Just my opinion they should be fine with just a little profit on the group buy. After its over they can charge whatever they want for the next 10 years. 20 years make changes and make new models etc. And they get a light thatâs better then there team of designers could come up with. The best minds are here on blf. Which they get designed free for them as well. How much would they pay a team of designers for a light like this? The little to breaking even on our run I see as payment for the months of work blf team of designers has put in. Any changes that can be made and still keep it $40 and a better light should be made. They can raise it to $70-80 afterwards for a 5k lumen light and it will sell. The s70 at 3k is $70 I wouldnât be surprised if they ask for $100 for a 5k lumen light. These builds should be little profit for us designing and see issues they canât see. Theyre getting probably thr best ligjt they have up until the point and get to keep the rights to it for free so i dont feel any sympthary if theres .01 cents per ligjt profit on our run of them. And they make their money from the general public on amazob and other forum. Tom and others would probay charge thousands in labor each with all thatâs put into this light. Giving us this light at cost is the way of payment and thanks for this light kind of thing
Yes speed4goal that is kind of what I am thinking too 
The rough calculations we have made show very little profit and he R&D, selling of first (larger) amount, good will and positive standing here on BLF is the major profit/benefit for Thorfire in this.
To be honest they didnât want to do it earlier on, saying it could not be done for this price. We stressed these not directly monetary benefits and they accepted it and continued the project with us. So we are all on the same page here
I think we need them more than they need us. We need to keep it attractive for them so that 1. This groupbuy happens at all, and 2. We get the quality product we specified.
A bird in the hand is worth two in the bush.
I like that there is a lot of copper area. As well as low resistance, that helps heat transfer and sometimes high frequency interference.
Mind if I give it a shot in diptrace? I am sure it would be easier then by hand.
What is the MCPCB size going to be? And LED spacing?
Sure, have a go, thanks 
The exact dimensions are not set in stone, it is a bit of a guess from other SRK dimensions and the latest drawings from Thorfire.
- The basic idea is the ledboard drawing that I made in June, repeated below. It has as wide as possible traces from the solder pads to the leds, which close to the solder pads will have to accomodate 20A current. The size and position of the screw holes in this drawing is not correct anymore btw.
- The ledboard diameter most likely is 45mm
- It has a central hole of 5mm to allow up to 16AWG wires to pass through,
- Two 3mm diameter screw holes opposite and both 6mm from the center. These should be repeated at a 45 (probably?) degrees rotated position to allow the second set of 4 led solder pads be positioned under the reflector openings.
- The solder pads for the led wires should be very close to the center hole because an as small as possible cavity for the led wires must be milled out of the underside of the reflector, this cavity must not cut into the four reflector cups
- Close to one of the 4 leds (between a set of XP and XM pads?) should solder pads for a 0603 (?) size thermistor and connected pads for the wires for it.
If this explanation in words is not clear enough I could make a sketch for you.
The external temp sensor (thermistor on Wikipedia) is the best, most reliable way to go, but of course requires yet another I/O pin, plus a lot of dev and testing. Calibration is the challenge - making it easy to do on a per unit basis is a challenge, unless there's a sensor out there that doesn't have a big variation from piece to piece, dunno.
An alternative is to ship the units with a default aggressive turbo-timeout. We will go beyond the safe zone of heat for what this host can take over time, plus the 4 cells will keep fairly high amps going for a while, though it's really equal to 4 individual single cell, single XPL lights, so should lose output/amps as it discharges. But of course going beyond the comfy zone is what this group buy is all about - as other BLF custom group buys have. Goal is to put a custom high output compact light into the hands of the many at a reasonable cost.
fyi... I'm a full-time consulting software developer, so my NY hourly rate is a little up there, but actually rates have dropped in the last 15 years with offshore and green card import pressures in the industry, and world wide access in general. Back in the 90's, 20 years ago, the rates were higher than they are now.
Ok, here is a very quick and dirty proof on concept for the mcpcb. I didnât feel like making it look pretty if changes were going to be needed lol.
I swapped the LED pads 90 dergees from your drawing because I felt this gave a larger trace size to the ground plane.
The LEDâs can be rotated back and a smaller trace to the ground plane used as well if people think that is better.
I only put a single pad for the LED + and - since that A: leaves room for the thermistor close to the wire hole and B: you can simple move the wires to the new location if you rotate the reflector, it would be super simple to repeat any cutouts at 45 degrees on the reflector.
I started working on a 12v / 4S version of the MCPCB but doubt people will really care since it would not effect most people and it would not be as clean as this setup. Although running the light in 4s mode would net noticeably higher amps/lumens but this was shot down early on.
In 4S mode you should be able to do the same as an A6 per LED, around ~5 - 5.5 amps and ~1500-1600 luemns per led. In parallel mode we seem to be getting around 4 - 4.25 amps per LED and around 1200-1300 lumens per led.
Not a major deal and not really an option since people said they want it setup in parallel. Having it work in both is doable but would not be as pretty / as uniform in the traces. Plus if someone really wants to get that crazy they could swap out for individual MCPCBâs I guess.
WAUW!
Thanks.
Neat how rotating the LEDs 90 degrees makes this elegant almost flower like patern!
Please do not be insulted by my poor sketching, terrible colors (easier like this in my crude paint thingy)
The screws should be in pairs at an angle of 45 degrees right?
And that would mean the thermistors cannot be placed where they are now.
Correct me if I am wrong, and well I probably am but want to put it out there in case there is not all wrongness here in this sketch:
Yep, you are correct, I honestly tossed the second set of screw holes in at the last min and forgot they need to be 45 degrees instead of 90. 90 looked so pretty I didnât think about the fact it wasnât correct. lol
I will get that fixed later.
Or should we swap the screw and wires? Have a single large screw in the middle with the wires coming out the holes on the side? Doesnât change much for the holes but you could go a lot deeper without hitting a reflector in the center. So then question then is a single large screw stronger then 2 smaller ones?
Two screws protect from accidental rotation during assembly with the bezel and lens.
That s why we went with 2 screws.
Nice work sofar between the 2 of you. The screw holes where they are probably clear the reflector cups already. We want the wires in the middle because the cutout (2mm deep) on the underside of the reflector must not be too close to the reflector cups, sending them to the side may cause a bigger problem than the screws.
Good points, I will correct the screw holes.
In the 4 led version I pointed the led wire solder pads, and also the screw holes in between two leds, so that the reflector cut-out and screw holes would have as much âmeatâ under it as possible, in this 2x4 led version that is still possible for the screw holes, but not anymore for the solder pads.
Edit: thinking of that, as the thick ledwires need the (rectangular) reflector cavity, the solder pads for the wires should rotate with it, in other words they have to be elongated in a circular way. The cavity in the reflector can not be widened to the side because that is where the screw holes are.
Please think with me how this should look like
The thermistor circuitry can easily be placed between two leds, so that the crowded middle of the board can be fiddled with more easily.
This is an option but was trying to keep the wires as short as possible but there may not be another option.
Here is an updated version with the correct holes and touched up here and there. I also extended the solder pads to allow them to line up with whatever cavity there is.
We could do 2 separate pads but it would be hard to fit 2 good sized pads in that space and keep them separate. I could do 2 overlapping circles though.
ha great! No need for me to hurt your eyes now wth this last revision!
Looks great!
Overlapping solderpads probably looks nice (almost like 8 holes there in the center) and from what I have been reading of the software you use are also probably easier to add then a free hand shape.
Yeah, 2 circles is a bit easier, I will swap them out real quick.
Naturally this will have to be redone by Thorfire so they can make sure everything lines up perfectly but this should hopefully be good enough to show them what we want.
Lets try this one on for size:
