keep temperature at safe level. And furthermore it supports connecting a thermistor to control LED temperature which we could easily use with L4P MCPCBs. Though the price of a 6A driver would be high. :frowning:

DLD101 datasheet doesn’t mention whether and how does it control temperature…

I’ve really taken a liking to the QX7138 lately. You can use it to drive a FET linearly (up to 3 amps according to the datasheet, I haven’t tried to push it higher).

Ahh, you’re absolutely right. I looked at that funky diagram on page 16 and saw either active or shutdown and didn’t bother continuing. I missed the “scale Iled down” box and didn’t read the next page. Throttling makes it much more interesting. It does appear to have a little higher dropout voltage than the CN5710s but maybe not too bad, an interesting chip indeed. Nice find!

Not much info in the datasheet on this one. Almost all of the smaller adjustable regulators I’ve looked at have shutdown if they have temperature protection. I can’t see anything mentioned in the datasheet. Also, if I’ve not misunderstood the datasheet, the dropout voltage appears to be rather high, a little too high for it to be of interest.

Thanks for checking it Mike. :slight_smile:

17mm Buck driver
calculated values for XHP35

Could I ask, what do you think of this op-amp for H1-A/H2-C style current control feedback?


I’m trying to find a balance between input offset voltage and gain-bandwidth product, I’ve seen different drivers use different ends of the scale.


I have been testing Convoy’s new XHP35 boost driver here: https://budgetlightforum.com/t/-/61809

It uses the MP3431 chip, and works quite well.

I include the schematic, board layout, and testing results. I invite anyone in this thread to check it out if you’re interested.