Flashlight Runtime/Output Question...Ie, output dropping after a few minutes..

Guys, I keep buying these flashlights with high output and of course, I read
the specs with runtimes and outputs and all that and I might for instance
get a light where the specs are 300 lumens at 1 hour but then it turns out
that after 3-4 mins, it drops to 40-60% of the max claimed output.

What the hell? If this is the case, how the hell can they specify an hour or an hour
and 30 mins runtime?

I WAS getting ready to order some lights and I just realized from reading
a few threads that this might be the case...?

Thanks

George

Definitely yes.
I reference the charts from Selfbuilt (in the reviews section) frequently.
(There were some Streamlight lights I was going to purchase at one time, but when I eventually found the manufacturer's runtime/output chart I changed my mind.)

Ok, so isn't this very misleading for dumba$$es like me or maybe I'm not the only one?

Okay, basically here's the idea.
They want to sell the high number lumens, but their flashlight can't handle the heat. So, they put out the high lumens for long enough time to slip past the minimum time limit to get the high output rating, and then step-down to where the light can actually operate.

Some people(including me) would call that "misleading". However, since just about every Chinese company is doing it now, you might call it "normal".
As far as I know, there are no American-made lights that do that.

It's very misleading and they should have to state how long the light will run at those given lumens not just a given mode. It's deceptive advertising but at least we CPF members have reviewers like Selfbuilt who give us the truth.

Yeah, you do need to look at Selfbuilt's output/runtime charts, and with a lot detail including reading between the lines. Unfortunately, he needs to keep a working relationship with the manufacturers and generally tries not to call out the egregious manufacturer claims. He also usually doesn't test low lumen runtimes, I think because there's large variability between different samples.

You get sense of which manufacturers are full of it though.

guiri said:

What the hell? If this is the case, how the hell can they specify an hour or an hour
and 30 mins runtime?

Click to expand...

The ANSI FL1 standard measures runtime to 10% of initial output. It allows (and almost encourages) designs that reduce output to claim long runtimes. For example, each of the following would be rated 300 lumens for 2 hours:
1. Start at 300 lumens, run 1 hr 59 minutes; drop rapidly to below 30 lumens in last 1 minute (weighted average output 298 lumens).
2. Start at 300 lumens, drop steadily for 2 hours to below 30 lumens (weighted average output 165 lumens)
3. Start at 300 lumens, run 4 minutes; reduce output to 31 lumens, run 1 hr 55 minutes; drop rapidly to below 30 lumens in last 1 minute (weighted average output 40 lumens)

..and why do we put up with it? Does no one care or is it that because they understand and know this better than
I do that they don't care?

When the original standards were adopted they were based off of the behavior witnessed with the current lighting technology at the time, incandescent bulbs. From my very limited understanding these standards were started by Surefire sometime back and they had a hard time deciding which numbers to give. With their directly driven incans, regulation was not all that stellar and light output tended to drop after a short while from things like heat build up. Surefire decided it was a good idea to rate their lumen ratings by the level of brightness after the light had been running a while. They also decided to state the runtime down to a certain level of brightness instead of dragging the runtime figure out for bragging rights when it wasn't putting out useful light anyway. That's why people often say that Surefire lumens are superior to over manufactures. They underrated their lights from the get go. Now we have much more efficient and cool LEDs available and can achieve near perfect regulation with out worrying too much about the heat, i.e. Fenix. Unfortunately measurement standards have not really kept up with the times, so this is what we are left with.

Another thing to think about is the fact that we are still in the wake of the great Surefire period where they really were leaders and standard of the market. A lot of police officers have been around longer than modern flashlight technology and will still be imprinted by Surefire and Streamlight incan lights and the way that they behave. Lights that target LEO's may keep that style of regulation because A) that's what cops are used to or B) they don't realize that we can have perfect regulation now.

That last paragraph is just me thinking aloud and wondering if it might be true. The rest of it is what I have perceived to be correct from my readings here on CPF. I haven't done any actual research into this subject so I may be completely wrong but, I think these reasons are large contributors to what we see today. Technology will always advance faster than the regulations that govern them, this is something that we will have to live with like it or not. Perhaps soon we will begin to see new ANSI standards adopted for LED lights that more accurately reflect what we can achieve with them. Another thing is the fact that some people actually like that style of regulation as it easily let's them know when their batteries are dying without leaving them in the total dark.

