Wet blanket on the E01 party
- Thread starter Don99
- Start date
Help Support Candle Power Flashlight Forum
this_is_nascar
Flashaholic
Guys, you've beaten the 10 lumens vs. 5 lumens vs. 3 lumens to death for Pete's sake. Sure, we all expect published specs to dead-on accurate, but at the same time, you need to exercise a bit of common sense. If you believe other's product claims of a certain lumen in a 1 x AAA with Nichia GS configuration and then consider the published run-time of the E01 and think that both the published runtime AND lumen rating is accurate, you're not firing on all cylinders if you really thought the brightness would be the same. Get over it and move on.
... and if you can't understand what I'm getting at and need me to be more detailed, consider this. Maybe this will sink in.
The Arc-AAA GS version is published to produce 10.5 lumens for 5-hours to 50% brightness. Assuming you believe that and take that as your benchmark, how the hell do you think another light using the same LED and same power source is going to produce that same amount of light with 2, 3 or 4 times the run-time. Do you really think Fenix found a way around the law of physics to produce a light like this? No, they did not, but what they did manage to do is actually produce a light that is not only flat-lined regulated, but runs 2, 3, 4 times longer than the benchmark, that's only slightly dimmer than that benchmark and by the way, it's actually regulated, unlike the benchmark.
And speaking of accurate specs, when's the last time you didn't purchase or did return an automobile because it didn't match the published MPG rating?
... and if you can't understand what I'm getting at and need me to be more detailed, consider this. Maybe this will sink in.
The Arc-AAA GS version is published to produce 10.5 lumens for 5-hours to 50% brightness. Assuming you believe that and take that as your benchmark, how the hell do you think another light using the same LED and same power source is going to produce that same amount of light with 2, 3 or 4 times the run-time. Do you really think Fenix found a way around the law of physics to produce a light like this? No, they did not, but what they did manage to do is actually produce a light that is not only flat-lined regulated, but runs 2, 3, 4 times longer than the benchmark, that's only slightly dimmer than that benchmark and by the way, it's actually regulated, unlike the benchmark.
And speaking of accurate specs, when's the last time you didn't purchase or did return an automobile because it didn't match the published MPG rating?
C4LED
Enlightened
Nicely summarized. I'll be getting at least two.
Brownstone
Enlightened
I agree with what you have implied; the E01 should be benchmarked against other similar products, namely 1xAAA LED lights from China.
We shouldn't be justifying the E01 by accepting any perceived faults because it is only one-quarter the price of an L2D (for example). Instead, we should justify paying 4x the price of a similar light due to the benefits the E01 brings.
stitch_paradox
Enlightened
I just got both of my E01, I was excited until I turn it on. I was thinking, " so this is it?" It has the so~so effect on me, neither here nor there. My promotional keychain light is brighter than the E01. IMO, it cannot be compared to the newer ARCs. The only consolation that I get form it is its regulated, it has long runtime. Its great for my surviving it the forest or for a long blackout, but not really an EDC for me. My 3 year old daugther loves it though.
A lot of 1xaaa lights can beat the E01 in brightness, if you want brightness you buy something ells, but then you sacrifice runtime.
I am not aware of anything that has better runtime/brightness ratio. I would be willing to pay $30 just for the regulation alone. And why should it be benchmarked against other lights from china when it's so clearly one of the best in it's class, no matter what country you pick the competitors from? It beats every similar light I can think of by a long shot. Perfect regulation, extreme efficiency. Why pay more for less?
It's twice as bright as the E0, what more can we ask?
Last edited:
Probedude
Enlightened
You've left out the driver in between the power source and the LED.
I find it ironic that JZMTL was saying the same thing when the E01 was first announced and many people were 'correcting' him using Fenix's data. Fenix's high efficiency driver was the reason cited back then for the near similar output but the much longer run time.
marduke said:
True, I have not yet received mine and in fact I don't own a lightmeter yet. I saw someone's post that said their's measured 7-8 lux in their lightbox which is what I'm going on. It's the whole "it's only $15.00, what did you expect" answers that strikes me odd.
The Arc dc-dc converter has been tested extensively by CPF'ers and found to be around 85% efficient if I remember right. Therefore no 2x-3x improvement through converter efficiency is possible. What is true, however, is the Arc overdrives the led to get more lumens out, at the expense of luminous efficiency in the led itself, i.e. the led itself is less efficient at the power level the Arc runs it at. This is a design tradeoff that has been discussed extensively in the Arc threads. The E01 gets some more efficiency, and more total output, by running at lower power. Some of us wanted Arc to do the same thing (or at least offer a low power model), but (at least according to Peter) Arc always found that the majority of its customers wanted the higher power level, so that's what they offer.
Arc's choice of semi-regulation instead of flat regulation is another design decision that has been discussed extensively. All these choices were made in pursuit of certain technical goals that were well thought out and reflected the designer's priorities, even if not every user had the same priorities.
Arc's choice of semi-regulation instead of flat regulation is another design decision that has been discussed extensively. All these choices were made in pursuit of certain technical goals that were well thought out and reflected the designer's priorities, even if not every user had the same priorities.
Last edited:
HKJ
Flashaholic
Nice summary. I am in favor of the E01 way to do things, I like the flat output and the long runtime. From the runtime test I can also see that the ARC is only best for a few hours, then the E01 wins.
I also believe the light has enough power, it is more than the lowest level i have selected on my NovaTac, NDI and LiteFlux.
