Question about L1 runtime on high - doesn't seem right....

[KDOG3]

I see that the E1Ls' runtime on high is 8.5hrs with a rated output of 45 lumens. The L1s' runtime on high is a mere 1.5 hrs with a rated output of 65 lumens. I think we can agree that Surefire underrates the output of their lights. But the L1 only puts out 20 lumens more yet its runtime is roughly 1/7th that of the E1L. Why is that? Has anyone done a runtime plot for the new L1s? I wonder if they underrated the runtime too. Just doesn't seem right for the output. In my feeble mind, at 65 lumens (which is probably 85 -90 ish in reality but the point being) It should go for at least 2 SOLID hours before dropping out of regulation. Anyone else see where I'm getting at?

 

[js]

Yes. I see what you're getting at. I have no experience with the E1L, but I do own an L1 Cree, and I can tell you right off the bat that the L1 is not a constant current regulation circuit. At the start of the run it is brighter than at the end of the run. OK. So there's that. Next, I think the runtime is a lot longer than 1.5 hours, like maybe 2 or 2.5 hours ??? And it doesn't drop out of regulation, just gets progressively dimmer. I did do a runtime test when I first got the light, but I can't remember the result right now, and am here at work, not at home. As for the output, I haven't tried to accurately estimate it yet, but I suspect that you are just about right at 85 to 90 lumens.

That's all I can add. Hope it helps.

 

[jzmtl]

L1's runtime is spot on, steady decline from on, 1.5 hours to 50% output. SF are kinda lacking in efficiency department, not much we can do about that.

As for E1L's runtime, I think it was around when P60L was released SF started to use inflated runtime which include both regulated and moon mode. So in reality E1L's runtime in regulation is nowhere near 8.5 hours. I don't think there are any tests on the two stage E1L yet.

 

[js]

L1's runtime is spot on, steady decline from on, 1.5 hours to 50% output. SF are kinda lacking in efficiency department, not much we can do about that.

As for E1L's runtime, I think it was around when P60L was released SF started to use inflated runtime which include both regulated and moon mode. So in reality E1L's runtime in regulation is nowhere near 8.5 hours. I don't think there are any tests on the two stage E1L yet.

SF is lacking in the efficiency department?!? Not in the L1 circuit, I can tell you that much! That's a very efficient converter, probably party because it is boost, not fully regulated, not PWM. Not sure why you would think the L1 Cree circuit is inefficient! In terms of output x runtime, it kicks a**, in my experience. The NG in my LS20, or GD in my Ti-PD-S don't beat it in terms of an output-runtime calculation. And those are highly regarded converters. But, I'm happy to be proven wrong! What comparison would you draw that would cast the L1 converter in a bad light?

 

[Kestrel]

 

[1996alnl]

I remember not too long ago i turned on my L1 with a fresh primary cell and with a stop watch nearby it stayed lit for just over 2 hrs. at that point then it got a bit dimmer.I was very impressed to say the least.
Don't own a E1L so i can't speak for that light.

Take care

 

[jzmtl]

SF is lacking in the efficiency department?!? Not in the L1 circuit, I can tell you that much! That's a very efficient converter, probably party because it is boost, not fully regulated, not PWM. Not sure why you would think the L1 Cree circuit is inefficient! In terms of output x runtime, it kicks a**, in my experience. The NG in my LS20, or GD in my Ti-PD-S don't beat it in terms of an output-runtime calculation. And those are highly regarded converters. But, I'm happy to be proven wrong! What comparison would you draw that would cast the L1 converter in a bad light?

My assessment is from comparison of runtime/output of L1 and other lights data found on CPF. Here I will compare L1 with EX10.

L1 output is said to be underestimated, but there is no actual test result, so I will take the highest estimation in this thread, which is 90 lumen. EX10 is estimated to be 110~130, and I will take the lowest which is 110 lumen. Now overlay the runtime plot of both adjusted for scale, we have this. Lighter blue is EX10, darker blue is L1, should've changed one of those to not-blue.

 

[270winchester]

My assessment is from comparison of runtime/output of L1 and other lights data found on CPF. Here I will compare L1 with EX10.

L1 output is said to be underestimated, but there is no actual test result, so I will take the highest estimation in this thread, which is 90 lumen. EX10 is estimated to be 110~130, and I will take the lowest which is 110 lumen. Now overlay the runtime plot of both adjusted for scale, we have this. Lighter blue is EX10, darker blue is L1, should've changed one of those to not-blue.

well, until a integrated sphere test is done it's all just "estimates".

 

[jzmtl]

99% of the data on CPF is done without an IS, that's what we have to work with, unless you feel like to donate one. Or I could graph it according to published spec, 130 vs. 65.

 

[270winchester]

99% of the data on CPF is done without an IS, that's what we have to work with, unless you feel like to donate one. Or I could graph it according to published spec, 130 vs. 65.

eh, most of the manufacturers don't use realistic methods to determine output at all and CPF seems to enjoy the coolaid, why should I bother arguing otherwise?

 

[jzmtl]

eh, most of the manufacturers don't use realistic methods to determine output at all and CPF seems to enjoy the coolaid, why should I bother arguing otherwise?

So manufacturer's number is not realistic, CPF number is no good because it's not done with IS, got any better ideas?

 

[270winchester]

So manufacturer's number is not realistic, CPF number is no good because it's not done with IS, got any better ideas?

yes, a side by side comparison/run time test instead of "estimate" the output then superimpose the "estimated" run time plots on each other.

 

[jzmtl]

yes, a side by side comparison/run time test instead of "estimate" the output then superimpose the "estimated" run time plots on each other.

Which is not avaliable and I don't have EX10 to work with.