Actual Lumen Readings in 10.5in Sphere.

[Noctis]

Yes, the stock Z41 twisty turned out to have a relatively high resistance, when Moddoo did a comparison.
What indeed makes me wonder, too, is that this should of course make no difference with a regulated light.

It's never really been clear to me how regulation works, but I suspect all it does is to limit the current running through rather than keeping the current from falling below a certain level.

If the resistance is high enough to restrict the current draw down to about 2 amps, then the module would only draw 2 amps even if it's regulated for 2.8A.

But if the resistance is near zero(say, with a solid piece of copper instead of a spring) and the light would draw 10A in direct drive, then regulation would force the current to drop to 2.8A.

That's what I'm thinking anyways. It would be interesting to put regulation on the Moddoo Triple so that it runs with 4.5A and then lower the resistance down to nothing. At the very least that ought to minimize sag from the battery while making sure it doesn't draw enough current to damage the emitters or the battery.

[jaundice]

Here is the Malkoff Devices WildCat XP-G R5 Tripple


I don't have AW 18500 cells to test, but I am working on that.

Jose;

Do you need me to bring over a pair of IMR 18500s? I'd be happy to do so, later this week. I'd love to get a chance to check out the Wildcat Triple in person.

-John

[elho]

It's never really been clear to me how regulation works, but I suspect all it does is to limit the current running through rather than keeping the current from falling below a certain level.

That depends on the type of regulation, you described a buck regulator (or a linear one, which limits current by burning excess voltage).
The whole point however is, that it as you say is supposed to limit the current to 2.8A at which the two XP-Gs are not supposed to become blue unless something is wrong.
There might be a problem with heatsinking of that particular drop-in (was it installed without the outer spring to have the brass base make direct contact with the host?) or it might be an unlucky pick of very different Vf for the LEDs so that one of them gets a notably higher current and turns blue, or....
If it was a fundamental problem of a LiIon cell having too low a voltage to go into regulation with that driver, then 2xCR123 should, even when sagging down quite a bit, have high enough a voltage to reach regulation and would make the LEDs turn blue, too. With probably quite some happpy Linger Special users on 2xCR123 out there, that theory sounds unlikely. Also note that the Z41 twisty turned out quite bad, a Z59 clicky is much closer to a McClicky, so it is also likely, that Z59 (and other non-SureFire) users could face that problem with LiIon cells which also does not seem to be the case.

It would be interesting to put regulation on the Moddoo Triple so that it runs with 4.5A and then lower the resistance down to nothing. At the very least that ought to minimize sag from the battery while making sure it doesn't draw enough current to damage the emitters or the battery.

No. You first have to lower resistance in the host before it draws that much. You of course can not reduce it to noting and it will not affect battery voltage sag (unless you lower the internal resistance of the battery), only total voltage sag as seen from the LEDs.

[bigchelis]

Okay just a quick update:


I have a Nailbender Cool White Tint Linger Special and tested it stock as received from Nailbender. My unit has a 2.8A driver too.


I put a McClicky switch and the tint turns blue, then I put Arctic Silver 5 on the threads and I get less lumens


I changed hosts bezels and multiple clickies/twisties to try and see whats wrong, but I still get 2.8A at the tail and the lumens are still less then when I got it.



bigC

[gswitter]

I imagine this has been covered elsewhere, but I'm curious...

I assume the Linger Special has the LEDs wired in parallel like the Moddoo triple? Do either the Linger Special or the Moddoo triple do anything to limit the current to the individual LEDs? Do the builders do anything to ensure the Vf of the LEDs are balanced?

[bigchelis]

Here is a Nailbender SST-90 P60 drop-in with a UCL lens and 6P hosts (bored for 18650 cells). The owner sent it with copper tape and AS5 on the threads so, I never tested it without those things.

[bigchelis]

Here is my Nailbender Linger Special cool tint with 2.8A driver, which provides 1.4A to each XP-G R5.

Please Note: That the AS5 readings were taken after the tint turned blue. I shut it off right away, but it should have never happened because its regulated and 2nd it may explain why I get less overall lumens.



The warm Linger Special with same 2.8A driver

[Noctis]

Here is a Nailbender SST-90 P60 drop-in with a UCL lens and 6P hosts (bored for 18650 cells). The owner sent it with copper tape and AS5 on the threads so, I never tested it without those things.

I feel so proud.
Though it seems like running the Li-Co cell is more beneficial in the long run due to less sag and heat.

[gswitter]

Please Note: That the AS5 readings were taken after the tint turned blue. I shut it off right away, but it should have never happened because its regulated and 2nd it may explain why I get less overall lumens.

And, again, I'll ask...

Are the emitters on the Linger Special wired in parallel? (Attempting to drive them at 2.8A suggests they are.) And, is the current to the individual emitters limited in any way? (I doubt it.) So did Linger, Nailbender or whomever put those emitters on the board check the Vf of the emitters to ensure they were equal (or extremely close)?

At 2.8A, you're close to the recommended limit for the XP-G. If there's a significant difference in Vf of the two emitters (and at 1.4A, even a small difference can be significant), one of the emitters is probably drawing significantly more current than the other, overheating and there's your angry blue.

[bigchelis]

And, again, I'll ask...

Are the emitters on the Linger Special wired in parallel? (Attempting to drive them at 2.8A suggests they are.) And, is the current to the individual emitters limited in any way? (I doubt it.) So did Linger, Nailbender or whomever put those emitters on the board check the Vf of the emitters to ensure they were equal (or extremely close)?

At 2.8A, you're close to the recommended limit for the XP-G. If there's a significant difference in Vf of the two emitters (and at 1.4A, even a small difference can be significant), one of the emitters is probably drawing significantly more current than the other, overheating and there's your angry blue.

The only way to get the 2 XP-G R5's to work off a single cell is if they are wired in parallel. I highly doubt that Nailbender would have matched to the best of his ability two emitters with close vf and this is probably why I saw the blue tint.

I didn't think that having one emitter with more vf (even just a bit) would have caused such a disparity in current dristribution, but it makes perfect sense now. Otherwise; a regulated 1.4A XP-G R5 would never have turned blue.

[elho]

Though it seems like running the Li-Co cell is more beneficial in the long run due to less sag and heat.

Definitely. At 30 seconds, the difference in Lumens already is less than 10% and IIRC 10% difference is what it takes for a human to actually recognize (or even tell apart? ).
Either way, the IMRs are nice to get a high peak reading, but in practice the runtime advantage of the LiIons is a much stronger argument.

Nice to see the nailbender drop-ins to hit the magic 1000 Lumen mark, even if just briefly.

[elho]

And, again, I'll ask...

Two posts before your first question which I also just read:

it might be an unlucky pick of very different Vf for the LEDs so that one of them gets a notably higher current and turns blue

And to specifically answer all of your questions:
The LEDs are in parallel, there is no "extra current limiting" (which in fact would be individual drivers). Linger had the idea to cram two XP-Gs into one reflector opening and approached nailbender with that idea, who picked it up and started to build the "Linger Special" as he liked the result.
I'm pretty sure nailbender does no Vf testing, I previously approached him about the possibility of it, but he refused to do the extra work of testing a whole bunch of LEDs instead of just taking the next one on the roll.
Given how interest in his drop-ins has risen since then, he would not have the time to do anything like that these days either.
But most importantly, he is very helpful, so if you have a problem like that with one of his drop-ins, approach him and I am sure he will find a solution for it.

[elho]

The only way to get the 2 XP-G R5's to work off a single cell is if they are wired in parallel.

No, you could design a driver with two individual 1.4A outputs, or simply connect two individual 1.4A drivers in parallel to the battery and connect one LED to each. In practice just wiring them in parallel to an existing single driver is of course a lot easier.

I didn't think that having one emitter with more vf (even just a bit) would have caused such a disparity in current dristribution

Just a bit obviously does not make such a dramatic difference, or every Linger Special out there had that problem. I think Vf can vary quite a bit and that drop-in might be a very unlucky pick of one LED with relatively high Vf and another with relatively low Vf. Put together you suddenly get quite a difference.
Furthermore, the higher currents you deal with (at such low voltages), the more difference a slight difference in resistance makes.

[psychbeat]

Nice to see the nailbender drop-ins to hit the magic 1000 Lumen mark, even if just briefly. [/QUOTE]


Yeah! PROPS to Noctis too for the research and BURNING sh!t up!!

[bigchelis]

No, you could design a driver with two individual 1.4A outputs, or simply connect two individual 1.4A drivers in parallel to the battery and connect one LED to each. In practice just wiring them in parallel to an existing single driver is of course a lot easier.


Just a bit obviously does not make such a dramatic difference, or every Linger Special out there had that problem. I think Vf can vary quite a bit and that drop-in might be a very unlucky pick of one LED with relatively high Vf and another with relatively low Vf. Put together you suddenly get quite a difference.
Furthermore, the higher currents you deal with (at such low voltages), the more difference a slight difference in resistance makes.

Maybe I should bug Nailbender for an exchange and perhaps the next one will be somewhat better

[psychbeat]

i still think it would be cool on the Linger if there was
a way to shorten the reflector and add some kind of
copper slug to it...
the reflector dosent need to be big since its a flood
light anyways right?

just an idea...

hope my warm one never turns blue on me

probably safe - ive run it til cutoff multiple times..

[Moddoo]

I have not tried the linger special unit yet.

But, I do want to bring up 1 important feature:

The 2 XPG are mounted on a single board, and this board only has a solder connection to about 1/2 of the thermal pad on each emitter.

This means that the heat from the emitters cannot be drawn away as well as it can be when a single emitter has complete contact to the MCPCB.

This may be part of the problem that you are seeing.

[csshih]

Definitely. At 30 seconds, the difference in Lumens already is less than 10% and IIRC 10% difference is what it takes for a human to actually recognize (or even tell apart? ).

more like 30%, it feels.

[recDNA]

Maybe I should bug Nailbender for an exchange and perhaps the next one will be somewhat better

Hey BigC! How about updating your data in GoogleDocs?

[recDNA]

I have been itching to test this Moddoo Tripple XP-G R5 real bad. This drop-in is all that and more way more wayyyyyyy more

I can't tell you how broken hearted I am that I missed out on this one.

[JCD]

Deleted

[Nitroz]

I can't tell you how broken hearted I am that I missed out on this one.

You and everyone else.

[bigchelis]

You and everyone else.

Well the you better sign up for the XP-G R5 Quad....Its not Moddoo's but at $155 you get 4 XP-G R5's driven at 1A~1.2A each so no problem using big IMR 26650 cells

http://www.candlepowerforums.com/vb/...d.php?t=267366

[gswitter]

The only way to get the 2 XP-G R5's to work off a single cell is if they are wired in parallel. I highly doubt that Nailbender would have matched to the best of his ability two emitters with close vf and this is probably why I saw the blue tint.

There are drivers out there (like Surefire's Lux V drivers from the KL4, U2, etc) that could easily drive a pair of XP-G's in series with a single Li-ion.

Ultimately the angry blue you're seeing is a heat sinking issue, but it's likely accelerated by a difference in Vf of the two emitters.

I didn't think that having one emitter with more vf (even just a bit) would have caused such a disparity in current dristribution, but it makes perfect sense now. Otherwise; a regulated 1.4A XP-G R5 would never have turned blue.

The XLamp binning sheet shows the Vf bins have a range of .25V. I doubt the distribution is even throughout that range, but pair enough emitters together without matching the Vf's and you'll eventually get a pair with a significant difference.

Now consider the Vf/If graph in the XP-G data sheet. If you start with a pair of emitters at opposite ends of their Vf bin (the aforementioned difference of .25V) and you drive them in parallel at 2.8A, the difference in If could be nearly an amp. Ouch. Even a difference in Vf @ 350mA of .1V would result in a difference in If when driven to a combined 2.8A of 300-400mA, and one of the emitters will be seeing 1.6A.

There's nothing inherently wrong with the Linger Special, but driving it at 2.8A doesn't leave enough headroom to accommodate these differences in Vf.

[jac2001]

Big C... PM replied. It's happening!!! I'm boxing up the stuff right now.

Here's what he is getting for testing since these are what I use for work. Hosts are as-in with no modifications so this is the real deal without fluffing (UCL's, aftermarket switches, etc).

1. Solarforce L2P host with Nailbender SST-90 one mode regulated drop-in. The one mode drivers are 2.8A iirc..
2. Thrunite Catapult V2
3. Surefire Z2 host with Malkoff M61
4. Surefire X300 weapon light (rated "110" lumens)
5. Surefire LX2 (rated "200 lumens")

[021411]

BigC, my lights should be delivered today. Shoot me a pm when you get them. I'll send you the donation and return shipping $$.

[orbital]

+

Big C, thanks for your work on the lumen readings.

Please consider making this list alphabetical,
getting serious vertigo scrolling through the list...

[ti-force]

+

Big C, thanks for your work on the lumen readings.

Please consider making this list alphabetical,
getting serious vertigo scrolling through the list...

An alphabetical list sounds like a good idea. Why don't you get started on that? The list is already there.

[flashfiend]

An alphabetical list sounds like a good idea. Why don't you get started on that? The list is already there.

No way to use data sort to accomplish this?

[ti-force]

No way to use data sort to accomplish this?

Not with the spreadsheets I made for BigC. I made them in Photoshop from scratch. If BigC wants to, that's fine, but personally I'm just thankful that he's giving us all of this data. I know it would be nice to have it sorted somehow, but I really don't think BigC has found anything that works with this forum and doesn't take more time to enter the data than to measure the data.