It's a classic case of "working the spec charts".
As I posted before, the mfrs know that all they need to provide for max lumens is the window that the ANSI protocol specifies. Then they can "officially claim" that amount of lumens. This pumps up their marketing ads.

Why do we accept it?
I don't accept it. I buy American made lights that don't do that.

guiri said:

Ok, so isn't this very misleading for dumba$$es like me or maybe I'm not the only one?

Click to expand...

What lights are you looking at specifically?


Some lights are rated to ANSI FL1 specs, these say nothing about maintaining MAX output for the entire duration.

Sure some lights are capable of constant output regulation, however in such situations they will have less run time.

Some lights will deliberately step down, although many of these will return to full power if you turn them off and on again.

Other non regulated lights will dim as the cells deplete and voltage drops. This will only get brighter if you recharge the battery.



I don't really think it's that misleading though, GM sell a 505hp Corvette, capable of 199mph and 26mpg (EPA). However it won't be able to manage 26mpg while making 505hp continuously, nor can it likely maintain 199mph for it's expected/quoted fuel range.


If you frequently turn a flashlight on and don't turn if off again for a long time or until the battery depletes and constant output is required, then you'll want to find one with a good constant output regulation and one that doesn't get too hot and step down due to thermal protection. If you tend to use your flashlight for 3 mins or under at a time, then almost all regulated lights will be fine and will offer you max output when you are using them. Although due to using max all the time their overall runtime will likely be lower than quoted by the manufacturer.

Chicken Drumstick said:

What lights are you looking at specifically?
.

Click to expand...

How about SWM M20A (2xAA): Manufacturer says, Medium 70 lumens for 24 hrs and Low 4 lumens for 200 hrs. Selfbuilt got a regulated flat line 10 hrs at output well under a 54 lumen Fenix and closer to a 36 lumen Quark Mini2. While he doesn't test low runtimes, SB thinks low is 0.3 lms!

Or a Klarus XT11 on an 18650: Low 10 lumens for 295 hrs. SB tested low to be <3 lumens.

These aren't even close.

Another trouble with runtimes is battery capacity. IIRC Selfbuilt uses 2200mAh 18650 cells. Now evidently if you use a genuine 3100mAh you are going to see a fair bit more run time.

Also different cell combo's can produce differing results, e.g.

1 x 18650
2 x 16340
2 x CR123A

In each instance even ignoring the fact you can get difference capacity cells of a same type, swapping what is powering the light is likely to affect output and run times.


All that aside though, I use my lights, when they do dim I change the batteries. Claimed run times mean practically nothing to actual usage so it's not even something I bother looking at when buying a light tbh.

Ahhh, perhaps your right. I think I see what Klarus might be quoting now.... the low lumens from 2xCR123 (5.7 lumens.... rounded up to an even 10?) paired with the runtime from an 18650 (call it 3100mah - as I believe this is the most watt-hrs of energy possible for this light). ZL, arguably the most efficient light manufactuer, can only claim 2.8 lms for 280 hrs from a 2900mah 18650..... Klarus is claiming 3x that? Will 2xCR123s outperform 1x18650 with less watt-hours of energy?

The SWM 70 lms for 24 hrs thing seems to be a "typo" as their US site says 9 hrs - good thing SWM isn't into banking. This light isn't rated for Li-ions so the best you can do is the L91 lithium primaries. I don't think the marginal extra voltage (~0.6V for 2) is going to turn 40-45 lumens into 70 and especially 0.3 lumens into 4 lumens, esp. in what appears to be a regulated light.

Things these days are sold by the numbers or "buzzwords!" It is all about marketing. I guarantee you if a grocery store took two stacks of bottled water, labelled one of the stacks as "No fat", those would sell before the other stack!!! :laughing::laughing:

Chicken Drumstick said:

All that aside though, I use my lights, when they do dim I change the batteries. Claimed run times mean practically nothing to actual usage so it's not even something I bother looking at when buying a light tbh.

