Integrated Emitter and Driver PCB?

I stumbled across this old thread with 3 XM-L emitters and a driver on one PCB, and love the concept. I’m working on a project where that thin profile would be amazing. So why don’t we see more of these? Do people make PCBs with the driver and LED and I just haven’t been able to find them?

for high performance drivers/lights the heat might be an issue, otherwise it is certainly possible. I have seen all-in-one pcbs on aliexpress and kaidomain.

Olight and DQG are a couple companies that use this design, among others. As was noted, heat is a concern, you can’t push them very hard (unless you have someone make a version that is also DTP, which doesn’t exist yet). They aren’t as popular with the modding crowd because your host has to be made/modified for it, they are rarely a drop-in replacement. You still need a contact board for battery+ in most lights, so you might as well have your driver there too.

Right, heat might be an issue, you have to make sure to use large copper planes and lots of vias. In that thread he claimed thermal performance in between an aluminum and copper star. I can’t find anything on Kaidomain or Aliexpress, I might just be using the wrong keywords. Mind posting some examples?

My contest light for this year has a three-channel driver circuit with pads for 1+4+16 AMC7135 chips and 6 XP-size LEDs. I also made a board for those cheapie 9-LED lights to upgrade to XP-size emitters, with a driver on the bottom. Both of these were recent (contest hasn’t even been closed out yet).

With the ability to get a thinner substrate and 2oz copper layer in PCBs from OSH Park, now might be a good time to experiment to find out what can actually work. We also have had DTP board manufacturers hanging around here willing to make custom boards! A good design, on a DTP board, could be a reality with just a little bit of work from an interested party. Those are going to require a MOQ around 100 pieces, though. So, if you’re just wanting to buy one or two, it wouldn’t be a solution for you.

Something to think about. :wink:

Oh cool, didn’t know about the new OSH Park boards. I might try that out if I ever finish my current projects. Until then, it looks like I’ll be ordering an XM-L2 and LD-4B.


Funny to hear about heat transfer problems again and again. Just use DRIVER and you wont met them.

I’m not sure what you’re trying to say kiriba.

Of course when a company designs these they keep the limitations in mind and they should be very reliable. We are saying (for example) you can’t integrate a FET driver and emitter pcb together and expect the components to last very long.

FET is not driver, it doesnt provide current limitation.

A mosfet alone, no, but a PWM’d mosfet does limit current (although somewhat crudely) and is the most common type of driver for modding. I would expect you to be familiar with this; it’s what most of your customers are using with their copper heatsinks/spacers

Last time I asked on CPF LuxRC modules are not DTP, but because they are driven around 2-2.5A there wasn’t the need for it. Also the driver is many times more advanced than our self designed FET drivers with components able to withstand the heat… main reason they are so ridiculously expensive.

I use them too, they are small cheap piece of power. But they dont limit it. You will change cells with new one which have lower resistance, make spring bypass and current in all modes except 7135 only will raise. Same thing with new leds that have lower voltage drop, more leds in parallel and etc.
Light engines are on the other level. If you are making it you should undestand that it will be used in different configs, with different number of cell in S or P and different sizes. So it should work stable in all situations.
Im trying to say that:
-here is no man that is able to make 20mm triple/24mm quad/35mm quad light engine with buck driver which is as powerfull that will burn leds (if there is good dielectric
-very often members are using configs that provide MUCH WORSE heat transfert that it could be with alu non-dtp board. For example, lots of you solder your best copper dtp boards to brass pill. Thus is much worse neither there was alu board and alu pill/shelf.

Yes, there is about 2Amps…in serial config :smiley: .
There is no need in 99% cases. Last one - big, heavy host with thick shelf and small led. All this 50W builds in small thin hosts have much more probability to burn due other reasons.

I very much disagree with this, unless you are referring to DTP aluminum. A “normal” aluminum mcpcb will fry an emitter in direct-drive within a few seconds. A DTP board on a brass pill may not be as good as a copper pill or even aluminum, but your emitters will be safe in the short-term.

But I’m a little off-topic now.

Short answer for the OP, if you find one being sold as a finished product, it is probably just fine and reliable. I was thinking more from the standpoint of designing your own: you must be mindful of the heat.

We cannot manage with results that consist word “direct drive”. It means nothing.
Hundreds of members have bought alu non-dtp boards made by dsche. Could anybody show me burned leds over them? Maybe somebody wants to test one?

i thought direct drive means no buck or boost, but direct (usually current limited by 7135)

…and what’s the point in a no PWM-ed FET driver if it just switches the battery full on anyway…?
(or am i (probably) missing something…?)

7135 driver called linear.
FET one makes what you decsribed in turbo and pwm turbo current in other modes.

ok thanks

believe it or not, some people actually do wire a battery directly to the emitter with just a switch. truly direct drive. And whether you agree with the accuracy of the terminology or not it does have meaning. “Direct-drive” will convey a definite concept to 90% of this forum when said.

I also have a few of dsche’s alu quad boards. They are a different thing than the cheap alu mcpcb’s that most emitters come pre-installed on. I am discussing the majority of alu mcpcb’s, not the few special exceptions.