Reflector is 51mm, lens is 52mm.
Does anyone know if the stars on the DRY driver are functional?
Ouchy, was that 13 amps at the emitter?
I agree with O-L. Same experience here in that every mounted MT-G2 I have received had a gap. Luckily, I have had no failures, but then, I have not driven one past 6 or 7 amps yet. Well, also I reflowed them all really early on.
Gj, here is a Dry thread.
Thanks!
It was at the tail.
Yikes!!! Now you guys make me want to reflow my MTG2s just to be sure they’re done right.
Don't worry too much. I have run two of mine at 10A+ for over two minutes, no damage. I have built 10+ lights for others with MT-G2, no damage. DBCStm has been running his hard at 10A+, no damage. All of these are factory reflowed and no there isn't solder all the way to the edge of the pad on any of them.
IF there is ever damage due to insufficient solder under the emitter, I will warranty it as a manufacturer defect. All I'm saying is that this isn't what killed this emitter so don't get too worried about it. You could have an MT-G2 with a perfect reflow on a block of dry ice and it would probably still fry within a few seconds running 3S cells without a properly working buck driver.
Head diameter = 2.2”
Tube diameter = 1.3”
Anti-Roll Handle Piece = 1.9”
How would a holster work with the anti-roll piece on the handle by the tailcap.
Thank you!
This is a neat build! I’m going to have to build one…
I have 2 questions if you dont mind…
What (approx) resistor would I add to up the current to between 4 and 5 amps with 2 cells?
Would that resistor mod work for 3 cells as well, or would the driver go over 4-5 amps with 3 cells?
Thanks!
Thanks
No resistors (bridge resistors) w/ two cells gives you around 5A. As of right now don’t attempt 3 cells with resistors bridged until more info is figured out on if the driver is having issues with the MT-G2 and 3 cells.
It is a buck driver, but when it fried this emitter it was modified with the sense resistors bridged. Without modification it bucks just fine, as tested by me and others. Many buck drivers have fried LEDs with bridged sense resistors, which apparently can cause them not to buck and voltage spike the LED.
I am not making this conclusion based on insufficient data, I have been talking with mhemling step-by-step throughout this whole process including before he ever did anything. I have also fried several LEDs testing buck drivers (including the new DRY driver) trying to push them too far, including brand new LEDs that I have reflowed myself and that were well heatsinked. It isn't a gradual overheating over a period of minutes, they pop like a fuse, probably the bond wires or internal pathways burning out. When an LED is thermally overheated it isn't going to happen instantly like this.
My point is that no amount of thermal management will help a part withstand a higher voltage than it was designed for. For all I know when we're messing with these buck drivers we could even get a momentary voltage spike higher than with 3S cells, but I don't have the equipment or data to back that up.
First off I would like to say the number I threw out there was just a guess off the top off my head, with no real data to back that number up. I have pushed a MT-G2 with 2 20r”s and got over 10 amps. In this thread DBCstm states he seen 12a at the moment of hook up direct drive with 2 20r’s. Now where dealing with 3 li-ion cells, is 15a possible? If you take a buck driver and bypass the sense resistor (solder across it) I would think there is no more regulation, because with no sense resistor it doesn’t know where to stop or regulate the current. Running 2 cells basically becomes direct drive, where the driver keeps trying to push current until there is not enough supply voltage to support the demand of the vf of the led. So if you use 3 cells its trying to buck the voltage and doesn’t know where to stop with the current, so it push’s it as hard as it can until the cells voltage stops it. Now what interest me is we can all look at HKJ’s useful graphs and know about what the predictable voltage of a cell will be at a given current load. But look at the very first second of those graphs, the cell’s have not had sufficient enough time to adjust to the load, so the voltage is much higher, almost at the 4.2v charged voltage. So if 3 good li-ions (not exactly sure which cells where used to blow this MT-G2 in this thread) where fully charged and hooked up direct drive (which I think this driver would basically be with the sense resistor bridged) At the moment of hook up what would be the current to the led?
.
In my testing with 20r’s direct drive, the current actual increased as the cell’s became warm, which is common with li-ons. I never actual tried fully charged cells and checked the current at the very second of connection, but I do believe as DBCstm says, that 12a is possible with only 2 cells so what could 3 good cells do, especially if they where pre-warmed.
.
@ImA4Wheelr, we know power in has to equal power out plus losses, so I’m with you on the math. I don’t know for sure but, at the moment of connection it looks like to me the cells voltage would be much higher and the driver losses maybe far less at that moment, since nothing has had a chance to warm up yet. This is all just me thinking out load, ya I know quite dangerous. :bigsmile: If anyone has something to ad or disagrees with my thoughts above, please post away. I don’t mind at all. I’m always willing to listen and learn.
.
I have 2 more 20r’s coming from RMM in a day or two, so I can sacrifice a MT-G2 to find out what 3 will do direct drive if need be. I just hope I don’t blow my meter up. 
Appreciate the offer to take one for the team, but it doesn't seem worth killing an MT-G2. Seems better to build a dummy load and test. Would have to be one tough dummy load though.
Your off the top of the head estimated current sounded right to me too. It wasn't until later that it hit me that it didn't ad up. Not sure if Ouchy is pulling my leg, but that tail measurement of 13 amp for 3S and Dry would scare the heck out of me if it happened to me.
As far as shorting the voltage sense resistor bank causing no more regulation, I think that depends on the circuit. I and others have done it to many 3S drivers with no harm even to xml's. Ouchy did it with the new Dry and didn't kill his MT-G2. I think it depends on were the bucking happens, but I have insufficient electronics knowledge to know for sure. The voltage sensor bank is usually for current regulator and shorting usually just sets to max current. The regulator attempts to deliver enough current to achieve a set voltage across the bank. When it can't, it just delivers all it can (which is generally limited by design).
I know that I killed two new XM-L LEDs reflowed to a Noctigon with the DRY trying to bypass the sense resistors with 2S cells, one died with an instant flash the other died within about 5 seconds. So it will kill LEDs if the sense resistors are bypassed when it should be bucking, that much we know. I am wondering if we should be trying to bypass some of the other resistors instead.
I agree. If you will remember back in this thread of yours
I blew 2 xml’s when I cranked the pot a little over 7 amps. I never figured out what was happening, I just know something had meet its limit. I think you had a few do the same.
I remember. I also remember shorting other drivers with 3S without issue. I never said this new Dry was not the cause of the OP's MT-G2 failure. I also never said it was. I've said I would like to know because it seems like a promising driver. I don't think we will ever know in this case due to a shortage of information. We will have to keep on experimenting/measuring with the driver to find out if it can safely be used for high current MT-G2 applications.
I wonder how the 10 ohm pot mod would do with this driver. It might also tell at what current the driver runs away, if that’s what is happening here. Anybody tried it yet.
I agree with what your saying ImA4Wheelr, I’m like you my electronics background doesn’t come with a degree.
Its my thoughts that a driver that has been designed to run high current could very well be pushing close to direct drive current if the sense resistor is bridged. With a lower current designed driver I can see where the limiting factory maybe the components or the design itself. I don’t own one of these drivers yet to test.
That's what I started out using (10 ohm trimpot). You can max it out over the sense resistors without doing any damage. It ends up giving around 4.5A to the emitter, if I remember correctly. It was when I started adding additional 220 ohm resistors on top of that that I started having problems, so I think that it is treading right on the edge when maxed out.
I think that I need to sit down and get a better understanding of this circuit because I think that we must be approaching this the wrong way.
What host it this? That is a huge pill. When I ordered some parts this week from Mountain Electronics, they didn’t have a host like this.
It's discontinued. 
>8 month old thread.
I guess it wasn’t new for too long.