How to ruin the runtime on a perfectly good L2P

[amanichen]

I finally got around to testing the runtime on my L2P which now contains a SSC P4. The light is definitely much brighter, but something is amiss - the runtime is quite literally HALF that of the luxeon (4.5 hours vs. 2.25 hours) on the same batteries.

According to Newbie's testing, I should have a lower power consumption assuming the driver is delivering the same current to the SSC as it did to the luxeon: http://candlepowerforums.com/vb/showthread.php?t=138503

The L2P driver a constant current driver, which would mean the SSC would have a lower Vf than the luxeon for the same drive current, and thusly less overall power draw. But then I thought maybe the driver is actually driving the SSC harder than the Luxeon. But this would only make sense if the driver is using voltage and not current as feedback, because the SSC P4 would flow more current for the same voltage, and thusly it would be driven harder, making the runtime to go he-double hockey sticks.

So, short of testing with electrical equipment I don't own (oscilloscope, DC power supply, etc.) is there anything I'm missing? Can anybody confirm whether the driver looks for the Vf or current as feedback?

 

[VidPro]

argg, this is all under the asumption that your light has some perfect current control reguardless of the Vf of the led and the voltage of the battery. the occurance is more rare than normal.

hey i dont own a lot of $150+ lights, but most of the cheap stuff is not some perfect regulation, the good stuff probably is?
lab tests are lab tests, they dont often translate into reality, that is why an MS operating system comes out to the users in BETA
ALSO
you dont nessisarily want Perfect regulation, because wouldnt that be hard on the batteries?

say you have a 2 cell light, and one cell is completly discharged, and now the magic "full regulated" curcuit is running 2 TIMES as much current through, that is sure to cause a reverse charge, and at HIGHER power. OR if the curcuit is JUST current control, you still get the same reverse charge you can get from incadescent.
that cant be good.

there is ramifications to everything, even to perfect regulation.

when you change the original design, or the original battery type, you change what they designed, any modder with METERS basically knows what will change before they change things. the best way to see it is with your bench supply, some things its better to not see

 

[amanichen]

(lots of stuff)

This really doesn't answer my question though, but thanks anyway.

 

[VidPro]

you can get a DMM (digital multimeter) that does ampmetering for about 30$ for a usable one of sufficient acuracy.

 

[amanichen]

you can get a DMM (digital multimeter) that does ampmetering for about 30$ for a usable one of sufficient acuracy.

I thought I could rely on CPF as a knowledge base, but I guess I'm wrong.

 

[VidPro]

i cant answer your question (obviously)
but i have tested the older fenix, and i can tell you IT did not go up in current consumption when the input voltage is lowered, which would indicate that the LED itself is NOT getting the exact same current reguardless of the input voltage..
in the same respect if you change the Vf i doubt that the current stays the same, ohh wow its lots brighter , but changing the led for a more efficient one is a little brighter, not a lot brighter.

you say its current controlled, mabey that model is, but i say its loosly controled and as it should be.

so if i am just bumping your thread, and ranting, then neither of us is getting anywhere. i can only provide a clue, unless you want to send it to california , and i will put it on the meter.

if you batteries were perfectally matched in capacity you tested, then i would have to assume, you see the same thing i have seen, tighten the things up (the Vin and Vf) and they get brighter, and the runtime drops?

 

[amanichen]

if you batteries were perfectally matched in capacity you tested, then i would have to assume, you see the same thing i have seen, tighten the things up (the Vin and Vf) and they get brighter, and the runtime drops?

The batteries I used (2500Mah NiMHs) were the same ones as my previous runtime tests, area year and a half from purchase date, and are only lightly used. I've tested with two pairs from the same set, and come up with almost identical results, so I've ruled out the batteries as the cause.

Again, my operating assumption is that the if the L2P does contain a "constant current driver," it maintains this by the voltage across the LED and not across a known constant resistance. The driver is set up to boost for a Luxeon I, thus, with a LED that flows more current at the same voltage (SSC P4), power consumption increases, and runtime shortens. I just figured there'd be someone on CPF who was familiar with the driver circuitry and could confirm whether my assumption is correct or not.