FW3A, a TLF/BLF EDC flashlight - SST-20 available, coupon codes public

13575 posts / 0 new
Last post
Tom Tom
Offline
Last seen: 8 months 2 weeks ago
Joined: 09/10/2017 - 08:30
Posts: 1163

TK, I think you need to map it. Even when running off just the 7135s this is a hot torch, but not something to be feared by the experienced who understand what turning on the FET can do.

Thermal feedback isn’t going to work on this one, I think you know that now.

So, map it, build a profile that works safely, subtly, and keeps the thing within limits (whatever they might be).

Or just go for it and to hell with the consequences. Lumintop might not like that though …

ToyKeeper
ToyKeeper's picture
Offline
Last seen: 54 min 47 sec ago
Joined: 01/12/2013 - 14:40
Posts: 10105
Location: (469219) 2016 HO3
Meyer wrote:
What about an additional “expert mode”?

Expert mode is when you set the thermal ceiling to the maximum it’ll allow, and intentionally skew the sensor calibration so it thinks it’s cold all the time. That’ll effectively turn off thermal regulation entirely.

But, like… don’t do that. Smile

adnj
Offline
Last seen: 1 month 1 week ago
Joined: 08/12/2012 - 22:04
Posts: 477

ToyKeeper wrote:
A question about thermal regulation:

At the moment,

1100 lm whenever it senses any overheating

So.

What do people prefer?

  • Smooth but slow universal method which allows full turbo to run longer but may require the user to exercise common sense and turn the brightness down manually sometimes.

Turbo ramp-down always seems better for me in real-world applications.

bushmaster
bushmaster's picture
Offline
Last seen: 51 min 25 sec ago
Joined: 07/09/2017 - 15:40
Posts: 358
Location: Colville, WA USA

PBWilson wrote:
I can see points for both options, but I do know that I use my D4 less often than I intended to when I bought it because of the potential for burned fingers and lighting pants on fire (even if I’m not lying!). I carry a variety of flashlights, use them most days, know how to be safe and understand the dangers and I still can see the value in the second option.

Turbo in a light this size is not a feature that I’d use often. I’d show off to friends and family with it or use it if I need a lot of light quickly. Neither of those situations require minutes (or even a single minute) of turbo so I’m not so worried about a stepdown at the first sign of overheating.

I can guess that there are a bunch of folks who hate the thought of throttling down a light with this much potential, but I think that the second option is wisest.

I have not been a participant in the creation of this light. I sure plan on buying one though. For whatever it’s worth, I’m with Wilson. Go with the safer option.

Keep your nose in the wind and your eyes along the skyline.
Del Gue

teacher
teacher's picture
Offline
Last seen: 4 months 6 days ago
Joined: 02/23/2016 - 19:04
Posts: 9596
Location: NE & SW Alabama

@ TK…… I say do whatever you think is best in the big scheme of things. Thumbs Up

You never know how a horse will pull until you hook him up to a heavy load./"Bear" Bryant 

 .................................. "Slow is Smooth, Smooth is Fast" ...................................

       Texas Lumens Flashlights / M4D M4X Deals : sign up - save $$$$  

         Rudeness Level _ mΩ _ {width:70%} _ LightWiki _ LED Tint Chart  

      Xlamp size chart _ BatteryU _ Flashaholic? Need Professional Help???            TheOriginal _ TAB _ LightSearch _ BatterySearch _ 14500's _ DiCal 

 

                                             

ToyKeeper
ToyKeeper's picture
Offline
Last seen: 54 min 47 sec ago
Joined: 01/12/2013 - 14:40
Posts: 10105
Location: (469219) 2016 HO3
PBWilson wrote:
I use my D4 less often than I intended to when I bought it because of the potential for burned fingers and lighting pants on fire

Exactly.

The FW3A implements a few things to reduce the heat issues compared to a D4…

  • Three emitters instead of four.
  • Regulated output up to ~1100 lm instead of ~150 lm.
  • Button on tail instead of side, so it’s harder to hit by accident.
  • Much faster access to lockout, and less need to exit lockout for quick tasks.
  • User-adjustable ramp ends, so the ceiling can be lowered.
  • Default ceiling is ~1100 lm instead of full turbo (though turbo is still accessible).
  • Smoother regulation which is less prone to overshooting and oscillating.
  • Temperature sensor calibration, for more consistent and predictable behavior.
  • Muggle mode, for use while lending the light to others.

