crash-testing a MT-G2 on a copper Noctigon, graph done, mod done and repaired :-)

At 7 ampere, the Vf of the MT-G2 is 7V, or at least close to 7V. It would have been great to have 7135 type lineair chips that could handle the 8.4V of two li-ions in series, in that case my no brainer ideal driver would be a lineair driver with enough chips to deliver 6 or 7A, running off 2S-2P li-ion batteries.

What is available is one of the available 2/3 leds-in-series buck drivers. Unfortunately they are big and comfychair found that for high currents they need 3 cells in series. But as said, for actual suggestions for drivers the MT-G2 driver threads gives more useful information than I can.

(slightly off topic: I actually think Eagletac has made the almost ultimate MT-G2 flashlight (SX25L3), the only weakness IMO is that they use 3 instead of 4 18650 batteries, I expect that it can only maintain that 2375 OTF lumens of the turbo-mode for a short period when the batteries are freshly loaded)

djozz, your numbers look great. I think they are perfect for relative output numbers. This is perfect for knowing how much is too much for the emitter. That part you have nailed right on.
Very few of us can measure absolute output of an emitter or light.
The equipment to do that truly proper is way above my pay grade. :beer:

djozz,
Perhaps using the ‘9V’ MT-G2 should somewhat alleviate the issue of achieving ideal overdrive current in certain cell configurations. It takes less current to drive it as hard as the ‘6V’ (e.g. Noctigon, 4S Li-ion, 3-18V IOS buck-boost driver: 9V vs. 6V). Ergo, driving the 9V at 5A is equivalent to driving the 6V one at 7.5A.

Thanks for the sweet graphs and your time invested in them.

The 6v part works pretty nice in direct drive off 2 cells, does around 5A. The 9v part would be interesting with 3 cells, either DD or a driver that does DD on high mode only.

Based on the test above, I dared to make a 2x18650 direct drive flashlight mod, and here are some results.

I must admit that I wrecked the light already, the reflector was electrically insulated from the solder connections and led by a piece of Kapton tape, I screwed it too tight, the tape moved and the plus and minus shorted via the reflector somehow, frying the switch. Have to order a new switch now and find a better way to tighten the reflector. But this all happened after I tested the torch out, so it might be interesting to see the results. EDIT: I repaired the light by bypassing the switch, see post#36

I took this torch that I bought on sale from cnqg a while ago:

It is the Shadow TC 300. I removed the driver and led board, cleared the driver from components to use it as a contact board. I tried to make every electrical path nice and thick for handling big amps. Soldered the minus led wire around the corner of the contact board directly to the brass contact ring, soldered some solid copper wire through 4 via's of the positive battery contact plate and soldered the plus-ledwire to that, and used copper braid to enforce the conductivity of the + contactplate spring:

It doesn't look pretty but it sure can handle some amps .

So here it is with the MT-G2 on the Noctigon in place. A weak spot is the reflector directly screwed against the top of the led itself, ruining the thin layer of silicon around the dome, it does not ruin the performance of the led, but eventually after screwing it too tight the light shorted somehow. The tests were done already by then.

The first thing I did was to put some fresh loaded Panasonic CG18650CH (IMR) cells in it, and measured tail current without switch: 8.5A . I decided at that point not to improve the conductivity of the switch spring with copper braid, and measured with the switch into the circuit: 5.6A, that is more acceptable to me . I took a beam shot from three meter from the wall, the first is how it actually looked the second one is underexposed to show that it actually has a nice hotspot:

Here's a video of the first two minutes measuring ceiling bounce lux. Multiplied by 20.9 it gives a lumen indication (so at start that is 2970 lumen). I stopped the test when the head got too hot to touch.

The smoke coming off the lens are evaporating fingerprints that I forgot to wipe away .

In the middle of the night I dared to film a very quicky beamshot into the neighbourhood, it is over before you start paying attention :

After some playing around I became a bit more confident about the light, charged the batteries to full again and did a stamina test of 10 minutes at room temperature without extra cooling (even not with my hand). I measured the output:

Afterwards the light was way too hot to touch, but not blistering hot. The batteries were also hot but not that hot, the voltage was 3.74V each.

I also checked some protected no brand Li-ions (non-IMR) that came with a very bad headlamp I had to review (EDIT: removed some unneccesary grumbling ), and it still runs at 4.6 A on those (initially).

Concluding I must say the MT-G2 runs quite happily on two 18650 IMR's direct drive. Now I have to repair that switch (or I may not do that and go back to small EDC flashlights where my heart lies ) EDIT: repaired, post #36

Hope you like the results, and that the information is somehow useful to someone.

