KD 7135 V2 3040mA LED Driver with custom modes
A while ago I contacted KD and asked if they can make a driver with custom modes. At first they told me an absurdly high number as a minimum order, but after a some persuasion they agreed to make it if I buy 5 (or more). Some people report that ordering 2 is enough for KD to make a custom version.
Here is my review of a custom made KD 7135 V2 3040mA LED Driver.
I chose mA settings based on Match's chart from his XM-L lumen test: https://budgetlightforum.com/t/-/2346
If you are interested in a detailed review of the stock driver as programmed by Kaidomain, look here:
Review: KD 7135 V2 drivers 3040/2280/1520 mA with mode configuration
You can order the stock driver here:
To order a custom version, you have to contact KD and specify details.
These are the mode sets I ordered:
Modes #1-4 SHOULD have been a significant change in brightness from one mode to the next, with a firefly mode, very low mode, etc. The other mode sets follow roughly the same logic.
I have mostly tested mode set #2, in a flashlight, these last few days since I got the drivers. For whatever reasons set #2 isn't perfect... but it is close. Read on.
I cannot reliably measure current, or so it seems. Either my DMM has gone bazonkas, or my cells are all crap, or the driver isn't really 3040mA. Or all of the above.
I believe my DMM is the culprit because it used to measure quite higher currents until recently, up and over 2.8A in some cases. Not so much lately.
It's possible that KD didn't respect my specs, but my measurements most certainly aren't precise enough to even suggest that.
I've also established that I should fully charge my cells before taking measurements. Who would expect different results for current measurements if it's truly current regulated without PWM? One would think that 200mA is 200mA from 4.15V as well as from 3.8V. Guess again, it really looks like controller was programmed in percentages, and actual current values change as cell gets depleted. :(
Mode set #2
Since current isn't reliable I'll put emphasis on the modes description in the order of appearance. That is how they appear to me, and in the end eyes ARE what will be watching the LED output. My current measurements are very rough.
- Wow! Ultra low or firefly mode. In complete darkness this is a useful amount of light and could go on for ages. Measured about 4-7mA, depending on the used cell. Still not very pleasant to watch directly at the LED, though possible.
- Nice, brighter, but definitely still a proper low. Measured about 30-50mA
- Ever so slightly brighter than 2. Wtf? The difference from previous mode is almost useless. This is still a low mode. Measured about 70-100mA.
- Nice medium, useful. About 220-400mA.
- High. Decently brighter, good enough for normal high. Measured 450-800mA.
- Turbo. Full blast. Bright as XM-L goes. Measured 1.9-2.4A.
- Fast strobe. Could have been a bit faster, like 15 or 20Hz.
- Slow strobe, nothing wrong with it.
- Light on for 1s every 10s. Cool beacon.
- Light flashes 3 times within 1s, every 10s. Even cooler beacon. :)
Now, from these current measurements you can say something is off. I have to stress this: current was measured with a cheap DMM, which recently started giving lower numbers than it used to. This was measured with two MF Trustfire 3000s, at 3.9V and 4.10V as well as a gray protected UF 2800 at 4.05V. Current was highest with the first cell, lowest with the last (on same modes). All are very budget cells.
Take my current measuring results with a big grain of salt.
And more text
This was the description of the mode set #2, from firefly upwards without memory, of which I had highest expectations. Bench tests are never real life tests, so I wanted it INSIDE a working flashlight that I use.
I'm glad I took my time testing this in as many real usage situations...
Other mode sets
I've ran short tests of other mode sets.
Sets #3-4 are same, no disco modes.
Sets #5-6 are VERY useful all-round modes without disco. Idea behind those was the same as modes #1--4, but with slightly different numbers and less modes in total.
Sets #7-8 are still untested, but the idea behind these was to have strong light for cases when you don't need low modes. Lets call them High and Turbo modes.
Mode memory activates (or resets) in ON mode. I'd prefer if the mode got reset after I turned it OFF, rather than having to keep it on (in any mode) for a few seconds, but I'll live with it. After mode memory sets in (few seconds of not changing a mode) you have to press clicky twice to go to the next mode.
I LOVE the firefly mode.. Yesterday when going to bed I left it on in firefly mode on the night stand.... then in the middle of the night I switched to low (1 above the firefly) for reasons I can't grasp at the moment (!?) and again fell asleep.
I was a bit worried today when I got up, but after measuring the cell voltage, it reads 3.93V. It's been a while since I charged this cell, and most of my tests were conducted with this one cell, so it was below or very near 4V yesterday before I went to bed. I love it.
It's actually really cool to see the firefly mode in high quality darkness. Not having a firefly mode was one of my biggest negatives about any of the other lights I own. Not anymore. :)
You can use the firefly mode to light up inside of your mail box, in complete darkness, while holding your flashlight right next to your eyes and still preserve your normal vision.
- crappy/useless mode number 3 ("150mA") or too high mode number 2 ("30mA")
- mode remember/forget only after few seconds in ON state
- with mode memory enabled, two clicks are needed to switch mode
- make mode 2 (after firefly) about 18mA
- make beacons on -> off and on -> strobe (instead of off -> on and off -> strobe); this was my mistake when writing specs...
On the positive side, if you like your LED flashlight to gradually increase brightness, this custom modes driver is just perfect.
I hope my review helps someone. I'd love to hear if you order something similar!
And finally some images!
My captions are above images.
KD 7135 V2 3040mA LED Driver with custom modes with shortened spring
XM-L emitter in firefly mode (on rougly 7mA)
Ultrafire V6-T60 in firefly mode
Ultrafire V6-T60 on firefly mode vs. Sipik SK68 (1 mode / hi mode) on a 14500
Ultrafire V6-T60 on firefly mode vs. FandyFire HD2011 (low mode)
Images are not to be taken as reference, they were shot on semi-automatic mode. I wasn't trying to establish no realistic representations, just a short comparison.
I'll be happy to post more impressions as I discover new stuff. I still haven't tested thermal protection nor low voltage warning. I wonder what it'll do on firefly mode. :D
Good luck driver swapping!
*turns on Beacon2 and smiles*
*turns on firefly and smiles*
*turns off all the lights and has heaps of fun*
*thinks about growing up and decides to do it tomorrow....*