Trustfire AK-47 9x XM-L Driver Defect / Repair / Upgrade / Photos

Tested the light with 18650’s this time to see how they handle the power of the KD driver.
Using Redilast 2600mAh cells fully charged, just as the light came onto medium mode, about 4.85 amps, one cell tripped it’s protection.

Removed them and tossed in 3 Redilast 3100mAh cells, the light worked on all modes. On high, tail current was only slightly low at 7.85 amps. I left the light on high for about a minute with no problems.

Hi folks,

Looks like there seems to be a problem with these AK-47’s that I had hoped would have been an incorrect assumption on my part from my original post.

I’ve got a second AK-47 in my hands that I’m upgrading for a forum member with the KD driver. All is fine with the light physically and is mint, but the stock driver I removed not only had one of the 3 resistors off, it was actually missing having never been installed.

Without this 3rd resistor, this has the effect of not allowing the light to push more than about 3 amps at the tail cap on high mode with 3 cells. With the resistor in place, the stock driver would pull well past 5 amps despending on how good the cells are.

At least the one I purchased recently, the 3rd resistor was present, just not soldered on sufficiently, so it was an easy fix. The one I’m working on now will have a 5-mode KD driver installed within the hour and pull over 8 amps at the tail on high mode with 3 MNKE cells in it. So this one will light up Broadway.

Sorry to give the bad news, but if you upgrade the driver, it’s simply a wall of light.

Have you tried to solder smaller resistors to the stock driver to increase current…?

I don’t have a spare R820 resistor the stock driver is missing. But the member is allowing me to keep it as a spare, so I’m going to see about getting some. My wife loved the output of my light, I ended up placing an order for the last AK-47 that Amazon dealer had on hand, so it looks like I’m going to have to fix a third one.

She wants to use it in a 2 cell configuration because it’s pretty compact, so that’s fine. I also have some KD 3-mode drivers on order, so I’ll be able to get the power level up to what I feel it should be.

I just receied my multimeter. By the tailcap I got 3.2a falling out quickly. In the stock driver are just 2 resistors, one is completely missing, and I thought this is a powerfull flashlight. I ordered immediately 3 Enerpower 26650 cells and a KD 3-mode driver. I can’t wait to blind myself;) I’m so thankful for all the information I got here. Thanks!

i so cannot wait to get my modded ak47 this week! thanks again richie for the helping hand with this mod.
wow, so you got the last one from that seller? when i ordered, he had 6 left hahah. and that was on thursday lol and today is what saturday . geez this thread must have gotten seen hehe

Strauss, sorry this has happened, but it has become a very familiar situation with these Chinese lights. The first run tend to be decent, but if they sell out quick and they have to restock, the newest ones usually have cheaper parts and components installed.

This latest version from Amazon did away with the GITD O-ring at the lens, no 3rd resistor on the driver, and the plunger on the tailcap spring is to short to allow the use of the shorter unprotected cells reliably. So I removed the link from my OP. I’d say if anyone still wants one of these lights, try the one from Manafont. Hopefully it will be better.

With all the different versions and changes going on with the manufacturing process for this light, I thought it would be best to post some additional photos for clarification.


This photo shows the proper length brass tail cap plunger on my first example AK-47 at .434 mm length. This plunger will allow
the use of any length protected or unprotected cells to make proper -B battery contact.


Sorry for the blurry photo. This is what the long plunger looks like removed from the tailcap. The smaller version
below it is what I removed from BLF member blueb8llz’s AK-47, which is only .278 mm long, and replaced it with the longer one.

Fortunately I was very thourough when testing his light with both the longer EnerPower protected button-top 26650 4500mAh cells
and then with unprotected MNKE 26650 button-top cells and quickly realized there was a problem. Due to the shorter length of the MNKE, which are similar to the King Kong cells, the light would flicker due to improper cell contact when moved around. I had several spare correct size brass plungers in my spare parts bin, so his light is perfect. I only wish I knew where I got them from.


This last photo is a minor issue, but still a change I didn’t like. My first version had this GITD lens gasket and blueb8llz’s had a standard black one. The size seems to be 73 mm diameter and just lays inside a recess and the lens simply sits on it. I’m trying to locate some replacement GITD O-rings because I have a second AK-47 that will arrive this week and I want it just like my original.

Last night my wife needed some extra light in a room she was doing spring cleaning in. There are two 75 watt bulbs in a fixture on the ceiling, but that wasn’t enough. She asked if she could use the AK-47. I had just finished reinstalling the stock driver in the light because my encapsulated KD driver was given to blueb8llz. So I brought it in and turned it on high mode (5.75a). It appeared as if a 300 watt light was in the room now, which was much to bright. Naturally you can’t run most any light on high mode for long, so I switched it to medium, then tail standed it on a table and left the room. She was really happy for about 15 minutes and returned with the light saying it went out.

I assumed one of the EnerPower cells tripped the protection, and check them but all and were fine. Further investigating revealed the +B spring on the driver annealed and the driver also fried. Damn, I couldn’t believe it.

Lucky for me blueb8llz was generous enough to allow me to keep his stock driver so I can fix mine until my new KD 3-mode drivers arrive. Since his driver was way underpowered, I removed one of the R820 resistors from my blown driver, soldered it onto this replacement driver, then removed the crappy Trustfire +B spring and replaced it with one of my heavy duty gold plated ones, and also added some copper braid.

The light is back to running great with a 5.75a tail cap current, so I’m happy again. I’m certainly going to learn how to encapsulate these drivers. Clearly it makes a big difference compared to leaving them as they come from the factory.

I think I’m beginning to see why the factory now makes them 3a on high mode leaving out the 3rd resistor. This stock driver is worthless driven to what it should be and simply frys.