Nice djozz. I’ll have to watch the vids when I get home.

Thanks MRsDNF, and I just figured that the tail-threading of this Shadow is anodized, so I can just bypass the switch and make it a tail twisty .

Great info djoss. Emboldened by this thread, I tried 2S protected NCR18650A’s last night DD. Only pulled 3.65 amps and the emitter seemed quite happy. I know it’s quite wimpy compared to what you’ve been doing, but it was a big leap of faith for me. I only have 3 of these emitters right now and I love each one of them.

Now I’m wondering if 2S2P with the same cells would work too. It seems that voltage sag would still be the same given how much current these emitters seem capable of handling. I will it out try tonight and report back.

I like that plan, and am curious about the results, you'd expect the current to be a bit less than double what you get with 2S because the Vf goes up a bit with the higher current, but who knows what happens....

Am I reading this correct? 3K lumens with 2A @ 6v?

Nope, red line is lumens, blue is current.

Thanks, I didn’t scroll to see the legend.

So it’s 5A@6v (30W) for 27-2800 lumens if I’m now reading it right.

Well, I tried it, but I think I wasn’t getting good battery connections. I really hodge-podged it with tape and wires as I had no battery holder for 2p2s. It was only getting 4.5 amps at the emitter at that was at start up. It dropped down to around 4.1 within a minute or two.

I will try again another day.

I repaired the MT-G2-modded Shadow :-) Here is some more on the repair and the (altered) mod:

First I put a cardboard spacer around the led, just a bit thicker than the led-base. the reflector is now tightened against the cardboard and not against the led-base (the short I had appeared to come from the led-base, and that actually means that if you scrape away some of the silicone around the dome, it is likely possible to solder your led wires on top of the led, like with the XR-E).

I decided not to bother with a switch anymore, the tail threads were anodised so I shorted the switch out, it is now a tail twisty light.

Thing is, with the CGR18650 it now it runs the full 8.6A which is not very healthy in this host: the head gets bleeding hot in seconds. On switching twisting on, I measure 3400 lumens, but within the minute the output has fallen to 2700 lumens and descending more slowly from there. So to run the MT-G2 at this 8.6A a bit more reassuringly you need a much bigger host with lots of fins (well, we kind of knew that already). Some other numbers with these batteries: throw just after start-up is 29klux@1meter, power is 60W, efficiency of the light is 56lumen/W, and after a minute 45 lumen/W. This efficiency at this output is actually not bad (I have build a XM-L2 (5A tint 80 CRI) on copper board flashlight at 3A that measures -the battery I estimated 3.9V under load- 66 lumen/W, if it was cool white it would have been 80lm/W).

I tried LiFePo batteries as well: A123 1100 mAh x2. The output gets really mild on those: I measure 2.1A, with 1136 lumen at start, 1072 lumen after a minute, and stable from there on. These batteries should last for half an hour but there are higher capacity LiFePo batteries around. The host gets luke warm. Efficiency is 82 lumen/W which is really good for a neutral white flashlight.

I do not have two matching normal Li-ions apart from the two unbranded protected ones I mentioned earlier that came with the headlamp. I could just do a current reading before protection kicked in: 5.6 A. I had two seconds for the output: 2500 lumen

I did a better video with the Panasonic IMR's :

So that's all you are going to read about this mod, I will keep the light as it is now: with the IMR's it will generate serious wow from my non-flashoholic friends, with LiFePo's it is a nice and safe light to use and can even be lended out to others :-)

I’ll confirm you can solder to the top of the emitter’s pcb. You have to clean it well though for the solder to take.

djozz, thanks for putting so much effort to come out those measurement.
I am wondering why in the 1st video at 5.8V measured by power supply, current is about 2.4A measured by both power supply and DMM.
But in 2nd video, at 5.8V measured by power supply, DMM measure Vf to be 6.7V, while current measured by power supply is 4.5A.
I mean why Vf which seems correct judging the current, but why Vf is much higher than power supply voltage while power supply seems accurate in first video?

It is a problem in the power supply. When I check the voltage separately with the DMM, the voltage on the power supply sometimes reads too low, and then suddenly jumps to the correct voltage. But sometimes it gives the correct voltage right from the start of the test.

Hi djozz. Are you maybe planning to crash test the XHP-50/70?

Hi shrick, I already tried, but could not kill them :-) . I buried a few XHP50 and XHP70 tests in Old Lumen's XHP50/70 thread instead of doing a separate thread about the tests, and that might not have been the best of ideas, sorry about that :-(

Anyway: this is the thread: https://budgetlightforum.com/t/-/30860

My tests are in post #274 and post #575

On my way checking it out! Thanks for the effort djozz.