Review: KD 7135 V2 drivers 3040/2280/1520 mA with mode configuration

How does mode switching work here?

The OP writes: "The mode memory snaps in about 0.5-1 second after a mode switch. I normally prefers memory after switch-off so I do'nt have to tap twice to shift level."

In a KD review is written: "I would rather have it switch modes with a single 'off/on', but it takes two 'off/of' cycles to switch modes. Not really a big deal, maybe even better. Just used to switching with a single click is all..."

Two taps to switch modes when the light is on? Or just one?

And what about memory - what does it mean that it "snaps in about 0.5-1 second after a mode switch"? Only that if I switch the light off less than 0.5 a second after a mode switch it doesn't remember the last mode or anything more than that?

Do the answers to the above questions change if I disable mode memory?

I have memory disabled so that it starts in high every time but the way I understand it is that it will remember the mode if left on in that mode for .5-1s. In order to switch modes you have to switch off then back on (to clear the memory) then quickly switch off and back on again to jump to the next mode. I'm not sure what happens if you switch modes and then turn it off within .5s but my guess is that it would start in the next mode when you turn it on again later.

Has anybody tried making a sandwich with these drivers? I'd like to make a 3 board sandwich with the 3A version and I was wondering if a single mcu would handle all 24 of the 7135s?

Thanks,

Tony

anyone tried programming the PIC12F629 with their own code? I seem to have problems communicating with the chip. I checked my connections, but all seem ok. could it be one of the components on the PCB that prevents somehow the programming signals from working correctly?

I gotta say, I love what you listed as a con. I loaded up one of the 3040 mA version with a XML U2 in a p60 and using the panasonic 3100 cells, the ultra low will run for hours and hours and hours and put out enough light to read by, read a map, check on sleeping kids, etc. If I find anything is wrong with the driver, the medium is too close to the high. If I could, my alteration to it would be to keep the high and low and lower the mid by about 1/2.

Wouldn’t a 2 board sandwich work better for our 26650trple XML lights?

You mean whole board or only AMC's

The 7135

Well, not the prettiest idea but if needed to remove a single or few amc 7135 chips i simply crushed them with pliers and carefully removed the leftovers. Works.

Probably.

I'm just exploring options. Right now I'm running mine basically DD using the UF XM-L P60 driver and according to my meter with 10ga leads it pulls around 12A (4A per emitter) at startup off of a fresh imr. I figured 9A from a 3 board sandwich would make for a nice "wow" or turbo mode leaving medium and low for normal use. Though I guess I'd be better off to keep it below 2C from a 4000mah cell so it's not limited to running imr. Maybe a 2 board sandwich with some extra 7135s to push it closer to 8A on high? Anyhow, I was just curious if there is a limit to how many 7135s a single mcu can handle if I remove the mcu from the other board and slave the "dummy" board to the board with the mcu?

Thanks,

Tony

Low i measured with 1mA, means my Fandyfire HD2010(with 26650) will now glow for ~5 Months

I find the low to be on the bright side!!!!!!!!! lol? Its still significantly brighter than the low on a D10/EX10 ramping, and about the same as the low on a D10/EX10 SP.

I have mine in a skyray G6. removed 5 of the 7135's to leave 3 behind to provide something I'm guessing to be around the 1.2A mark for an XP-G.

Mode spacing for low/med/high is not bad at all IMO, a low that is *FINALLY* low enough to call a low (IMO), a medium that is not too close to being a high, could be a tad lower, but not bad at all, and plenty of power on high. Im currently running mine without memory, and on grouping number 1, so I can get 2 sets of l-m-h, followed by the disco. Because of the lack of mode memory, this works great because I rarely have to see the disco unless I actually want to.

Pretty happy with everything except the fact that it is still "on-time" memory. Off-time memory would have made this perfect. This means that if i don't turn it on for more than a second, it will go to the next mode next time I turn it on. In other words, if I were to pick up the torch one day, I cannot always guarantee that it will start on low. This does annoy me, and is why I prefer the off-time memory, in addition to the fact that mode switching requires double tapping initially.

I bought the 3A version because I get a good supply of 7135 chips for practically 10c extra over the lower current models. I find them easy to remove too!, just use 2 soldering irons that are well tinned and a "helping hands" alligator clip on a metal base to hold the board. Heat up the heatsink tab with one iron, and the 3 pins on the other iron, and when its all heated up enough, it simply slides off, usually done with a couple seconds with no collateral damage at all.

So would I buy more? likely, but I only have 2 torches that actually need replacement, the rest already have UIs that I'm happy with.

For anybody that wants a brighter low, simply break the trace to the Vdd pin on one of the 7135s and solder a jumper from the Vdd pin to B+. This will put that 7135 on full all the time so low becomes roughly 380ma. A word of caution: Only do this if you're using protected cells as this will defeat the driver's low voltage protection.

Any change of a picture of that Tony?

Good news I just got a chance to take apart a Shiningbeam P-Rocket and it had one of these new drivers and I was able to reflash it with the standard setup I have been using on the old Nanjg drivers.

I've already got that driver installed in a light but here's a link to the data sheet so you can see the VDD pin I'm talking about. It just needs to be isolated from the mcu either by breaking the trace or unsoldering/lifting the pin and then jumpered to B+.

http://www.soiseek.com/ADDTEK/AMC7135/datasheet.htm

I'm gonna guess that explaining what B+ is too might be good.

Battery positive. If you use one of the 7135s that's on the battery side of the board, the jumper wire can be run from the VDD pin to the pad that the spring is soldered to.

Just be sure that the VDD pin for the 7135 you're using is disconnected from the trace on the board or it will backfeed the others and you'll lose all modes except high.

What are the modes settings with the reflash you did?

.002, .085, .650, 1.5, 3.15