Smiling Shark SS-5039 1 x AAA size flashlight, modification notes, Now 3 x 7135 with modes, on a BLF Tiny 10 board.

My jewelery class teacher was not mathematically inclined. I put some on the brass plate and put my propane torch to it, and it didn’t melt as it did with the acetylene/air torches we used in class. I think the alloy is mostly silver and zinc. It works nicely for jewelery because the zinc, which lowers the melting point, oxidizes or evaporates or something, so the previous solder tends not to melt when you solder on something else.
For the key ring eye, it might have been stronger than tin lead solder. It is also available in various melting points.
It should not be confused with ordinary “silver solder” which probably has no silver in it.

I got it to go with only one original spring and thicker wires. (That spring is now collapsed.) It read 2.4 A with a 4.0 volt cell. It has multiple problems about the screw lengths and especially about the (x) driver board (I think) having too many conducting edges. It worked long enough to prove the concept, but is not on my key ring.

The jewel thief driver boards have more connections than are actually used for the two components. They are cut down from a board for a bigger and more complicated driver. There are half vias around the edge for connecting the bottom outer negative ring with a top outer ring that has been cut off. These may be one of the problems with the twisty. They come off with a triangle file.

I think you’re talking about jewelry “hard silver” brazing — that comes in several different temperature melting points so you can assemble something out of several different pieces, by using lower-melting silver braze for each successive joint.

This is a useful discussion thread; points out the temperature range you need (much lower than what oxy-acetylene produces):

Here are some, I don’t know this supplier but it seems to have a wide temperature range:
Welding Supplies | Linde Gas & Equipment Inc - formerly Praxair*

This torch is often used with it — it’s oxygen + propane (I’ve had one for decades): http://www.smithequipment.com/files/pdf/spec_sheets/Little_Torch.pdf

Thanks. As I said, what I have melts with air/acetylene and not with air/propane. I could use some that does melt with air/propane but hotter than the 60/40 solder I have. I could have used that today on my second direct drive conversion, but what I really lack is practice. I have done welding, but oxygen/propane would be too hot for the sort of things I am doing. I don’t think I want to buy anything else at this point, as I have put out enough for leds, stars etc.


They seem scolloped or almost cusp shaped, cut with a round end tool.

The shortened one is working now. With an SP-G2 it has a bit more throw.

It draws 2 amps. charged to 3.8 V, with only one short spring and 22 gauge wires. It has a 10 mm. aluminum star, so I don’t trust it long with 4.2 V.
Cree XP-G2, 10mm Round MCPCB, R5, 3D Tint (5000K) from Illumination Supply.


There is a smaller inner copper tube bent and soldered to the outer one.

Nice mod. I don't know why I have a soft spot for these cheapo lights. This one reminds me a bit internally of the "Police" lights I modded, but threaded head and nice texture on the body.

Does the pill, reflector, lens come out the front of the light?

I'm kind of confused by the various links towards the beginning of the thread. What is the best source for this light with the interesting tube texture?

The window (lens), plastic reflector and plastic pill with fiberglass star and driver can be pushed out the back of the head by pushing on the window. I use an AAA cell.
I have been getting them from Banggood, as that is the cheapest I have found them. The 5038s are from http://www.shenzhen-wholesale.com, which is an odd site that tells you how much the shipping will cost by emailing you after you place the order. BangGood has them cheaper also, though.
One that I was carrying on my keychain for a while shows slight traces of silver color on its edges. So far, I have been able to remove and replace the clips without scratching the anodizing.


A split is not an important defect, because it is not in the main heat-flow direction.

At 4.1 V charge, I read 2.9 A at the tail. Probably more without the multimeter leads.
I also am carrying a soft steel knife that needs sharpening frequently.
I am seeing what appears to be thermal droop when it is fully charged. This one is not direct copper.

Checking my history with them, Banggood sometimes sends an email confirming an order and sometimes one notifying one of a shipment, or sometimes both or none. The orders history at the web site appears to be complete.

Thanks Fritz. Gonna get me one from BG.

The SS-5039 is listed with a picture on http://en.cnsmilingshark.com so it appears to be legitimate, except that the size listed is impossible with an AAA cell.

Of 5 Nanjg 102s I received from Fasttech, 3 have the top flange of the inductor’s coil broken. These ferrite devices increase the field induced by the coil. The field goes through the center of the coil and wraps around to close the magnetic field lines. This core does not have ferrite around the full circle, but these flanges at the top and bottom spread the field out where it has to go through the air, which reduced the maximum field that passes through air and therefore reduced the field energy, which allows a stronger field, which induces more current.

Broken off pieces of ferrite can also be seen. They came in a soft envelope, which is probably the problem. I am hoping they will replace them and send them in ridged boxes in the future. I have 5 more on order that may arrive in similar condition.

I finished the slim hybrid with one of the two good drivers, an XP-G2 R5 1A, a thin aluminum star and a slotted copper pill. It draws 1.8 A at the tail, with a Ni M-H charged to 1.3 V. This may hopefully correspond to the 0.55 A at the led that 102s are rated at, or a bit more.
I also finished a 5039 with the other good driver, similar star and led and a long enough pill that it needs only one spring. (I should see what I can do about the tail springs.)
I tried two (of the damaged) drivers in parallel. I read 2.8 V with one driver and 2.9 V with two, into an XP-G2. I suppose that the duty factor is less than unity and that those are some sort of averages. That is better than the two jewel thieves did, but, glancing at the electrical characteristics, it may not be worth the trouble of fitting two drivers into a 5039.

Sounds great. You take any pics of that mod on the slim one?

I got some of those FT drivers, but I never inspected them. Too late for me to check them out now though. Thanks for the heads up.

If you prefer Ni-MH like me, you may want to check out this driver in Post 10 that DenBarrettSAR endorsed. I got some, but haven't tried them out yet. Probably will try on my SS's when they arrive.

I think it is unlikely that I could get a 17 mm. driver to work cut down to 12.3 mm. When I worked with electronics before, the parts either had BNC connectors or wire leads.
I’ll try to take pictures again of the next mods. I have three still in their boxes. I am thinking of giving most of them away with Energizer (Eveready) lithium cells. I think one of my brothers can use a direct drive one.

I use a spot of duct tape to protect the led when it is already mounted when I file down the star. I clean it off with mineral spirits.

> this driver in Post 10

Wait — it says 3 to 5 watts, out of a single AA cell — how?

Fritz t. Cat wrote:

I think it is unlikely that I could get a 17 mm. driver to work cut down to 12.3 mm.

I wasn't thinking. Forgot these are AAA lights. Sorry for the confusion.