I am looking forward to receiving the E01!:twothumbs
I could actually see the value of a single stage cree producing about 10 lumens. I'm sure plenty of you have witness the Zebralight's 10 lumen setting and know how long it runs on a single primary. A single Lithium cell could potentially power a warm tinted, 10 lumen, EO2 cree for about 15-18 hours in regulation. It would have better color and efficiency with a slight cost increase. Even if a light like that were to cost $24.99 it would perform magnitudes better than the very blue and less efficient 5mm, while costing far less than anything like it. For me it would be worth the extra 8-10 bucks.
Last edited:
baterija
Flashlight Enthusiast
- Joined
- Feb 7, 2008
- Messages
- 1,053
First - interesting article thanks.
Jarl suggested a way to manage the pain that is the bloated, slow Adobe Acrobat. You might also look at replacing Acrobat Reader with Foxit Reader. It's much, much quicker to open. Using Firefox and Foxit, I never locked up. The download window opened for a about 20 seconds. Once download was complete it took Foxit a second or two to display the first page.
We now return to the thread's regularly scheduled gnashing of teeth and wailing...
nerdgineer
Flashlight Enthusiast
Jarl and baterija - thanks for the pdf advice. I'll give 'em a try.
The GS efficiency is comparable to a Cree. The Cree has better tint. The GS has a built-in focusing object while the Cree would need a separate optic or reflector to create a directional beam. The light you're describing as an E02 is really basically the L0P or L0D from Fenix running at a low enough power level, or the Baltic XLR from Peak. These lights are considerably more complex than a 5mm light and tend to cost more.
Marduke
Flashaholic
Or a E1 v2.0
Seems half the people want an E1 Cree in a moderate output version (no change to old circuit, 3-4 hours), the other half for a long runtime version (lower current output to ~8-10 hours)
I thought that the Cree still had an efficiency advantage over the GS. Also, I was wondering if it would really need a reflector. Would the GS throw much further that the H30 for example? As a single stage light how much more complex would it need to be? Wouldn't a low regulation be relatively simple Paul?
Last edited:
2xTrinity
Flashlight Enthusiast
I checked on the White Lumen LED testing thread:
Nichia GS: 85lm/W @ 30mA
Cree R2: 150lm/W @ 30mA
The XR-E is more efficient, even if you take into account the fact you'd probably lose 20% efficiency by adding an optic. The real advantage of the XR-E though is you can enjoy better tint and absurdly high efficiency like that at 30mA, and with the flip of a switch multiply output by a factor of 20 :thumbsup:
Probedude
Enlightened
One thing that plays into runtime, and I only bring this up as information (just thought of it), is how low of a voltage can each driver work down to? On paper an 85% efficient driver that only works down to 1.1V is not going to drive an led as long as an 85% driver that works down to 0.8V.
Has anyone plotted battery voltage along with light output for both the Arc and the E01? What's the forward voltage of the GS LED?
2xTrinity
Flashlight Enthusiast
I cherry-picked 30mA to emphasize my point -- XR-E efficiency peaks there sharply, then actually DROPS at lower current. Meaning that using a PWM with "on" current of 30mA is the most prudent way to go dimmer. I actually wanted to investigate this effect myself, and hooked up an XR-E q5 to a bench supply. Brightness does in fact go WAY up between around 10-30mA, far more than 3 times. <30mA is also where perceptible color shift begins to occur. Basically, what I believe is happening is that that is where the Vf is just high to begin producing the optimum wavelengths. Below that the diode conducts current, but quantum efficiency (converstion of electrons to photons) is worse. At high currents, resistance losses begin to diminish efficiency.
Anyway, I decided to hotlink some of these graphs from "White LED lumen testing":
Peak of about 1106lm/W @ 3mA. So these "peak" at a much lower percentage of maximum output than the power LEDs.
It turns out the Rebel100 would be the most efficient alternative at 20mA:
And if you REALLY don't want any blue in your beam, here is the curve for a 3300k warm white Cree LED. Efficiency @ 20mA is about the same as the GS:
Last edited:
Well, the E01 has ultra flat regulation, i.e. the light output doesn't vary with the battery voltage. What that means in terms of efficiency as the battery runs down, I don't know. There were some tests and graphs of the Arc converter in the electronics forum a long while back. The runtime curve is also well known and can probably be mapped to battery voltage using the various discharge curves for AAA cells (or maybe extrapolating from AA cells) in the battery testing threads.
The Arc does not really try to maximize the total number of photons per battery. As I understand it, Peter through various considerations decided that an EDC light of this type should have 5 hours of runtime. He then made the light as bright as he could within the constraint of having a generic alkaline battery last that long before its sharp drop-off, with some safety margin and taking into account that batteries have varying capacity (so with good batteries you get more like 6-8 hours). That is the reason for semi-regulation as I understand it: he wanted a good output level at 5 hours, but if more power than that was available earlier in the run, he wanted to use it and get more lumens during that part of the run, not save it for more runtime beyond 5 hours.
It looks like 5mm leds are most efficient at around 5-10 mA so perhaps the ultimate in efficiency would be to use three of them like in the Peak Matterhorn 3-led version.
I personally prefer the Arc's semi-regulated scheme since I don't use the light in long enough stretches for runtime to be an issue. Batteries last me for many months. I notice the light getting dimmer when it's maybe halfway down the curve, which means there's still an hour or two left before moon mode, and I know to get around to changing the battery. With flat regulation, I have no such advance warning, it's at a constant level til boom, the light becomes very feeble in an instant.
Ray mentions that the E01 was able to light up at 0.45V, which is pretty interesting--does that mean at a super low glow, or ...?
With the Arc, you can run an AAA cell to near-extinction, then turn off the light and rest the cell for a few minutes, then turn it back on and get a good lighting level for quite a long time (1/2 hour or more), and you can repeat this several times. I'll do a similar test with my E01 when it arrives.
Last edited:
