New WildTrail (former LuckSun) BLF-D80v2 Sale is open.

Tom_Tom · March 11, 2019, 6:35pm

everydaysurvivalgear:

Tom Tom:

To be more positive, the D80 is a pleasing looking handy thing with unique characteristics, and could be brought up to date with a decent FET+n driver, e.g. a BLF/Lexar/Texas Ace, running whatever modern firmware that can still operate with a tail clickie, rather than an e-switch. 7135, or dual bank 7135 for efficiency and steady output at modest levels.

Preferably a straightforward simple intuitive UI that doesn’t require a crib sheet to comprehend. Perhaps a couple of programming options, but no more.

The original driver was very poor, it even had an incorrect value resistor that prevented the very weak FET from turning on fully, restricting tailcap current to 1A or less. DEL discovered this, and showed us how to improve it. However it still suffered from objectionable PWM, inefficiency at the lower levels and poorly spaced modes. It must be binned.

That short reflector and head is rather unique, and the concept should be retained.

Keep the tail clickie, don’t be seduced by side e-switches.

Together with a DTP copper MCPCB and a modern smaller LED (original was XML-2) with higher surface brightness for better throw. As supplied, with non-DTP Al MCPCB it was thermally challenged as soon as you tried to pep it up. Though the basics were there, integral shelf and twin screws to clamp down the MCPCB solidly.

However IIRC some who tried smaller LEDs at the time found that it was difficult to get the reflector/LED focussed satisfactorily and without artefacts in the beam, it is very sensitive to e.g. the thickness of the locating piece, and the large reflector hole, sized for the XML-2 doesn’t seem to work so well with smaller LEDs.

Perhaps the reflector could be re-tooled, just by reducing the size of the hole to suit smaller LEDs. Or it might require re profiling of the flatter part.

The beauty of the D80, to my mind, is the neat compact design, and it feels good in the hand. The workmanship and surface finish was great.

It makes a C8 feel crude and unwieldy.

However for an enthusiast it was no good, basically just a host to spend the same again on as a modding project. Not a well rounded BLF design that could be used as-is.

So, my list of essential changes would include:

1) New driver, in a standard size (e.g. 17mm)

2) DT|P copper MCPCB

3) Updated LED with reflector modified and tuned to it.

4) Body tube length optimised to take flat top, button top and even protected 18650 cells. Sadly this may require changing the stud contact on the driver (which I like very much, much better conductivity than any springs) for a spring.

5) Spring(s) to be good BeCu, e.g. Blue type. The days of faffing about with spring bypasses should be over, at least for BLF designs.

6) Short 18350 tube also to be available, with liner to also take 16340, CR123 etc. Some have cut down the 18650 one to make a very neat and handy package, it would be great to have this option.

7) Forget the latest fad for 21-70 cells, that would bulk it up and lose the neat compact nature of this, and further complicate use with 18650, spring lengths etc. And I’m pretty sure they will continue to be a fad, as well as the dimensions not being well standardised, making tolerancing of the tube and selection of compatible cells a guessing game. Having a 1” OD tube is important for some applications, it is the standard for many mounts. Yes, it might be possible to fit a 21-70 into a 1” tube, but I think you would lose the chunky grippy nubs and stout wall thickness that make the D80 so tactile.

8) Change the thread of the switch retaining ring to LH, so it doesn’t keep coming loose when you change cells. Convoy knew this years ago.

9) Optional FET remote tailswitch assembly for hunters, curly lead and pressure switch, with UI option to lock into turbo-only output. Maybe in a package with some flip on colour filters.

Give me 1-8, at a BLF price and I would buy a couple. But overall that would be a lot of re-tooling to get to the product that it should have been originally.

The actual BLF D80 was around before you where a member so i don’t know what model you got? It had a worked driver doing 4.5amps and a DTP MCPCB its a proper Noctigon board with specialized XML2 tints. The driver is fairly standard its a 20mm driver but you can make a BLF driver fit in with no mods. Maybe you bought the normal D80? Does yours have a Noctigon board?

I have been a member here far longer than the existence of the BLF D80, , but ducked out for a little while and re-joined under this new name. Fresh start.

I bought the BLF D80 as soon as it came out, one of the first delivered, and my description of mine is correct. No it did not have a Noctigon. No it did not have “specialised tints”, just a choice of tint with no binning specified.

It should have had a Noctigon, but mine did not, I guess it was a mongrel using a standard D80 part instead of the BLF component. I was unlucky.

The driver, as for many, did not perform properly, due to an incorrect resistor fitted. Even when corrected, the two SOT 23 FETs are weak. Works ok with a full cell at 4.2V, but as the voltage drops the on-resistance becomes the limiting factor.

You could refresh your memory at the original thread. Things started unravelling about 2000 posts in.

I’ve remembered that mine did not have MCPCB securing screws either, that was a mod I did, because I was doubtful that pressure from the reflector was adequate, having suffered thermal issues with my Al mcpcb.

It would be good if they could be included on a new design.

Sorry for any confusion.

DEL discovered that the FET gate drive resistor was far too high a value, meaning that, in combination with the gate pulldown resistor, the FET gate was only being driven with half cell voltage, not enough to turn it on fully particularly at lower cell voltages, i.e real life. Testing torches only with fully charged cells is unrealistic. Though it seems to be the norm, ie. topped off strong cell in perfect condition, straight to turbo, peak lumen/candela/tail current readings for a few seconds, job done. As for real-world performance …

Perhaps that was corrected in later builds. Otherwise fixable with a solder splash, or a much lower value resistor.

But the driver was crude and the two paralleled FETs poor. Poorly spaced modes, objectionable visible PWM in lower modes, no low voltage protection.

First thing to replace. I think we are agreed on that.

I could not get a 17mm FET+1 to fit, except with an adapter PCB, the cavity on mine was more like 21mm. I think that reduced to 20mm later in the run.

I suppose the basic driver being so under-developed and cheaply made helped in hitting the selling price. The firmware tailoring for BLF was half-hearted, PWM could have been made higher frequency, modes could have been better spaced, and somehow the original LVP disappeared. This should have been picked up during prototype evaluation.

All this is fixable for V2.

I am not trying to knock the original, it was what it was, and is history. It could be brought up to date, this time with much more attention to detail, and more rigorous evaluation of prototypes.

When looking to refresh this, it is wise to also understand some of the history of the original, it was far from perfect.