Click to expand...

I find output/runtime info very helpful (when truthful) in determining a light's efficiency, esp. in the most often used med/low modes. For example, the Fenix LDs are more efficiency than the equivalent Quarks on most modes, but at a cost of narrow voltage tolerances that can go direct drive on new 2xAlks, while the Quarks can run a Li-ion. SWM magnetic rings have very elegant UIs, but at a loss of 50+% of the low lumen runtimes of a equivalent clicky.

Low lumen efficiency is one of the reasons I don't bother carrying spares.

reppans said:

Ahhh, perhaps your right. I think I see what Klarus might be quoting now.... the low lumens from 2xCR123 (5.7 lumens.... rounded up to an even 10?) paired with the runtime from an 18650 (call it 3100mah - as I believe this is the most watt-hrs of energy possible for this light). ZL, arguably the most efficient light manufactuer, can only claim 2.8 lms for 280 hrs from a 2900mah 18650..... Klarus is claiming 3x that? Will 2xCR123s outperform 1x18650 with less watt-hours of energy?

The SWM 70 lms for 24 hrs thing seems to be a "typo" as their US site says 9 hrs - good thing SWM isn't into banking. This light isn't rated for Li-ions so the best you can do is the L91 lithium primaries. I don't think the marginal extra voltage (~0.6V for 2) is going to turn 40-45 lumens into 70 and especially 0.3 lumens into 4 lumens, esp. in what appears to be a regulated light.

Click to expand...

Klarus' literature cites the XT11 will run 295 hours on low (10 lumens) using a Sanyo 2600 mAh 18650.

I tested my XT11 using a new premium RediLast 3100 mAh 18650.

My actual results:

Oct 01 - 3:15 am on
Oct 10 - 5:42 pm off

About 231 hours. :thinking:

^^ Your runtime testimonial is awesome and certainly close enough in my book... but once again, the issue I had was the lumen output claim of 10 lumens when SB found it to be ~ 2.7 lumens on an 18650.... and at that lower level, it should get about 3x the runtime of a 10 lumen mode.

EDIT - Hope I'm not being too picky... but I think if output x runtime is anywhere within 2/3rds of manufacturer spec, I'd be very happy. The Klarus spec sounds like (based on this independent data) it's not quite reaching 1/3 of spec... and SWM, not quite 1/10th?

Aren't some runtime claims theoretical too? As in based on amp draw vs battery capacity? Also when on a low mode when do you actually say it's run out of power? When it loses regulation maybe? Or when it no longer produces usable light? But what exactly is "usable light"? Or is it based purely on being able to power the emitter? If it's the latter then it would explain some discrepancies between what is observed and what might be happening if you could still measure it.


For me though, I can think of no occasions where I'd need 295 hours of low light that during that time I wouldn't be able to charge or replace the batteries. Also if it is important then it comes down to asking yourself how much leeway do you allow?

All good questions...

I would have thought independently tested around two-thirds of manufacturer output x runtime claims to be reasonably leeway - is that too high of a standard to expect? We're not even talking 1/2 manufacturer claims with these examples.

Selfbuilt generally doesn't test low end runtimes, but he does test the lumens, and I believe with fresh batteries, so I do find it misrepresenting to claim a low mode so far off the mark at the starting point! The Klarus, not so bad with a 5.7 lumens on 2xCR123s (after you pointing out the battery diff), but the SWM claiming 4 lumens when it's actually sub-lumen? (other user reviews have said the same, btw)

I also agree not ever really needing to run a light 300 hrs on its lowest mode, but I prefer not having to constantly top-up batts, or carry spares, so I generally try and deplete my batts by ~ 2/3s before recharging. Sometimes I get that wrong and find I lose my max mode meaning my battery is ~ 90% depleted? I just like knowing that 10% remaining capacity still has 30 hrs or so of emergency runtime. I'm also a camper and SHTF type and multi-hundred hr moonlight modes (I'm actually a single cell Eneloop/14500 light fan) is the only mode that I'll use continuously like its plugged into A/C.

Anyways, it's that B/S marketing claims completely turn me off to a brand altogether.. buyer beware, I guess.