[ GXB172 - 50W Single Cell 17mm Boost Driver! ]

good news as my design has same floating ground for USM mode

could you test together with an MCU already if the chip needs the LVP pin populated with 2 resistors, this is really a pain to integrate with all 0603 components
I had later a design adding 2 resistors but then later have to make a wire bridge as absolutely no space to place a via for V+

While well out of the loop I agree with your decision, that is the trade off I would have made.

When you are talking about a mode that will run for months in either setup, it really doesn’t matter much lol.

BTW can anyone share the MP3431 datasheet with me? I got troubles getting it from Monolithic…

Impressive driver! What host could even handle that much heat? A C8 light is the biggest thing that comes to mind.

ICSK, from a practical point of view, I am the first to acknowledge that the driver doesn't make sense , since heat will be the biggest issue. However, this is intentional! Moreover, it will be able to drive 50W for a short period of time, before thermal regulation kicks in; and if you turn it down to say 10-20W, then you've got a much more practical but still powerful driver. I've got a prototype PID thermal control firmware tested and it seems to work OK, just need to put it in an actual light to fine-tune the variables. However, you're right that the C8 is likely one of the larger lights that still uses a 17mm driver. I'd imagine that having a C8 and swapping out the XPL HI with a XHP35 HI would produce some pretty nice results , even if the XHP35 HI die is a little bigger than the XPL HI.

Yes, the XHP35 is a great upgrade from the XP-L in every way. More lumens, more throw and generally better tints from what I have seen.

I don’t agree it makes no practical sense.

  1. If you lack distance to shine it where you want, you can use a turbo or walk. I consider the former to be a practical option.
  2. Aside from ultimate power it offers superb efficiency.
  3. Larger variants will find use in larger hosts.

NarsilM v1.2 for the GT brought a good feature ramp to a certain percent like 80% and Turbo only by double press to 100%

I am not sure what you are referring to.

loneoceans said this is not a practical driver - that’s what I disagree with.

Problem that if your host could carry for example 10W, after 50W boost you need a great step down to normalize temperature. And in most cases illuminance difference between this 2-3W and 10W is times bigger than between regular 10W mod and 50W boost.

Hello everyone,

Quick post to share some updates ^_^. I was out of action (caught a flu!) for a while but I'm better now and managed to complete the preliminary firmware for the GXB172. So far I'm happy to say everything is working pretty well! I'll get into this in a bit more detail later on but yes, I've reviewed Bistro and it should port over easily (with some changes specific to the GXB172, and likewise with Narsil), but I've also added a bunch of new features to the current firm as previously mentioned, such as accurate PID temperature control with a dedicated temperature IC (no fudging around with NTCs!), several fun optional modes like candle-flicker or strobes, and smooth brightness transitions which help reduce eye strain and ease mode transitions without blinding the user immediately!

Anyway I put some boards together and I'm happy to report that the only mistake I found was a single silkscreen error, but electrically it all turned out good! Here's how it looks like, top and bottom.

Above is it pictured sitting in a Convoy S2+ brass pill which is my candidate flashlight for this driver. Note the big pads for + and - to the LED, as well as programming header pads, E switch pad, two additional jumpers, and one more auxiliary pad which can be use to power an extra LED or something else (e.g. I used this for a debugging LED during firmware development!).

Above shows a side view of the driver. The big inductor can be easily swapped out for a smaller shorter (3mm tall) one if your host doesn't have space for a ~7mm tall inductor. Using a bigger inductor allows better performance though with a higher inductance (desirable) and lower DC resistor (desirable).

But does it work?

First I had to solder on a spring. It looks a little funny but it works just fine with an inverted spring and solder wick for bypassing.

To test the driver, I paired it together with a Nichia 144 90CRI LED. Now this is not the most efficient of LEDs in terms of being a lumen-monster! Djozz has tested this E1000 LED and it 'only' puts out ~2500 lumens at a 6A ~6.4V drive.

But I really wanted to make a high-CRI flashlight while still being very bright! Don't worry, I have a J4 XHP50 up next for a >4000 lumen light

... and it works exactly as expect with all the modes and functionality working!

I'll be testing this extensively in the field since I'm sure this preliminary firmware needs a bit more tweaking, but much more to come soon!

Because I've configured this light to be a 'shorty' build, having it run at 6A output would lead to a ridiculous ~3 to 4+ minutes of runtime (thermal throttling will kick in before that though), so I changed the modes for a more reasonable 4.2A LED output (measured and verified on the scope) for about 2250 lumens of beautiful CRI 90 Nichia light.

For those interested, here's the current version of the firmware, with one particular mode structure shown.

Obviously the number of modes, memory, special effects, battery cut-off, thermal regulation etc can all be adjusted to whatever you want!

So far it's not quite the same as can be done in some single-button programmable UIs, but I'm planning to iron out any kinks in the core functionality first before moving on to more firmware development. Hopefully when more people get to try out this driver, more talented firmware engineers here can help write better firmware ^_^.

Meanwhile I'll document my build of the flashlight above in another thread (there are a bunch of tricks to do to make the system work well, plus some cool new features!) so I can keep this thread more focused on the GXB172 driver proper.

See the thread here: https://budgetlightforum.com/t/-/50373 for more information on that build.

More to come and thanks again for reading!

Very nice! Can’t wait to see this project progress.

What size are those components? They look smaller then 0402?

I am also very interested in seeing your bistro and Narsil ports. I wish there was a way to program the 1617 and 1616 mcu’s easily (maybe there is now, been a few months since I checked), it would be nice to step up to one of these MCU’s but sadly not an option if we can’t program them.

Do you have schematics and parts list available?

Where can we buy this driver?

From u?:slight_smile:

I put together a really quick video to show the GXB172 in action (preliminary firmware). Also I realize this video production is super poor quality so please excuse this, I'll make a better video in the future!

www.youtube.com/embed/M-_C1XHUgQw

In the video I go over the basic firmware functionality, and to show you all that I'm not making this up!

The light is extremely bright, but my camera stops down pretty quickly. Also the candle-mode is a little silly right now since I'm using a 5700K LED. However I have plans for some nice, warm emitters which will be great for camping! Or I suppose I could use an orange diffuser too ^_^.

Wonderfully done! Looking forward to the moment when you’re able to offer these for sale. Out of curiosity how long does it take the 144 to thermally ramp down on the highest mode?

Very nice!

@ICSK, developing a driver is one thing, but offering them for sale is a very different game. I look at Lexel in full admiration that he is able to maintain a custom driver sales operation, and at not much more than component costs, it takes a lot of time and dedication, not many are able to do that.

Totes Mazeballs work.

Would be very interested if these can somehow end up for sale… Cross fingers…

Even if loneoceans etc cant make enough of them to meet demand, if someone else can (mntelectronics maybe), and give them credit (and maybe a cut of the sales) etc…

TA, I've used no smaller than 0402 for all the passives. The big caps are 1206. They're small, but still hand solderable. The other components are more difficult to solder but it's ok if I'm just making a few. Eventually when I get the driver nailed down I'll just send them off for PCBA ^_^. Mike, as you can tell the driver is still very much in development but like with my previous drivers (see my signature), all of it will be released so everyone can build their own

I'm not quite sure yet; that's the main reason why I build the driver into a test flashlight so I can verify and fine tune the features, such as thermal PID management. Right now I have it designed so once the temperature sensor hits 60C (adjustable), it adjusts the output current in such a way that it maintains the 60C temperature setting and adjusts accordingly. So for example if you dunk it in a bowl of cold water, the brightness will ramp back up again. So the light will remain at full output until the set temperature is hit, and this can depend on how well sinked the LED is to the MCPCB and the pill, the driver board to the pill, the starting temperature of the light etc.

As described, the original point of the project wasn't to develop a driver for sale, and I'm certainly not planning to make any money out of this at all.

It's one thing to offer a single-sided driver board with fairly large components for sale - that's pretty easy since I can do a single pass solder paste application, hand place the components and reflow it all at once. However it's really not time or cost effective to do this for a dense double-sided board with small components like the GXB172, and I certainly wouldn't want to offer such a product for sale where I can't ensure every build would be up to my quality standard, within a reasonable cost or time.

However if there is sufficient interest, I think here's what I can do. I'd like to fine tune the driver electricals and make sure that it's reasonably robust, work on getting at least a decent firmware going (still a bit more work to do), and optimize the BOM for production and cost. I'm not keen of the idea offering something for sale if it's not robust, but don't worry all of this will be open source regardless

Then if people like what's on offer, I can just send for an entire batch to be fabricated and assembled say in the ~100 to 1000 quantity with some sort of decent built-in firmware. I suspect once we hit the 1000 order threshold, the driver cost will be pretty affordable! 100s quantities should be decent but will still be fairly expensive I guess, but certainly doable. It'll come with some stock firmware but those interested can then do their own firmware mods.

I hope in the future more talented people than me can improve on the firmware for it and future revisions too! ^_^

But right now, I'd like to focus on making the driver actually a decent driver first, before thinking about production! It's easy to get the driver working (it's working now!), but it's a bit more work to make sure all the edge cases are covered!