Finally getting around to posting a few photos of the KD driver Richie used. His first was a dud, this one I had on hand and potted for him. Unfortunately for Richie, we discovered it is 5 mode, so I’ll be doing another for him.

stock

upgraded wires

I put 26650 size battery wrap around the driver and shrunk it, only around the upper board.

I filled that w/ epoxy, let it settle, then shrunk the rest of the wrap. This is the first time I’ve done one this way, and I only did that because I didn’t have the host in hand to actually pot the driver in, and I wanted to be sure I didn’t create any fit problems

after peeling shrink wrap off and trimming excess off the top

next time I’ll be adding alumina. it helps with heat transfer, but also helps eliminate air bubbles.

I didn’t with this one because I was concerned about the mixture being to thick to flow between the driver’s components. Since then I’ve found a way to add the right amount of heat so that the epoxy flows well once poured in place (on top of the toaster oven :stuck_out_tongue: )

Hey Dave,

I didn’t know the KD driver had that huge capacitor. I think someone mentioned to me the new drivers don’t have that and have been changed. I assumed the driver was similar in size to the driver we used in the Torchlite, but could never figure out when the encapsulated driver was so tall…LOL. Now I have to wonder how these two new KD drivers I have on the way will perform. Hopefully they didn’t screw them up. Thanks for the great photos.

You’re welcome. I need to add some lighting, or make a light box, but at least I stopped using my phone :slight_smile:

I did notice the change in their photos, but I sort of assumed it was a mistake. I guess we’ll find out soon!

If the driver had been mine, I would have removed the rubber insulation from the toroid and capacitor, for better contact w/ the epoxy.

There’s no need with the toroid in a horizontal position - I can just poke a hole in the insulation it so the center fills up with epoxy.

I wonder if the new version is as good?! I suppose your ceiling bounce and current measurements will give us some idea if efficiency has changed.

Well, my second AK-47 arrived today that will be used in a 2 x 26650 configuration for my wife. This one will be left in its stock form because I don’t want my wife using a light pushing the power of a KD driver in her hand. When my wife walks past our Skyray KING to use the AK-47, you know it’s because the light output is pretty incredible. However, her AK will be less bright, but will still outshine the Skyray King.

As with the other lights I received from this same Amazon.com dealer, this one also arrived in mint physical condition. Amazingly, this one had the longer brass tail cap plunger, so I didn’t have to upgrade it like on blueb8llz’s version. But it didn’t have the green GITD lens O-ring, rather the standard black version, but that isn’t a big deal. I may make the time to replace the Trustfire +B spring and add some copper desoldering braid for better contact. The Trustfire springs typically are crap and very thin.

As for the driver, on high mode with 3 x 26650 EnerPower cells, it started at about 3.45a and began to drop off from there, which I’m used to at this point with these stock drivers. Using 2 x 26650 cells, it started at about 4.15a and fell to 4a and stayed there. So this driver is also missing the 3rd sense resistor, but it’s going to remain that way.

On 3 cells, it grossed about 38 watts before inefficiencies

On 2 cells, it grossed about 30 watts before inefficiencies

All the threads were excellent and generally I have no complaints. Hopefully this stock driver will last longer being less stressed using 2 batteries. We’ll see :~

Just couple of photos of it next to my original version.

In case anyone is interested how the Trustfire AK-47’s series wiring looks under the reflector, here is a shot of it. Below that is a photo I borrowed from member, FMcamaroZ28, showing his Skyray 9x T6 factory wired in 3s3p.
Funny thing is the Skyray driver would not work in my AK-47. I’m sure the grounds where fine when I installed it, but perhaps the different wiring configurations between both lights make the drivers non-interchangeable, but I’m not fully certain.


AK-47 factory wired in series. Also, the brighter color red wire at upper right side and black wire
are what I also upgraded on my example to 22g silver wire down to the driver.


Skyray 9x T6 factory wired for 3s3p. Photo credited to FMcamaroZ28.

I couldn’t leave well enough alone on this second AK-47. So I opened the reflector and de-soldered the stock positive and negative wires and replaced them both with 22g silver wire down to the stock driver. Replaced the Trustfire +B spring with one of my heavy duty gold plated variety with some copper desoldering braid added to it. When all that was done, I mashed up some thermal conductive pad and inserted the pieces under the electronic driver components on both sides using this stuff from IOS.

http://intl-outdoor.com/thermal-conductive-pad-10x10mm-5pcs-p-411.html

Trying it with 3 cells, tail cap amps remained the same, so perhaps there is some regulation going on with 3 cells. But on 2 cells, whether using MNKE or EnerPower’s, tail cap current increased from 4 amps to 4.60 amps on high mode. Using my Trustfire 26650’s, current decreased to 3.75a.

So all of the above, or perhaps simply the two wires I replaced increased the output on 2 cells from 30 watts to 34 watts. Not great, but seems okay for the stock driver.

LOL…the KD driver outpowers the stock driver by triple the amount.

I have returned my defective k47 in exchange for a j18. Manufacturer rated at 8000 lumens but seemed to be only as bright as 3000 plus lumens only. Anyone knows how to make the light produce 8000 lumens as rated. Maybe a change in board or something?

HI Dexter, we live very close to each other…
Maybe can meet up drink tea…haha

Ur ak 47 is not defective. The numbers is printed saying that light theoretically if driven with full power can give out so much lumens.
However, the highest for ak 47 is about 6000 lumens as modded by Richie. J18 is bout 5000 lumen by Tom. All using diff type of batt to get it.

You know how to modify lights? If so, can help me mod my j18?

In theory i know. (Just follow what others do). BUT in actual fact, i have not done it b4.

Dexter,

Member TomE has been doing a great job upgrading his J18. This is one thread below, but I’m sure he has a better one. Best to PM him and have him refer you to his upgrade thread.