BIG KD driver questions... TR-J12 / J18 mods?

You can parallel wire any number of emitters, odd or even. It doesn’t necessarily make sense due to the varying Vf per emitter which can cause current hogging, but it can always be done.

I would double make sure the drive will fit and is a simular size is what i ment as its not a cheap driver. You can parallel wire odd and even numbers.

I posted detailed measurements of the 9xLED KD driver in another trj12 thread

Hi all:

Can some kind soul put up some pics with instructions on how to swap the driver out of the TRJ12 with the improved one from KD? Soldering rookie here that has absolutely no idea about electronics.

Thanks.

I started this thread little while ago, just to scope out a mod for my aliexpress J12. Out of the box stock, my light at 8.4V pulls around 28Watts, and delivers ~1.35A to each LED. The T6 XML will emit ~475 Lumens at 1.35A, for a total of 2375 bulb lumens. It gets warm at the 5 minute mark and needs 8-9 minutes before it reaches "hot to the touch" state. It needs another couple minutes to reach "painfully hot" to the touch. These numbers extend a little if I hand carry the light, with air circulating. So the stock driver is a moderate driven design, I think the sock light is a well thought out balance between output, thermal stability and power consumption.

According to KD specs (which look accurate) this new driver will pull ~55 watts. Factoring in an 80% efficiency (typical for FET-based drivers), this driver should deliver 44Watts to the 5-XML array. That equates to 8.8Watts & 2.6A into each LED. The T6 XML will emit ~800Lumens at 2.6A, for a total of 4000 bulb lumens.

Benkie and I have had some excellent J12 discussions in his "calling all J12 owners" thread here...

https://budgetlightforum.com/t/-/8573

...Where his light draws 55W at 8.4V and pushes his LEDs hard at ~2.5A each. So I think Benkies J12 is a good approximation how the J12 will perform with a ~55W driver upgrade. The idea of HOT 18650 cells (being pushed at a ~6.5A draw) and a HOT heatsink after a couple minutes, doesn't really appeal to me that much. I generally use my J12 as a constant ON, task flood-light kind of thing. So for me a cooler running ~2000 Lumens is very nice. Now if I could re-flash this KD driver to add another medium-high output mode at ~60%, and set the PWM frequency higher that would be really nice.

I would consider 5-XML pushed with 2.6A each more of a short burst only kind of light, like the DRY Turbo.

Question to others regarding 18650 safety.... how safe is it to run two cells in series with a ~6.5A current draw? Is there increased risk of a flame/vent incident using 18650 cells in a ~2C (+) application?

thanks!!!

Ask at KD, they might reprogram new one for you. They did it for me for tripple XM-L. Not exactly they way I wanted, but still got 4 modes and no strobe/sos ;-)

Greetings, this is my first post, but have been watching for some time. I did the conversion using the KD 5 LED driver a month ago. The biggest problem was the circuit boards were very slightly larger than the allowed opening on the light. I gently filed the diameter down and got it to fit. Overall the light operates pretty good. It's definitely brighter than my DRY in Turbo mode. I'm measuring 17.2 volts at the head. I didn't change out the switch. I figured I'll just replace it if it goes. I'm using 3 batteries. With two batteries the output is visually less.

Cheers

Hey thanks for taking the time to reply and welcome to BLF!!

How hot does the light run at 8.4V?... would you say its "painful to the touch" hot in 1-2 minutes or does it need more like 5-6 minutes to reach that level? With 2 cells on HI, could you place it on a table and leave it unattended for 5-6 minutes? If its visibly dimmer and cooler running with 2 cells then thats a big plus for me.

Also how is the PWM at lower modes? is it visible at all?

thanks again!!

Actually the light functions better with 2 cells verses three. Even the KD spec states to use 2 cells. Anyway I don't detect any PWM at the low setting at all. As far as heat build up, I'll have to check that out tonight and see what I get.

Cheers

Welcome to the club, brightlights!

Aloha and welcome to BLF brightlights!

Is there any need to “pot” the new driver? How would one do that, just tank the driver cavity full of thermal grease or arctic silver and let it set like concrete???

brightlights: Can you clarify if the new KD driver runs better on 2 cells or 3 cells because in post #19 you said it runs brighter with 3 bells, but in post #21, you said it actually runs better on 2 cells??

FWIW, I will only run mine on 2x26650s anyway even after I mod it with the new driver.

Greetings, using three cells, some of the KD driver functions didn’t work, specifically the emergency flashing mode. It had no problem running in high, medium or low mode however. With 2 cells, the driver functioned just as advertised at KD. All modes including flashing. I just noticed that with three cells the high mode was brighter than with 2 cells. With regard to heat buildup, 1-2 minutes wasn’t a problem. At 5 Minutes, it was getting pretty hot. I discontinued the tryout because I didn’t want to burn things out. One other note, you can run the flashlight on medium mode as long as you want. I found the brightness to be about equal to the high mode of the unmodified flashlight. As I recall I originally was getting about 14 volts on high mode on the original light, and was pulling about the same on medium mode with the new driver.

Cheers

sounds ideal to me…

Great thanks for the feedback!! I think I’m going to keep mine stock for now. It wouldn’t take much to change my mind though.

Brightlights, do you have the ability to create regulation plots, light output measurements, current draw measurements, etc…? I’m very interested in doing the same mod, but I’d like to know more before doing so. Thanks.

Agreed.

With 8.4 volt on constant high mode, I'm really pushing the limit of the unprotected NCR18650A's panisonic cells as I'm at the maxim claimed recommendation of 2C max (6.5 amps) discharge. Ive proven in my old discharge test's of the same cells they will do 7 amps from 4.18 to 2.5 volt with out venting but they do get dam hot but with in there safely limits, but the light gets hot and the batteries are inclosed not opean like my battery discharge test.

On 26650 cells your still up around the 1.6C mark and with my protected trustfire cells going from my battery test,s i will only get around 15 Min's of constant running on high till the battery protection trips, the battery protection tripping is not an issue as i would not run the light for 15 Min's on high on 8.4v due to to much heat, but on 12.6 volt no problem as my current draw is low approx 2.2 amps.

So for short burst on high I'm not worried, but for longer runs i use 3 x protected 26650,s with 0.6C discharge with about 2 hours run time on high, this is how i normaly use the light.

Well, you could use A123 26650 LiFePO4 cells. Only 2300 mAh, but you can pull 100C from them! A 4S pack will start a car 20 times in –20C weather. I’ve also seen them charged at 100C (under 45 seconds). They also make a 1100 mAh 18650 cell. And a wicked 20 Ah pouch cell. And a few other sizes of larger format cylindrical cells.

In my case i have 7.4v 2200 40C constant lipo cells that are cheaper and i use them with my daughter for RC, they would do the trick no sweat, but on 3 x 26650 or 3 x 18650 0.6C or 0.8C draw from the light is fine.

Wow…. I want one now. :smiley: