The product is great. But just a heads up to beginners who buy them. Make sure that the wires does not short, or can not easily make a short circuit to the reflector. If your wires are thick, or your soldering point is a bit high. Or the wire is not functioning as isolation, you can get a short circuit.
There are many ways to do things. But if you use thermal compound (silicone glue). Then you can first use one of the centering insulation thingies ImA4Wheelr linked too. Then change it out with one of those plastic insulators that cover the wires for maximum safety (after it has set, and is perfectly centered).
These days I prefer to not use anything with glue under the emitter if I can avoid it and I have an reflector pressing the emitter down. I use high quality CPU paste instead. Sometimes I use Fujick around the emitter just to hold in in place (but good thermal paste under it). Not only is it better for getting the heat away, but changing the emitter is super easy too.
I’m not familiar with the driver designs or the design for this one, but maybe the driver is designed to just pass the incoming voltage through to the emitter if the incoming voltage is, say, > 4.2V?
In my (unfortunate) case, I had 2x18650, so about 7.4V - 8.4V, which, from earlier post is > than the emitter maximum voltage, so maybe that, plus the theoretical driver behavior (pass voltage through to emitter) would kill the emitter?
Based on the recent messages, the XML emitter can work ok with a Li-ion battery for testing.
I’ve been wondering for awhile: Assuming that I only want high mode, why couldn’t I just wire directly from the +ve end to the +ve end of the emitter PCB, i.e., basically have no driver at all?
With the emitter PCB in the pill, and the pill in the light, that should provide heat sinking, so wouldn’t this work?
It seems like this’d be a lot simpler, and cheaper, than needing emitter + driver?
Or, maybe another idea is to put a resistor in series with the +ve coming in from the battery, going to the +ve on the emitter PCB? This could be done with just wires, without a driver PCB?
And, maybe fill the entire driver end of the pill with Fujik, to pot/seal it?
Ok, I take my original predicitons back. I guess it’s a direct drive driver with resistor and PWM for lower modes, so the added voltage got through and fried the emitter with a massive amount of overcurrent.
You could do that, but you may get very high currents with fully charged cells. It is also more complicated than re-using the driver or installing a new one as you would have extra soldering work to do as well as connecting the +ve an -ve wires to the LED.
Thanks for the comments. I was just curious, since I’m still learning. Having said that, if someone was going to do that, maybe a LiFePO4 battery would be good?
Edit: BTW, this was not completely off-topic. I started thinking about it in case it turns out that BOTH the emitter AND the driver were dead. I mean, at the end, I’d like to get a working light back again :).
Having said that, looking for recommendations for a driver, assuming I go with this emitter:
What I have found works best is to use a pair of needle-nose in the ears on either side of the LED star/base and twist to break the adhesive. This typically works better than prying the edge up; I usually end up with a bent star when I do that. Another technique is to press the driver from underneath through the wire through-holes if the base is exposed.
Lol. Yeah, I second that, for sure. Disassembly, especially, can get pretty messy depending on the light. That’s why it’s a lot easier to start over with new parts most of the time.
Do you still want the option to use 2x18650?
If you’re alright with using 1x18650 all the time, then the Nanjg 105c is probably the best way to go.
Just thought of a way to get your money’s worth out that fried led. If you keep modding lights, you will eventually get to a situation where you want to create your own led/star combo that is either not available or too expensive. At that point you can buy the raw emitter and star of choice. Then you reflow them together. Keep that fried emitter for practice reflowing. It will be worth many times its cost to you from an educational stand point.
EDIT: For some reason an s in parens creates a smiley face. Edited the above text to get rid of the smileys.
BTW, FYI, I finally got around to testing the driver board, and it appears that it is working. I was getting voltage out of 2.47V and 4.17V (e.g., if I changed mode).
You might also want to consider getting a new insulation disk for the P60 drop-in. The old one won’t work with a 16mm pcb. If you don’t want to buy one, you could just use high temp electrical tape cut to size.
I forgot :(…, and I think I also forgot the glue :(… I’ll use your suggestion temporarily, but what can I use/substite for the glue (to attach the emitter/board to the pill? Wouldn’t the pressure from the reflector hold that in place when it’s screwed into the pill?
Yeah, you don’t need glue. Any CPU thermal compound will work fine. The quality of the compound won’t make a huge difference. You just need to use something instead of nothing.
It’s probably not too late to add items to your FT order.
Between a U2-1A and a U3-1C? I dunno. I don’t think there will be a big difference.
Either emitter will do. I would probably go with the U3 in the C8 because it is a slightly cooler tint which is good for throw.
I recommend the 8 chip driver in the C8. The C8 can handle the heat and you will want the power for throw.
I have used this thermal grease for emitters and non-conducting thread connections in a good 40+ lights. It’s super cheap at about $9 for 3ozs. I haven’t seen anyone else endorse it nor have I seen any tests on it. When it comes down to it though. The most important thing is that you have a very close connection between the emitter base and the pill. So very little paste should be needed. It should only need to fill the pores and very small pits. Several people on this forum sand and file the emitter bases and pills to make sure the have flat, true contact. This is a very good practice when using high amps (3+).
EDIT: For electrically conducting thread connections (excluding the tail cap), I used the following: