Finally, I have a light that is starting to approach my holy grail for an outdoor flashlight. A light that can light up far down field and that also gives nice near-field illumination. But not so much near field that I can’t see well down field or that makes me squint and totally loose my remaining night vision. A light with decent cell capacity and heat sinking.
The great news is that anyone with beginning mod skills can replicate this light for about $90 and a fraction of the time and effort I put into this build. The recipe is:
· UF-T90 from Wallbuys ($70 after discount code)
· MT-G2 on Noctigon
· R10 Resistor
· Better springs (or just copper spacer in my case)
The light already has a great driver for the MT-G2 (HX-1175b) and also an Omten switch.
The beam pattern on this light is so like a good thrower, but on a bigger scale and with incredible spill. Most useful spill I have ever seen. I'm getting such a thrill seeing stuff that I don't see with all my other lights. I have serious throwers and flooders. Both leave me wanting some of the other. I have 3, 4, 5, 6, 7, and 9 emitters lights. They are the most useful, but they just put too much light in the near field. I want to see far and outside of the spill of light.
Right now I'm running this light at high current to see how it behaves. I will probably shoot for a 9 amp high mode with a time out turbo mode. Nine amps is the last sweet spot on the MT-G2 curve.
Summary of mod:
- Dedomed MT-G2 - No MCPCB in order to get emitter higher in reflector
- Stock Driver with 07N03 FET swapped in.
- A R10 resistors added.
- 9 amps measured to the emitter before light was assembled. Something isn't adding up here. I think I must have had a poor connection on the emitter current measurement.
- 171kcd throw with .035" shim 198kcd with .56" in shims and at 11.3 amps. Still working on focus. MT-G2 throw measurement thread here.
- On high, you can feel the heat coming out almost instantly. The light gets very hot within minutes with 2 R10 resistors at 11.3 amps.
Light has 2 lenses stock. Took one out and will not used emitter base. So had to build up integrated pill with 3 .07" disks and one .035" disk spanning the wider area. Mouse over is holes in integrated pill being stuffed with copper.
These are the leads to the emitter. They are 10 or 12 gauge copper wire wrapped in a layer of kapton tape. They are a slightly snug fit through the head so that they also conduct heat. Very hard to solder to after pressed into head.
Poor emitter has been through hell a few times. Had a couple accidents machining copper with the emitter in place. Also, grounded off the corners to clear reflector hole. Not symetrical at all. Luckily, MT-G2's are tough. My MT-G2 dedomes never look good as I just pick off as much silicone as I dare. What it looses in attractiveness, it makes up for in performance. Emitter was too high into reflector in the below pic. Had to add another .056" in copper disks (with hole for emitter) to lower emitter in reflector. Bottom of the reflector presses against the surrounding copper. So it helps move the heat too.
Copper round bar that was eventually pressed into built up head. Mouse over is bottom of bar pounded out and flattened for next copper disk. I kind of got sloppy trying different utensils to pound with. Hence, all the damage in the surrounding copper disk.
Very hollow head. Lined with copper sheet and insert copper and aluminum disks.
Bored center and pressed in 3/8" copper round bar. Below pic is second liner getting fabbed. First liner spans entire length of head. Thickness under emitter is almost 1.25". The liners seem to help transport the heat across the long head nicely. They are sized to very tightly press fit. Mouse over is how I determine size of copper liner. Use cardboard mock up.
Ran a 5 minute test on a HX-1175B driving MT-G2 with 3 KK ICR's. The N-Channel FET was replaced by a 70N02's and 2 R100 resistors added to the existing stock R062's. I changed to the lower resistance FET's because Relic had reported heat issues with his T90. The HX-1175b driver is from a UF-T90 (The long light with SST-90 from Wallbuys). I think it is also used in the Lustfire 3U2-X3. The cells were down a little (4.12 volts resting) from several short runs last night. The emitter was getting 10.7 amps (down from 11.3 last night). I don't know if the current was down to changing the set up or from lower cells.
The driver was loose and not heat sinked. At the end of 5 minutes, it was at 203 degrees F (95 C). The heat appears to be generated mostly by the FET's Source Pin resistor bank. The FET didn't feel hot. The MT-G2 lux remained unchanged through the test. Temperature of the head was 170 F (77 C) at the end of the test. The head is heavily heat sink modded. Current was down into the 9's at the end of the test, but I don't remember the first decimal.
I think this indicates the driver might be able to handle driving an MT-G2 around 11 amps (Maybe more?) with 3S cells, but I the driver should be well connected to the heat sink. I ordered a few of these drivers from 2 different ebay vendors. Hopefully, they send me the real deal and not the HX-1175c RaceR86 received. Here are what I ordered (Not yet received). The second buyer I have no experience with either. The specs in his ad are low, but the picture is of the HX-1175b.
Here is the driver potted with my potting recipe. Mouse over shows driver retaining ring I made from 10 or 12 gauge copper wire (pressed in tight). That spring had to be replaced by copper slug.
Tail spring fried instantly. Switch was going to next. So I turned light into a twisty. Mouse over shows Kapton tape that was needed for power to disconnect. This is temporary. Switch is back in. It is an Omten and seems to be fine. Just removed the spring and soldered the brass plunger directly to the switch. I used the tail cap to control pressure on cells.
The light illuminates a big area nicely. Good general purpose light for big and deep areas. Atmospheric conditions are bad tonight. It's been raining and the humidity is high. So much of the beam gets absorbed by the moisture in the air. Since I have not pictures posted in those same conditions I took a comparable using a DST with a domed XM-L2 T6 3C driven at 6.5 amps (Mouse over). The MT-G2 has the type of beam I want 99% of the time.
On a clearer night. Unfortunately, it looks like the beam hit the right side of the tree and sky.
Here are 2 driveway shots (one is mouse over). The below pictures were took on the first night with all the moisture in the air.
Another part of the drive way. I measured once. I think the utility pole is somewhere between 500 to 600 feet (200 yards) away. Mouse over is just another shot (not driveway). Don't know how far away the transformer is.
- Need to piggy back an different MCU on the driver.
- Add side momentary switch.
- Consider removing phosphor layer around (not over) the grid of crystals to get rid of yellow around corona. Not sure I want to do this as I think this improves tint and cri down field.
- Didn't need to do all the machining to get wires under the emitter. The T90 reflector is designed for emitter to be below bottom of reflector. Currently, the bottom of my dedomed MT-G2 is . 056" below the bottom of the reflector. I need to try .042".
- Can't think of anything else at the moment. This is my favorite light now. Happy, happy, happy.