ROV 3 watt (Nuwai X-3) Runtime Graphs

[Curious_character]

Here are runtime graphs of a Ray-O-Vac "3 watt" flashlight (identical to a Nuwai X-3 I got earlier). One graph is the original light and the other after replacing the emitter with a Seoul P4. The estimated total output of the original is 49 lumens, and the modified light 97 lumens using a Quickbeam type measurement box. As you can see from the graphs, the ratio of lux measurements isn't as large as the 2:1 lumen increase, showing that the beam isn't as tightly focused after the LED replacement.

This is my first attempt at posting a run time graph from data recorded by a new meter. The meter is a NIST-certified ExTech 401036, so the readings should be about as accurate as a hobbyist can get. They are, of course, representative of only a single light, and others of the same type might be significantly different. Comments and suggestions about the graph format are welcome. If they're useful, I'll try to post others from time to time.

c_c

 

[john2551]

The new rebel version of this light claims to have up to 148 lumens: http://www.nuwai.com.tw/luxeon_rebel_led/tm_325xr8.htm

 

[lexina]

The new rebel version of this light claims to have up to 148 lumens: http://www.nuwai.com.tw/luxeon_rebel_led/tm_325xr8.htm

and that's only because they are using the Rebel 080 and not the newer Rebel 100!

 

[Curious_character]

The new rebel version of this light claims to have up to 148 lumens: http://www.nuwai.com.tw/luxeon_rebel_led/tm_325xr8.htm

If I claim that my mini-mag puts out 500 lumens, does that make it put out 500 lumens? What makes you think that Nuwai's or any other manufacturer's claim, is true? It looks to me like most of the folks on CPF have been suckered into believing whatever claims the manufacturers choose to make, so they're happily claiming whatever output they need in order to sound better than the competition. It's a shame that the light outputs aren't increasing nearly as fast as the manufacturers' claims. At this rate, we should have "500 lumen" single LED lights before Christmas. Or maybe 1000. The sky's the limit.

While my lux measurements should be pretty good, I can't make any accuracy claims for the lumen estimation. Quickbeam found the method to be reasonably accurate, but it's surely subject to considerable question which he emphasized in his description of it. Yet it's a decent way to compare lights. And I certainly have more faith in even a casual measurement I've made than I do in some marketeers' dream "specification".

c_c

 

[john2551]

The stock X-3 light claimed 80 lumens on the package but tested out at 49, so it turned out to be about 60% of the claim. Then maybe in turn if the rebel versions claims 148, about 60% of that is around 90 lumens. Not bad for a ~$20 light.

 

[lexina]

This subject comes up quite often. There will be much less misunderstanding if people realise that the manufacturer-quoted lumens are usually bulb lumens at a specific current. The actual output or torch lumens will depend on inter-alia, the current you drive the led at, the resistance of the body, the type of lens, reflector, wires etc etc. Surefire is the only manufacturer I know that quotes torch lumens. I believe even Maglite recently switched to bulb lumens. As a rough guide, you can use a factor of 65% to arrive at the torch lumens. Ideally all manufactures should quote torch lumens. However, most do not want to spend the time and expense to do such tests - expense which would be passed on to the consumers anyway.

 

[Curious_character]

Well, I've gotten some good feedback about my run time posting. What I've learned is:

1. There's a lot more interest in flashlight lumen output, which we can't easily measure with reasonable accuracy, than the brightness of the main beam, which we can.
2. There's no point in posting lumen estimates from a Quickbeam-type box because it only results in argument from people who choose to believe manufacturers' claims instead.
3. There's apparently little interest in lux measurements.

I got the NIST-certified meter to satisfy my own curiosity, and will be using it to measure and compare various lights, determining the benefits of various mods, and so forth. It looks like there's not enough general interest to make it worthwhile to put together the graphs for posting, though.

Thanks for the comments!

c_c

 

[tarponbill]

It looks like there's not enough general interest to make it worthwhile to put together the graphs for posting, though.
c_c

I bought this same light, says 80 lumens on the box, had no misgivings that this would be real and would stand up to testing. And what is with the 3Watts, a real strange way to rate a light. It is understandable that Surefire and a few other manufacturers rate their lights with real test values, stands to reason when you consider their target user. The other consumer manufacturers fudge the data to say "it's just as good as ..., or better than ...".

I find your charts worthwhile, but I'm just a newbie here not a professional flashaholic, it's good to know how much fudge is in the dish. CPF is a good source of real test data, this alone should help others pick lights that fit their needs.

For $25 dollars I was happy, but knew it was unlikely to burn a hole in the garage door ...

 

[MattK]

This subject comes up quite often. There will be much less misunderstanding if people realise that the manufacturer-quoted lumens are usually bulb lumens at a specific current. The actual output or torch lumens will depend on inter-alia, the current you drive the led at, the resistance of the body, the type of lens, reflector, wires etc etc. Surefire is the only manufacturer I know that quotes torch lumens. I believe even Maglite recently switched to bulb lumens. As a rough guide, you can use a factor of 65% to arrive at the torch lumens. Ideally all manufactures should quote torch lumens. However, most do not want to spend the time and expense to do such tests - expense which would be passed on to the consumers anyway.

+1

I would add that even Surefire's numbers are still only a relative guide. Remember; their lumen claims are based on testing a few units and arriving at an average. As we all know output within every bin/batch of LED's vary so the number quoted are only an estimate. Also, if you take 3 integrating sphere and 3 operators my guess is you'll get 9 different results. Bulb lumens, which some people here 'rage' against is at least a consistent measure so long as you know the emitter bin/batch and current to the emitter - data much more easily available. The current lumen claims are, for the most part, based upon the LED manufacturers data sheets for a given batch at a given current so I'm entirely comfortable with that. Measuring LUX at a hot spot might be gratifying but it in no way gives an accurate measurement of total output - even if it can be more easily measured.

A lot of people here like to post quickbeam/Doug's data as 'gospel' and his data, as he readily points out, is imperfect and just an educated guess. What made it valuable was the consistency of the system - it was accurate in relative terms if not absolute terms.

I think the 'purists' are just going to have to accept emitter lumens for the time being since IMO NO ONE is telling you actual 'torch lumens.' Also, as has been said many times a FLASHLIGHT is the sum of more than it's lumen count.

Lastly, some manufacturers/dealers (I won't name names) have clearly been smoking crack even on the emitter lumen counts - no doubtaboutit - and that's a shame.

 

[Curious_character]

I maintain that there are at least three kinds of lumens:

1. Torch (flashlight output) lumens.
2. Bulb lumens.
3. Marketing lumens.

Using a Quickbeam box, I conclude that the ratio of number 2 to number 1 is usually on the order of 4:3, that is, it looks like you typically lose around 25% of the light in the reflector and window.

The ratio of number 3 to number 2, or number 3 to number 1, however, isn't as easy to quantify. I'm sure it sometimes exceeds 2:1, but the sky's the limit -- it's limited only by the dreams, eagerness to compete, and willingness to exaggerate of the marketing departments. Those factors can be powerful indeed. And marketing lumens is what nearly all manufacturers specify. Some manufacturers seem to have a fairly fixed exaggeration factor, and others seem to use whatever numbers they think will impress their rightful prey. Because there's no way for the average or even advanced user to tell how (un)truthful the claims are, the ground is fertile for unrestrained inflation.

I realize I'm fighting a losing battle. People want really, really badly to believe that their $2,000 speaker cables sound better than the $10 ones, and that an astrological chart will predict their future. And they really want to believe the manufacturers' lumen claims. The evidence is equally compelling for all, and there's really no way to convince them otherwise.

c_c

 

[adirondackdestroyer]

Very impressive results. Was this a difficult mod at all? Was the head glued or did it just screw right off?
The reason I ask is because I've seen this light at Walmart and would consider trying it as my first real mod if it is easy enough.

On the topic of lumen claims....If one person had a device like Doug and was able to compare most up to date lights to each other than at least we could get a relative idea as to which ones are brighter than each other. The person could also test an original U60 which were proven to be right around 60 lumens and compare the numbers to that (ex. if the U60 scored a 1000 total ouput then if something scored a 2000 we could assume it is around 120 lumens). Yes/no?

 

[Curious_character]

Very impressive results. Was this a difficult mod at all? Was the head glued or did it just screw right off?
The reason I ask is because I've seen this light at Walmart and would consider trying it as my first real mod if it is easy enough.

It's one of the easier ones. Look for postings by EngrPaul - he posted a very good step-by-step tutorial with pictures on replacing a Lux with a Seoul. That's essentially the method I used. You can sometimes improve the X-3 focus a bit afterward by sanding down the thread end of the head so it will screw down a bit more. Also, I took a piece of thin tape (I used Kapton), punched a hole just the size of the SSC dome, and put it over the emitter so it covers the black part surrounding the dome. Otherwise, the metal reflector can short the LED conductors when the head is screwed down tightly.

On the topic of lumen claims....If one person had a device like Doug and was able to compare most up to date lights to each other than at least we could get a relative idea as to which ones are brighter than each other. The person could also test an original U60 which were proven to be right around 60 lumens and compare the numbers to that (ex. if the U60 scored a 1000 total ouput then if something scored a 2000 we could assume it is around 120 lumens). Yes/no?

I made a Quickbeam type box some time ago and have used it a lot for just that purpose. However, I've slowly learned better than to post even approximate lumen measurement numbers. People choose to believe manufacturers' claims, because it makes them feel better to think they have, say, a 170 lumen light instead of a 110 lumen light. And if you post that flashlight A puts out more lumens than flashlight B, I'm sure it'll bring responses arguing that flashlight B is "rated" (by the manufacturer or dealer) as having more lumens than flashlight A. All I've gotten from posting numbers is a lot of flak -- the problem is, I can only measure approximate flashlight lumens, and don't have any way to measure marketing lumens.

I'm leery of any "standard" light, because I see quite a bit of variation from one light to another, both in candlepower (lux at one meter) and in light box lumen approximation measurement. Also, there's a lot of variation among new batteries, let alone ones with different histories and charge states. All those variables are enough to result in considerable disagreement among measurements of the same flashlight type. (Meters can be different, too, although I've got one now that's NIST-certified. My old $30 eBay meter turns out to be surprisingly good, though, so at least some of the other cheap meters must do a decent job.) Nonetheless, I've often gotten both lux and lumen results which are reasonably close to those on the Flashlight Reviews website, so I think that if we did have exactly the same light with the same batteries, our measurements would agree pretty well. But I'll keep my approximate lumen measurements to myself.

c_c

 

[NA8]

Lux is fine with me. Good work.