Could anyone be kind enough to give a one sentence summary of where this is at?
Pages and pages of snips and snipes, giant collections of quotes nested 4 deep,…

Moving forward at it’s own pace.

bansuri wrote:

Could anyone be kind enough to give a one sentence summary of where this is at?

“Current status (2018-11-28): Trying to get production started.”

jon_slider wrote:

bansuri wrote:

Could anyone be kind enough to give a one sentence summary of where this is at?

“Current status (2018-11-28): Trying to get production started.”

Tom Tom wrote:

Time is either ticking away (negative thinking) or moving forwards (positivity). Meanwhile this thing seems to be in limbo. And borderline chaos. Neal, since you seem to be "the man" here, please get a grip and try to sort things out, and communicate. Don't bottle it. Again. If you can't do such simple stuff (procurement etc.), send TK a proto3, nor set up the distribution chain in advance, well. I'd better say no more. Not expecting a reply :(

Great post. Thanks.

ActiveAl wrote:

Tom Tom wrote:

Time is either ticking away (negative thinking) or moving forwards (positivity). Meanwhile this thing seems to be in limbo. And borderline chaos. Neal, since you seem to be “the man” here, please get a grip and try to sort things out, and communicate. Don’t bottle it. Again. If you can’t do such simple stuff (procurement etc.), send TK a proto3, nor set up the distribution chain in advance, well. I’d better say no more. Not expecting a reply

Great post. Thanks.

If TK is cool with it, I’d vote for her to give it a once over before production starts.

Hoping this gets some more traction and we get a final prototype/render.

Current status: Trying to get answers. Attempting to pull something solid out of the cloud of possibilities, so there can be a more concrete plan and schedule. Most of the remaining tasks depend on Neal nailing down details and making commitments.

Aside from trying to get answers, about the only task for me right now is to build new firmware with the latest updates, test it on proto2 for a few days, and send it for production to use.

Interested in one.
Please put me on the list for one (LH351D)

Does anybody know what kind of driver the light will use. It says regulated up to 1000 lumens. Is it going to be a Linear 7135 based driver?

Driver:

Anduril based 3 channel driver, with 7135+7×7135+FET.

nick779 wrote:

If TK is cool with it, I’d vote for her to give it a once over before production starts.

Hoping this gets some more traction and we get a final prototype/render.

We want the production version sent to someone close to Hong Kong as a final check.

Thanks BlueSwordM

BlueSwordM wrote:

No problem there too. I’m just currently writing the email to Digikey’s quoting department, so it’s gonna take a bit of time.

you are being facetious, right?

Pavlo, it’s a FET+7+1 driver.

So, three power channels:

• 1xAMC7135 chip for 0 to 350mA
• 7xAMC7135 chips for 350mA to 2.8 A
• FET for 2.8 A to full power (direct drive)

This keeps most of the levels pretty stable and reasonably efficient, as far as linear drivers go. But it still has a direct drive circuit for turbo.

In a totally non-scientific qualitative manner, the overall efficiency curve compared to a constant current linear driver looks something like this:

LEDs also change tint according to the power level. This driver attempts to minimize that effect by blending power channels, but it may still be noticeable. The curve for that looks roughly like this:

If I remember to, I suppose I should probably put this info in the first post somewhere.

Color temp / power ??
Should be efficiency / power, right?

@chinookee, no, not at all.

I really sent Digi-Key a detailed quote for 800-4500pcs+ of LH351Ds in the chromacity bins we want.

Jerommel wrote:

Color temp / power ??
Should be efficiency / power, right?

The color can be effected due to the power delivery. The FET is full power, many amps, and uses PWM which pulses the full power to create the lower brightness levels.

The 7135 limits current, it only allows 0.35A through. So when it’s using 1 or all the 7135 chips fully, there is no pulsing or PWM and the color is slightly different.

It’s a bit complicated for me to follow, but the tint does shift slightly based on the type of power delivery to the led.

Check ToyKeepers vid.
https://youtu.be/l94py6C_ld8

Toykeeper wrote:

Using a prototype of the BLF-A6, I configured the firmware to alternate every half-second between 100% power on a 350mA channel and roughly 6% power (via PWM) on a direct-drive FET. The result was a constant amount of lumens but a changing beam tint.

Of course it’s gonna take a bit of time.

The folks at Digi-Key are not going to be responding during the week end.

irony is hard to master, especially for me, I guess.

chinooker wrote:

OK BlueS, I was referring to your “it’s gonna take a bit of time.”

The price will be quoted when it is quoted.

Joking aside, my experience is large suppliers like this can usually provide quotes on items they deal with regularly in hours to a day or two during their regular hours, and if they have to work directly with a manufacturer on a special order of something that requires a little bit of special effort, it’s usually not more than a week.

ToyKeeper wrote:

Pavlo, it’s a FET+7+1 driver.

So, three power channels:

• 1xAMC7135 chip for 0 to 350mA
• 7xAMC7135 chips for 350mA to 2.8 A
• FET for 2.8 A to full power (direct drive)

This keeps most of the levels pretty stable and reasonably efficient, as far as linear drivers go. But it still has a direct drive circuit for turbo.

Thanks ToyKeeper for the detailed response.

Being that its a Linear + Fet Driver, it should be more efficient with a high Vf LED correct?
The LH351D has a higher Vf than your typical Nichia so it shouldn’t heat up as much right at an equivalent output right?

Last but not least, does this light have a mechanical switch or an electronic switch?

It is e-switch only. A mechanical switch would not allow for ramping (not easily anyway) and would take up much more space.

.

Pavlo wrote:

ToyKeeper wrote:

Pavlo, it’s a FET+7+1 driver.

So, three power channels:

• 1xAMC7135 chip for 0 to 350mA
• 7xAMC7135 chips for 350mA to 2.8 A
• FET for 2.8 A to full power (direct drive)

This keeps most of the levels pretty stable and reasonably efficient, as far as linear drivers go. But it still has a direct drive circuit for turbo.

Thanks ToyKeeper for the detailed response.

Being that its a Linear + Fet Driver, it should be more efficient with a high Vf LED correct?
The LH351D has a higher Vf than your typical Nichia so it shouldn’t heat up as much right at an equivalent output right?

Last but not least, does this light have a mechanical switch or an electronic switch?

If you define efficiency by led-watts/input-watts then a higher Vf would be better. However I would argue that the most important efficiency is lumens/watt-input and in that scenario you want the LED with the most lumens/amp where the VF doesn’t matter as long as it’s low enough for the circuit to work.

I’m looking for a dark grey light, so please add me to the list for one.

Definetily interested for one, please.

Thanks all for the answers to my questions.
Love the fact that its a 3 channel light for better efficiency and regulation.

Looking forward to see the final version come to life.

Well, instead of waiting around or moaning about the FW3A's production  I had to get creative and make my own version ;)CRX SR4-T

Copper, Brass & Titanium construction
Quad Nichia 219C 5000K emitters
20mm Copper DTP MCPCB
20mm x 7mm solid copper heatsink
22mm x 2mm dual coated ARC glass
Emisar D4v2 ramping driver
Copper & glass tritium momentary tail switch
Lockout-able
Magnetic cell contacts
56, 2mm x 2mm magnets in tail
18650 cell compatible

Length – 97mm
Width – 25mm
Weight – 185g (with 18650 cell)

I could have shaved another few millimetres off the length of the light but wanted a decently thick copper heatsink, magnetic tail and tritium vial in the momemtary switch button, still managed to keep it below 100mm.

It is not 100% finished yet but near enough.

Size comparison: