My first mod of the new year. ZY-T08 with DRY driver

This is a Small Sun ZY-T08 with a DRY driver. Yep you heard that right. I modded the battery carrier to serial, wiped the stock driver clean to use as a contact board and installed a XM-L2 de-domed on one of the new 20mm Round 2mm deep copper stars from LCK. I did NOT do a heavy gauge wire mod or a spring mod and I am only running sony unprotected batteries and not the Samsung 20r’s. For numbers I am showing 1,250 lumens and 200Kcd. Very nice tint and I love the four modes. Obviously I could bump these numbers up even further by using the Samsungs or putting heavy gauge wire on. But I’m happy with the numbers were they are. I think to keep the light practical and useful this is pushing it hard enough.

It was a bit of a goofy mod for me trying to guess what the amps would be with the smaller gauge wire. I know with heavy gauge wire the emitter would have lasted about one second.

Happy New Years everyone and thanks to everyone last year for all the great tips and tricks.

Congrats on the first posted mod of 2014 , nice!

So what are those cheap lck copper boards like: do they have direct thermal pad connection to the core?

I have a DRY driver on order because why not; what's the story with these things, using them in setups other than 3 cells/3 LEDs? Is it truly direct drive on turbo, or just very high current? Is it still bucking when Vin is over Vout?

+1 on that. I have three dry drivers I haven’t tried because I haven’t had a host with a three cell, three emitter configuration. I’d like more info on this drivers uses with single LEDs and two cells. Will it power three emitters with two cells?

Not 3 LEDs in series from 2 cells, it's not a boost driver. Series cells with parallel LEDs, I don't know.

This driver has been around for a while and it seems we don’t know a whole lot about its capabilities.

You can buy a DRY and check out what are the features with it.

DRY used to be a lumen monster which was frequently discussed around here in the year of 2011. I owned a DRY before but I broke it due to my brutal handling, and did not take any measurement that time. It has both 3 cells and 3 LEDs in series. It is a nice light except that it comes with tail switch and having visible PWM to some people, other than that I think its modes are better than most budget multi-emitter lights.

Would like to know too if the DRY drive can work as a buck driver and offering real direct drive.

Well, real direct drive would be A Bad Thing if you have fewer LEDs in series than you do cells in series.

Bare driver available here: http://www.cnqualitygoods.com/goods.php?id=1349

Thanks. It looks like direct copper to me. I didn’t scratch up the middle to see but I did scratch up the spots where you put the wire. It was copper without anything on it.

Cool. I’ll be glad to see what you find out.

Did you do a continuity test on the thermal pad to see if it is direct? I hope to see a few of these pretty soon, have no clue what shipping times will be like for em.

I didn’t even think of doing it that way until you asked.

When using my DMM it’s not the same as a sink pad. There must be some kind of layer between the middle and the copper itself.

Learned that from Comfy when I was modding my S2200. Now I always check.

Another small but useful tip worth knowing. It would never have occurred to me to test it with a DMM, although it’s basic common sense.

Looking at those 20mm copper stars here: lck-led.com/20mm-round-copper, and they are really cheap at 75 cents each, 2mm which is quite thick, and it's a true round, not star shape, so even more mass. The bottom though doesn't even look flat. I got a couple cheap copper stars from FT and they had the dielectric layer in them as well, and also a non-smooth bottom.

I'm surprised you are getting that much power out of this - interesting...

You could try sanding off the thin copper thermal pad and dielectric layer below to expose the copper core.
I’ve done this on a few of these cheap KD copper pcbs and reflowing works fine even with the small gap that’s created between the bottom of the emitter and the copper pcb, just have to use a bit more solder paste to fill in the gap.

Don’t know how big the benefit is compared to just using the stars as stock but I’d imagine even a slightly thicker solder layer is going to do a better job of conducting heat than having a dielectric layer in the way?

The bottoms are not that bad and I am using that Deltabond epoxy so I imagine that helps.

Thanks for the suggestion. I might do just that if I decide to go any higher amps.

I wonder if anyone has actually tested the ones with the layer?

Not sure, just heard anecdotal stuff really. Apparently the layer is what really hampers aluminium pcb heat transfer rather than the material itself so I’m sure it’s doing a lot of harm here despite the copper right below.
I doubt it’s anywhere near as good in stock form as a direct path pcb for that reason alone, might still be somewhat better than a standard aluminium pcb though.

In any case it’s a cheap option and I feel happy using them at higher amps as long as the pad is ground down.

-

Oh and if you give it a go make sure to grind through until you get past the insulation layer, don’t just stop at the first sight of copper thinking it’s the pcb core.

That tip doesn’t apply to these pcbs since the exposed contact pads/traces seem to be bare copper already, but the other type have a slivery metal coating (nickle or tin or something?) covering the copper contact pads so you’ll see copper peaking out once you’ve removed the plating. That’s not the core though, just the trace layer. Under that copper layer is the insulating dielectric and then finally you’ll get down to the true copper core.

I have found that dielectric layer to be very much like G10 High Pressure laminated fiberglass. A thin layer, but very tough still and an insulator. What are they thinking?