TK's Emisar D4 review

The copper can hold more heat, which allows the light to run on turbo for a longer time before step-down. It won’t really increase the brightness post-stepdown, but it should at least last a little longer before reaching the lower level.

OTOH, wicking heat away somehow will make a lot more difference than using a copper pill section. One of the most entertaining tests I’ve tried is to touch the pill with an ice cube while thermal regulation is active. The ice cube quickly forms itself to the metal, while disappearing. It looks almost like what happens if you touch cotton candy to a pool of water.

Anyway, touch some ice to the pill and it can make the entire light get cold to the touch… pretty quickly. It shows just how good the thermal properties of aluminum are. Meanwhile, the driver reacts within seconds and starts stepping back up. It’s kind of fascinating to see. The step-up is intentionally slow and gradual, but it should keep going until shortly after the ice is removed.

Last night I tried some thermal tests on a Convoy S2+ triple with Crescendo. After melting some ice on it, the bezel was in a small puddle of water about 3mm deep. I didn’t really think anything of it, but the step-down stabilized almost twice as high as it had been before. Then I soaked up the water and let the test continue, and it stepped down to the original level again. Apparently a few millimeters of water at the front end of the bezel was enough to double the stable level for regulation.

That code isn’t released yet though… I have some bugs to fix first. It worked fine in the D4, but it’s behaving a little weird in Crescendo.

You’re probably right. It can’t be much fun taking those apart to reflash them. I still haven’t taken the driver out of my green one; have been waiting until the firmware was confirmed to be actually final with the updated version available on Hank’s site. I don’t want to open it more than once.

How about an open weave Kevlar glove with strands of copper woven through it? Carry the light in Turbo mode, the glove acts as an active passive heat sink with so much more surface to air structure that the light could run til the cell dies. Just a thought.

True. But I feel very uneasy when the head of the light is burning hot. Especially when the switch is mounted on the head so that heat can’t be avoided. My Manker E14II can run a decent period of time at max turbo due to its copper head, but that doesn’t help me when the head is dangerously hot to the touch.

Thermal transfer of heat from the LED is nice, but I don’t want it to be too nice.

A year or two ago I modded a DQG Tiny III 18650 with triple XPL HI and a FET driver. The head got burning hot to the touch in a matter of seconds. This felt completely unusable to me even for short burst use. I added an aluminum heatsink around the head. This helped… but not much.

I then painted the head to match the coloring of the anodizing on the body. And in an attempt to keep the paint on I laquered the heatsink with brush-on super glue. This helped a lot mainly due to the insultation caused by the layer of plastic from the glue. The metal switch still got too hot too fast so I built a new housing from it out of Sugru synthetic rubber. Now the switch could be touched even when the head is hot.

So what was the result of all the mods to that DQG?

  • It looks like crap - yeah yeah… superglue, paint and sugru is definitely not aesthetically pleasing.
  • The LEDs stay hotter longer. The insulation on the outside makes it harder for heat to get out of the light.
  • The head stays comfortable to the touch much longer. When it does heat up, both the head and body seem to heat up at the same rate. Instead of only being able to run the light at 100% FET output for 10 seconds on a fresh 30Q I can now run the light for about 1 minute, without burning my hand. To me that’s a big improvement.

In short, heat transfer is all well and good, but ergonomics and hand-safety shouldn’t be neglected. A full copper head in a hot-rod light might be fine … but only if you have some kind of insultating plastic or rubber cage to keep your hands from touching the hot parts. Maybe someone can make a safety cage on a 3D printer for the Manker lights. Maybe a cage could be jury-rigged from Sugru.

I’ve tried copper, brass, aluminum, steel and titanium flashlights. Personally I think aluminum is the most useful choice of materials, though for lights with pills I prefer a copper pill.

Awesome upgrades Toykeeper! I’ll definitely order one soon. Likely from hank due to shipping costs.

… or instead of sacrificing the flashlight to save the user, maybe it could use regulation, thinner wires, or moderation so it doesn’t get scorching-hot in the first place? This way, it wouldn’t hurt the user or the flashlight.

If it’s too hot to touch, it’s probably too hot period. Incandescent lights may operate at 5000 actual Kelvin, so insulation is appropriate for those… but LEDs die faster the hotter they get.

I’ve never had an LED in a light die from too much heat. I tend to go on to the next LED light toy before that happens. As long as my LEDs aren’t getting so hot that the solder melts I’m fine.

I’m also not a programmer. In the DQG I was using a driver with moppydrv firmware with a programmable stepdown. Only problem was the stepdown only went to 50% FET, and in that light even 50% was too much for extended use. I either need a rampdown to 20% or I need a temp sensor.

On the upside, it looks like RMM may start selling drivers with the new firmware on them in different dimensions. So hopefully I can replace the drivers on all my old e-switch lights with drivers with your new v2 firmware :slight_smile:

Oh to be a fly on the wall inside your titanium supported brain!

I have to admit I was just contemplating a line of thought very close this.

Just got my S4 2B, and it does get dangerously hot. I took the optics out just to see how it would be as a mule and the end started smoking vigorously within 5 seconds. This flashlight doesn’t seem safe?

The LEDs’ circuit board is not glued or screwed to the shelf. The optic and lens press it against the shelf when the bezel is tightened down.

Removing the optic prevents the board from properly conducting heat through the shelf and into the body of the flashlight.

I hope your light is okay… :frowning:

It’s fine now, the domes aren’t cloudy or anything. It was more surprising than anything else. I do love the light, I just wish LEDs were more efficient here on the ragged edge of emitters in hotrods. Love the UI.

ok this emisar light need a proper exposure, i didn’t even know D4 before this thread. Now there is a thrower and a D7? I haven’t even got my D4

Yeah, without pressure holding the MCPCB against its heat sink shelf, it’ll cook itself pretty quickly. That’s normal. It shouldn’t be operated at high current with the heat sink nerfed.

(this also applies to wrapping the pill to make the surface feel less hot, though that has less immediate results)

D4’s MCPCB is 2mm thick, all other MCPCBs are 1.6mm, it is highly unadvisable to use a thinner MCPCB which will create issues with thermal transfer between the head and the MCPCB as the Bezel is holding down the MCPCB by pressing down on the optic.
Neither should the flashlight be used without the bezel and optics in place, doing otherwise will overheat the LEDs.

The D4 will produce tremendous heat on turbo mode, and heat up very fast (10~20 seconds for the thermal protection to kick in depends on the cell), please make sure the bezel always be tightly enough on D4 although it’s checked very well before leaving the warehouse.

Hi Hank,
do you know when the D4 will be available at intl-outdoor ?

Hank, why did you make this MCPCB thicker?

Yes, it’s up now,
http://intl-outdoor.com/emisar-d4-high-power-led-flashlight-p-921.html

The price has been increased once again… :weary:
EDIT : ordered anyway, I’ve waited for such a compact e-switch flashlight too long… and with that FW I could not pass !

Ordered!!! Muahahahahaha!!! :smiling_imp: