Resistance of RCR123 3.0V versus primaries

[Blindasabat]

I have noticed that when I put a 3.0V RCR123 in just about any of my lights, it is twice as bright as the same light with a fresh CR123 primary.

I get less run time, but for lights that can't handle 3.6-4.2V this is a nice side benefit (when I want more light, of course).

I have seen it in my Surefire E1L Cree, Peak Rainier, Peak McKinley, SF L1 head DD on E1 body, and a couple others.

I know 3.0V RCR's are actually at about 3.2V in use, but is that small difference making such a big difference in output in regulated and DD lights? I suspect the internal battery resistance is much different - like Alkalines versus NiMh.

Anybody have any other explanation?

 

[HKJ]

This only happens with single cell lights, not with multi cell lights.

The problem is that the light goes into "direct drive", i.e. the current runs around the electronic control and directly into the led. This disables any current (brightness) control and forces the led to some undefined* high current.


*It is not really undefined, but the parameters to calculate it are not specified anywhere and will depend on the flashlight electronic, the led and the battery.

 

[Mr Happy]

Hmm. The interesting question here is whether your lights are running much dimmer than they should be when using a CR123 primary, or whether they are being over driven on the RCR and possibly getting a shorter life?

I would guess that the 3V RCR cells can maintain voltage under load much better than the primary cells. If they are the regular Li-ion chemistry with a regulator circuit they have a lot of head room for voltage regulation and also a very low internal resistance and high current capability.

It depends on the brand, but primary CR123 cells are not so good at moderate to high discharge currents. I tested some UltraLast cells and their output voltage dropped through the floor when I tried drawing more than 1 amp out of them. I suspect the quality brands perform better than this, but maybe not tremendously so.

 

[mdocod]

well, in reality, under the load of a flashlight, a CR123 is not a 3.0V cell, they dip under their resistance to ~2.25-2.75V depending on the load in question, so it's not 3.2V vs 3.0V, it's 3.2V vs 2.XV. So the voltage difference is a little wider than you originally anticipated.

Personally, I have heard various reports of regulated RCR123s actually operating higher than 3.2V, see, as I understand it, for the purpose of simplicity, they are *probably* just using a diode with a 0.7V drop on the cell. If my math is correct, it would be as high as ~3.4V under the light load of a LED light fresh off the charger, diminishing to ~2.5V when dead. I'm doubting they use any sophisticated or accurate voltage regulating electronics. With that in mind, the difference could be even higher, with 3.4V output on a freshly charged cell, and a light with a low Vf Cree, it's very possible that they could be going into direct drive and running noticeably brighter, but I still feel this slightly higher operation is reasonably safe for the flashlight in most cases.

 

[winston]

well, in reality, under the load of a flashlight, a CR123 is not a 3.0V cell, they dip under their resistance to ~2.25-2.75V depending on the load in question, so it's not 3.2V vs 3.0V, it's 3.2V vs 2.XV. So the voltage difference is a little wider than you originally anticipated.

Personally, I have heard various reports of regulated RCR123s actually operating higher than 3.2V, see, as I understand it, for the purpose of simplicity, they are *probably* just using a diode with a 0.7V drop on the cell. If my math is correct, it would be as high as ~3.4V under the light load of a LED light fresh off the charger, diminishing to ~2.5V when dead. I'm doubting they use any sophisticated or accurate voltage regulating electronics. With that in mind, the difference could be even higher, with 3.4V output on a freshly charged cell, and a light with a low Vf Cree, it's very possible that they could be going into direct drive and running noticeably brighter, but I still feel this slightly higher operation is reasonably safe for the flashlight in most cases.

Would the fact that most of these cells use 4.4v chargers affect your math? Just because the charger outputs that voltage doesn't mean the cells actually come off the charger at 4.4v, do they?
-Winston

 

[Blindasabat]

I think he is starting at 4.2V minus 0.7V, then sagging only 0.1V under light load. While a CR123 starts at a lower voltage and sags more.

Would the fact that most of these cells use 4.4v chargers affect your math? Just because the charger outputs that voltage doesn't mean the cells actually come off the charger at 4.4v, do they?
-Winston

I think it is time to do some investigation of operating voltage. I have taken a few amp measurements in the past, finding that Lux3 draw a lot more current than SSC (both direct drive again), but not much difference between CR and RCR in current. I did not measure voltage at the time.

 

[Blindasabat]

OK, test results!!!

I used Two Surefire L1 heads, one with SSC and another with Lux3 and three different batteries, 3.7V RCR123 (measured 3.95V open), 3.0V Tenergy RCR123 (3.9V open(?)), and a moderately used CR123 (3.05V open). I meaured each data point several times to make sure they are consistant. Remember, this is all direct drive using an E1 body so there was no driver in the loop.

The data:

Battery ...... L1 SSC ....... L1 Lux3
3.7V RCR ..... 560mA ........ 750mA
3.0VRCR ...... 130mA ........ 270mA
CR123 ........... 42mA ......... 50mA

HUGE differences!

Yes, as I have noticed before, the SSC draws less current than the Lux3.


Just for comparison, I measured a Fenix P2D with the same CR123 and the current draws were 50, 220, 470, and 1100 mA with a boost driver.