… but it has less thermal mass and is still rather overpowered on turbo. The highest mode is like 500% of its sustainable power level, and the potential change would make it drop down faster to ~150% or ~200% where the regular algorithm works better.

Scallywag
Scallywag's picture
Online
Last seen: 12 min 51 sec ago
Joined: 01/11/2018 - 22:23
Posts: 775
Location: Ohio, United States

I think you should err towards B as well. Whether it's the full-on "drop straight to regulated" or altering the ramp-down to be a lot more aggressive, it seems clear the "invisible" stepdown just doesn't work. It wouldn't on the D4 either. About how long will it run on turbo before stepdown?

EDC Rotation: ZL SC62 | Jaxman E2L XP-G2 5A | Purple S2+ XPL-HI U6-3A | D4 w/ Luxeon V | RRT-01 
EagTac D25C Ti | DQG Slim AA Ti | UF-T1 by CRX | Olight S1 | Klarus Mini One Ti
L6 XHP70.2 P2 4000K FET+7135 | Jaxman M8 | MF02 | Jaxman Z1 CULNM1.TG | Blue S2+ w/ ML Special
In-progress: Supfire M6 3xXHP50.2 
Others: Nitecore EC23 | Nebo Twyst | Streamlight ProTac 1AA | TerraLux LightStar 100

Eraursls1984
Offline
Last seen: 3 weeks 6 days ago
Joined: 11/08/2014 - 01:44
Posts: 211
Location: North Florida

I definitely think the quicker step down should be standard. If there is enough memory, maybe an option can be added to turn it off in programming.

bento_blf
Offline
Last seen: 1 year 1 week ago
Joined: 03/20/2018 - 09:18
Posts: 2
Location: semarang

If not too late, please put me in the list for one (1).
Thanks

cabfrank
Offline
Last seen: 2 hours 5 min ago
Joined: 11/19/2010 - 17:25
Posts: 2962
Location: healdsburg, california usa earth

Option A.

klrman
klrman's picture
Offline
Last seen: 5 hours 36 min ago
Joined: 11/07/2016 - 22:44
Posts: 2350
Location: Canada

I'm in for one thanks.

ARMS.AARON
Offline
Last seen: 2 weeks 2 days ago
Joined: 03/24/2017 - 07:46
Posts: 10

Please add me to the list for one!

Respectfully

pepinfaxera
Offline
Last seen: 1 day 6 hours ago
Joined: 06/25/2015 - 12:53
Posts: 269
Location: Gijón - Spain - Europe

ARMS.AARON wrote:
Please add me to the list for one!

Respectfully

You are number No. #728 on: Interest List , “Post # 4”:“http://budgetlightforum.com/comment/1120496#comment-1120496 , “ Page 1 ,”:http://budgetlightforum.com/comment/1120496#comment-1120496

See:

ARMS.AARON wrote:
My Q8 is sweet apple pie!

Please put me on the list for 1 of these. … Cheers , See # Post : Fri, 10/20/2017 – 13:19


For a total of 1 lights .
How many Flashlight do you want in total? … x … units? .

Sorry for my bad English
(google translator) , (https://www.deepl.com/translator)

Noir
Noir's picture
Offline
Last seen: 1 hour 30 min ago
Joined: 06/03/2018 - 09:13
Posts: 87
Location: Germany

I would prefer option B (with a fast but smooth step-down over the course of a few seconds).

I don't like lights that can become too hot to hold (and I don't want to have to worry about the light). Also if the light becomes too hot (with option A) it could potentially cause the thermal regulation to overshoot (which is of course undesirable).

 

BTW: have options to lower the sensor lag been discussed? For example these silicone rubber things could improve things a bit.

JasonWW
JasonWW's picture
Offline
Last seen: 5 hours 31 min ago
Joined: 10/22/2016 - 11:41
Posts: 11885
Location: Houston Texas

pepinfaxera wrote:
ARMS.AARON wrote:
Please add me to the list for one!

Respectfully

You are number No. #728 on: Interest List , “Post # 4”:“http://budgetlightforum.com/comment/1120496#comment-1120496 , “ Page 1 ,”:http://budgetlightforum.com/comment/1120496#comment-1120496

See:

ARMS.AARON wrote:
My Q8 is sweet apple pie!

Please put me on the list for 1 of these. … Cheers , See # Post : Fri, 10/20/2017 – 13:19


For a total of 1 lights .
How many Flashlight do you want in total? … x … units? .

Pepinfaxera, sometimes people forget they already joined the list. If they ask for one light multiple times, they are still asking for just one light. Wink

Texas Ace Lumen Tube calibrated with Maukka lights

Click this to go to signature links.

pepinfaxera
Offline
Last seen: 1 day 6 hours ago
Joined: 06/25/2015 - 12:53
Posts: 269
Location: Gijón - Spain - Europe

JasonWW wrote:
. Sometimes people forget they already joined the list. If they ask for one light multiple times, they are still asking for just one light. Wink
Thumbs Up
But, you must ask them

Sorry for my bad English
(google translator) , (https://www.deepl.com/translator)

pepinfaxera
Offline
Last seen: 1 day 6 hours ago
Joined: 06/25/2015 - 12:53
Posts: 269
Location: Gijón - Spain - Europe

ToyKeeper wrote:
What do people prefer?
  • Smooth but slow universal method which allows full turbo to run longer but may require the user to exercise common sense and turn the brightness down manually sometimes.
    .
    * Fast drop to a hardcoded level which is safer but may potentially be a little oversensitive and uses code instead of common sense to enforce “turbo is for brief use only”. .

Coffee without coffee?
Powerful flashlight without power?
.

That’s a lot of 3000 lumen? Make it 2500 lumen.
.
5 or 10 second turbo? I don’t accept it.
If that’s what it is, I don’t buy.

.
I have two, D4 Emisar, with Nichia 219CT 83CRI, 5000K and XP-L hi,
Nicha gets very hot.
XP-L hi also heats up, but less so.
Step Down works
and the leds don’t burn out.
ToyKeeper wrote:
The FW3A implements a few things to reduce the heat issues compared to a D4…
  • Three emitters instead of four.
  • Regulated output up to ~1100 lm instead of ~150 lm.
  • Button on tail instead of side, so it’s harder to hit by accident.
  • Much faster access to lockout, and less need to exit lockout for quick tasks.
  • User-adjustable ramp ends, so the ceiling can be lowered.
  • Default ceiling is ~1100 lm instead of full turbo (though turbo is still accessible).
  • Smoother regulation which is less prone to overshooting and oscillating.
  • Temperature sensor calibration, for more consistent and predictable behavior.
  • Muggle mode, for use while lending the light to others.
    ToyKeeper wrote:
    … but it has less thermal mass and is still rather overpowered on turbo..

Sorry for my bad English
(google translator) , (https://www.deepl.com/translator)

ARMS.AARON
Offline
Last seen: 2 weeks 2 days ago
Joined: 03/24/2017 - 07:46
Posts: 10

My bad.!

Just one light please!

Agro
Agro's picture
Offline
Last seen: 3 hours 50 min ago
Joined: 05/14/2017 - 11:16
Posts: 5021
Location: Ślōnsk

ToyKeeper wrote:
A question about thermal regulation:

At the moment, Anduril uses a thermal regulation method which knows nothing about the actual host characteristics or driver. It simply senses “overheating” or “underheating” like a user-adjustable thermostat, with some information about magnitude, like the degrees of excess multiplied by how long it has been in that condition. From this, it attempts to find a happy medium, with the adjustment speed determined by the magnitude. It’s fairly universal and typically works well with no need to hardcode anything about the host itself. Also, the adjustments are so smooth they’re basically invisible without a lux meter.

But.

On such a small and overpowered light, like the FW3A, overheating can happen so fast that it barely even has time to react. By the time it senses that things are trending too high, the light could already by too hot to touch. It works fine on a larger host like a Q8, or at a more reasonable power level like 3 Amps, because sensor lag is much less of a factor there.

So I’m wondering if I should change it. It currently regulates fine from the default ceiling level of ~1100 lm, but it doesn’t drop fast enough from full turbo.

Instead, I could make it drop immediately (or at least very very quickly) to the highest regulated level of ~1100 lm whenever it senses any overheating condition while the FET is active. Even if it’s over by just 1 degree for just a few seconds, it’d trigger a full turbo step-down and shut off the direct drive circuit. Maybe it’d still do the smooth ramp-down, but it would no longer be invisible — it’d be fast enough to see, like 4 seconds. And since there’s thermal lag, it might continue to regulate downward for bit afterward, overshooting its ideal target level, but at least the host wouldn’t get anywhere near as hot as a slower method.

Although dropping from 100% power to ~25% power is a huge change in the amount of heat generated, to the eye it looks more like it’s going from 100% to 80%. Effectively, it looks like dropping only one level.

So.

What do people prefer?

  • Smooth but slow universal method which allows full turbo to run longer but may require the user to exercise common sense and turn the brightness down manually sometimes.
  • Fast drop to a hardcoded level which is safer but may potentially be a little oversensitive and uses code instead of common sense to enforce “turbo is for brief use only”.

There is also still a muggle mode, which probably couldn’t start a fire even if someone wrapped it in dark-colored insulation with a low ignition temperature. So, use that for the kids. The question is how it should behave in non-muggle mode.


D4 managed to have a generic algorithm. Is FW3A hardware harder to regulate?

On which side of the driver is the MCU? Does if face battery of shelf?
Maybe it would be possible to shorten the thermal path by installing a silicone cube or adding a blob of thermal glue in the factory?

ToyKeeper
ToyKeeper's picture
Offline
Last seen: 54 min 47 sec ago
Joined: 01/12/2013 - 14:40
Posts: 10105
Location: (469219) 2016 HO3
Noir wrote:
BTW: have options to lower the sensor lag been discussed? For example these silicone rubber things could improve things a bit.
Agro wrote:
D4 managed to have a generic algorithm. Is FW3A hardware harder to regulate? On which side of the driver is the MCU? Does if face battery of shelf? Maybe it would be possible to shorten the thermal path by installing a silicone cube or adding a blob of thermal glue in the factory?

There’s really not room for a thermal cube. A blob of thermal glue between the MCPCB shelf and MCU might work, but then it makes the light significantly less mod-friendly.

The D4’s regulation algorithm worked, but its initial peak was too hot and it had a tendency to overshoot, oscillate, and jump around. It has a reputation as a firestarter or a “nut roaster”. The FW3A improves on those things, but the initial peak is still too hot. So I’m pondering whether to make it initially act more like a turbo step-down before handing control back to the usual regulation algorithm.

In either case, the user can still force it to stay at turbo by setting the ceiling to max, ramping up to it, and holding the button down. Or they can change the thermal config values to make it run hotter, which will extend the turbo time.

varbos
Offline
Last seen: 9 hours 11 min ago
Joined: 02/16/2017 - 07:36
Posts: 426
Location: Europe

My intention is to leave ceiling at 1100 lm (or even lower it some) so I don’t really have a preference about thermal regulation at higher levels.

yokomus
Offline
Last seen: 3 months 2 weeks ago
Joined: 06/06/2018 - 13:28
Posts: 7
Location: Norway

I’ll sign up for one more light, making it a total of two (2) to my name.

Need one as backup/giveaway

Yohan110
Offline
Last seen: 2 weeks 2 days ago
Joined: 01/08/2018 - 20:56
Posts: 27
Location: United States

I would prefer option B. I always prefer one or two hard drops vs a slow dimming.

Tom Tom
Offline
Last seen: 8 months 2 weeks ago
Joined: 09/10/2017 - 08:30
Posts: 1163

TK, could it be that your thermal control works better at the lower levels because a lot of the heat is being generated in the 7135s rather than the LED. The 7135s being closely thermally coupled to the MCU and its internal temperature sensor. I suspect you are mostly controlling the temperature of the driver, not the LEDs.

ISTM there are two cases here:

A: Power (manageable) burned off in the 7135s, controllable. Driver temperature is a reasonable indication of overall head temperature.

B: FET in use, power (lots) burned off in the LEDs, no good thermal path to MCU, long time constants, driver runs relatively cool since most current passes straight through the FET.

Without having a temperature sensor on the LED MCPCB itself, I suspect trying to control it in B: is a difficult exercise. So a timed step-down bringing the 7135s back into use seems the most practical way forward.

Otherwise I think you would need two different control algorithms, but then it gets even more complicated as cell voltage drops, and in the transition region between 7135 only, and full FET.

TBH getting rid of the FET could make things much easier. A straight linear driver is the way to go in future IMO.

djozz
djozz's picture
Offline
Last seen: 3 hours 28 min ago
Joined: 09/07/2012 - 17:04
Posts: 16161
Location: Amsterdam

I’m all for thermal regulation B.

Reason: I have a D4 that I like to pocket-EDC but the lock-out requirement (manual or via software) makes a stiff tail-clicky EDC much more convenient. I really tried: my S2+-shorty switches itself on spontaneously in my pants pocket about once a month (starts at low), the D4 does that at least twice a day, and in one occasion I found it (actually my nuts found it) on turbo.

The FW3A has the e-switch at the tail and that should help a bit but I expect not much.

So with some mapping using perhaps two different emitters you could make the stepdown start at the very first detection of rising temperature. Perhaps even use the first half second after first detection to detect the rising speed as recorded by the MCU, correlate that to what the actual temperature course is of the flashlight head and calculate an appropriate stepdown response. This could also work to get the correct stepdown when not fully ramped (but over sustainable current).

But your current temperature control algorithm may already work at this level?

spaceminions
Offline
Last seen: 1 month 4 weeks ago
Joined: 07/20/2015 - 14:18
Posts: 149
Location: Texas, USA

My first reaction is to say to either use the universal method, or to use a significantly more aggressive regulation behavior when the fet is active but NOT go as far as to force a fast stepdown like with option B. Mostly it’s just uncomfortable to the user rather than actually harmful when the body rises above the setpoint momentarily; actual burns are from having things in front of the beam.

I’d say use normal behavior up to max 7135, then in FET use a more proactive mode. (Would a trend projecting part work here?) If there isn’t a set of parameters that can make the regulation work right at fet mode (a set which would have been too jumpy at lower modes) then maybe loosen up the constraints a little. Let it get a bit hotter than normal while it quickly regulates down to somewhere between turbo and the 7135 mode (it will be visibly changing, but that’s not nearly as bad as a fast forced stepdown). I should try some of this myself, of course. Hmm…

Card Carrying CRI baby https://b.thumbs.redditmedia.com/laMC7BGhREfYTRQjYe2I5FMjTF1IjrmM6bPLK9SKEXU.png

 

KJ4BXT
Offline
Last seen: 1 week 3 days ago
Joined: 06/29/2018 - 21:24
Posts: 5
Location: Fairfax VA, USA

In for 2

Also would prefer option B

Higher CRI!

spaceminions
Offline
Last seen: 1 month 4 weeks ago
Joined: 07/20/2015 - 14:18
Posts: 149
Location: Texas, USA

Unless there’s only a handful of bytes left, maybe this could be a config option; it does seem like people could have good reasons to go both ways.

Card Carrying CRI baby https://b.thumbs.redditmedia.com/laMC7BGhREfYTRQjYe2I5FMjTF1IjrmM6bPLK9SKEXU.png

 

Kindle
Kindle's picture
Offline
Last seen: 46 min 34 sec ago
Joined: 04/22/2011 - 10:24
Posts: 715
Location: USA
ToyKeeper wrote:
I’d argue, though, that many things actually have gotten better over time, and are still doing so. There was certainly very good music made 50 years ago, lots of awesome songs. But looking at recent music, the amount of good songs is an order of magnitude more, and the production techniques have evolved to improve the quality and sophistication of the result. In both cases, it may be that only a 10% slice of the pie isn’t crap… but the entire pie has gotten bigger and fancier over time.

You know, there is a body of evidence that argues otherwise. Wink

The loudness war, decreased complexity of chords & melodies, fewer instruments, less diversity in lyrics, increased use of sampling, the growth of assembly line song writing (ie ‘track & hook’).

fritz15
fritz15's picture
Offline
Last seen: 1 year 3 weeks ago
Joined: 01/27/2012 - 13:32
Posts: 307
Location: Denmark
Kindle wrote:
ToyKeeper wrote:
I’d argue, though, that many things actually have gotten better over time, and are still doing so. There was certainly very good music made 50 years ago, lots of awesome songs. But looking at recent music, the amount of good songs is an order of magnitude more, and the production techniques have evolved to improve the quality and sophistication of the result. In both cases, it may be that only a 10% slice of the pie isn’t crap… but the entire pie has gotten bigger and fancier over time.

You know, there is a body of evidence that argues otherwise. Wink

The loudness war, decreased complexity of chords & melodies, fewer instruments, less diversity in lyrics, increased use of sampling, the growth of assembly line song writing (ie ‘track & hook’).

Does this body of evidence also have a source other than your subjective perception of today’s music? Would be interesting to read up. Thumbs Up

Pages