Well the specs were always the same, we just didn’t give anything more then ~120mm ID and roughly square dimensions as we want to see what the manufacture has to say before we commit to anything.
Okay, but this latest depth is an increase of 20mm or so, compared to the drawings we’ve seen.
I did some research about the perfect parabola on my own, and it seem that we are just a little bit of in our measures..
I came in contact whit Bob from a math forum, and we discussed the parabolic equation, and this is the tread:
http://www.mathisfunforum.com/viewtopic.php?id=23575
When the diameter is 120mm the depth for a perfect parabolic diagram is 92,63mm (rounded upwards to 93mm for simplicity),
not 112mm taken from an upscaled TN42 design..
Maybe it can give us an edge? ;)
Morten.
Interesting info, I have no problems going with a different depth reflector, as we know the depth has little effect on the throw anyways.
Going with a 93MM deep reflector also makes the light a bit more compact and/or allows more room for cooling fins.
Something to consider for sure.
Almost 20mm shorter head
Less weight
Less material head
Less material reflector
Sounds more then interesting I think.
Uh, where does “perfect” come from?
Here are four sections of different parabolas, all with diameter 120mm:
There is nothing non-perfect about these parabolas, they just have a different coefficient of X^2
As you can see, you can have different heights of a parabola while limiting it to 120mm diameter.
The thing is, the amount of frontal reflector area you gain from increasing the depth has diminishing returns. You can see in the circle at the bottom that the area between the large and small circles only increases a bit when making the reflector much deeper.
(value f is the area between the large and small circle)
That would bring us back to the commonly used proportions, namely (roughly) 4 x 3 size, like a C8 reflector and many (most) others.
I would prefer it that way too.
@ Endeman:
I think ‘perfect’ is meant in the sense that the addition of depth becomes a bit of a waste of materials (and looks too) beyond a certain point.
I agree.
This does mean though that i’ve spent quite some time for nothing, trying to fit a 125 x 112 cm reflector in a good looking exterior… 
Makes you wonder why though, Thrunite decided to go for relatively deeper one.
Nice find! But unfortunately it’s nearly impossible to make that work. According to [this review](Nitecore "Tiny Monster" TM11 (4x18650, 3xXM-L) Review: RUNTIMES, BEAMSHOTS, VIDEO + | Candle Power Flashlight Forum , the battery tube diameter is 50 mm. The inner diameter of the battery tube of our light has to be “at least 46 mm”http://i.imgur.com/KQZazrQ.png so there isn’t really any space for threads.
However, the handle is 30 $ at bg. So if Nitecore can sell it for 30 I think we should be able to figure out something cheaper. I still think the best idea would be to offer an optional handle for those who want one (if it’s around half of the interested people).
The word Perfect was just my “non complicated brain” that suggested… 
I still fail to see the need for or advantage of a handle on the GT, which will have an approximately 50mm diameter handle of itself (the batt. tube).
I would say skip the handle for the GT altogether.
It’s the most efficient proportions.
I agree, the guy at the maths forum just showed a simple more or less proof that the light rays with origin at the focal point are reflected parallel to the y axis. His 9x something mm just come from our 112mm minus the height of the focal point.
Which is where the 112mm depth came from, that is the scaled up proportions of the TN42.
There is no such think as the most efficient proportions. Theoretically, the longer the reflector the more light it catches and consequently the more throw. However, as we are limited to a length about at most 10 - 12 cm we can just decide what we want. Again, the longer the reflector the more throw and the angle of the spill will be smaller.
We could calculate though how the extra length is proportional to the intensified spot. That actually sounds like fun, I’ll maybe do that tomorrow.
Edit: it’s actually pretty simple. It’s just the angle of the triangle between the focal point, the edge of the reflector and the orthogonal projection on the y axis.
And we have to have in mind that the LED this light will use have a beam angle of 115 degrees…
http://www.cree.com/LED-Components-and-Modules/Products/XLamp/Discrete-Directional/XLamp-XHP35-HI
Morten
Efficient in the sense that you have size (and weight) and performance meeting eachother on a ‘sweet spot’.
You’ll (probably) find it won’t add too much throw without adding a lot of hight / length / depth.
Also the base (focal plane) will decrease in diameter a little, which is also not what we want.
I got that part down on this graph, you can see there are two lines at y=–1 that can go from –90 to +90, this indicates the degrees of light captured.
Depending on the width and height, different amount of degrees are captured. (value p is degrees from horizontal)
All you need to do is overlay this graph on top and you can see how much is being collected:
Although this is just for some random LED, you need to get the correct graph for the XHP35 from the cree data sheet.
Also I need to work on integrating this function over a circle, because the –90 to +90 is just a slice that needs to be rotated around the LED.
So just keep in mind that the graph is only two dimensional and won’t give you the real numbers, just a rough idea.
This is the graph for XHP35 HI
Dang it miller… you’re killin me! Got me for 3 Q8s already. Now i gotta go in for 2 of these also. Keep it up and my wife will divorce me for sure!
P.S. that means put me down for 2 please…
Thanks!