What impedance in an L91 lithium AA?

[bcwang]

I always thought lithium AA cells had higher current capabilities than Nimh, and especially Alkaline. Not having actually used them before, I was reading things around the web that make me question whether that is actually true.

I looked up manuals for 3 Nikon flash guns, and all 3 show Lithiums take the longest to recycle the flash, even longer than Alkaline. They do list that they last longer, but recycle slower. Here is one example.

Nikon SB-900:
Alkaline-manganese (1.5V) 4.0 sec.
Lithium (1.5V) 4.5 sec.
OxyrideTM (1.5V) 3.0 sec.
Ni-MH (2600 mAh) 2.3 sec.
Ni-MH (eneloop) 2.3 sec.

I've also read users reporting that lithiums in camera flashes get super hot during extensive use. What is the internal resistance of the L91, and the current drain capabilities of the cell?

 

[SilverFox]

Hello Bcwang,

The L91 cells have around 0.100 ohms of IR and are conservatively rated for current draws up to 1 amp. They can do better than that, but they do heat up.

Tom

 

[45/70]

Just thought I'd point out that while the Energizer Advanced (EA91) and the Energizer Ultimate (L91) both have the same IR spec, the Ultimates do have a slight advantage at higher current rates.

Dave

 

[bcwang]

That kind of internal resistance seems better than Alkaline. Couple that with a higher voltage, and it should be much better at charging a flash than Alkaline. Can you think of a reason why alkalines can outperform L91s in a flash?

 

[VidPro]

That kind of internal resistance seems better than Alkaline. Couple that with a higher voltage, and it should be much better at charging a flash than Alkaline. Can you think of a reason why alkalines can outperform L91s in a flash?

built in thermal slowdown on heating? the backup safety stuff in the battey for cooling out so they dont overheat? :thinking: PCT
i cant think of any reason, but i wont put $$$ in a flash like that, the voltage on lithium is somewhat high, which might be harsh on some flash units, and flash units eat batteries for a living, good place for a rechargable.

mabey it is highly dependant on the flash units (curcuits), cooling itself out when the voltage is so high, or (more wild speculation) the way the flash and recharge is occuring, the POP of the flash also rode off the batts more making the charge capacitor even lower after the arc ended? <--- i donno, some 3rd dimetional phase shifting so in reality the lithium output a brighter or longer flash (discharge arc), so the capacitor is more discharged.

the electical components in the flash going into THIER own thermal cooldown modes, because the "safety" in the cuircuitry itself was triggered, and they slowed down?
Unit overheating, i have seen flash units that were flashed so often (harsh testing), and had heat holding diffusion on them and melted the plastic.

actual recognition OF lithium cells by the flash curcuitry, that sees the higher voltage lithium cells are being used, and follows spec for thier use to keep from blowing them up?

 

[bcwang]

I'm pretty sure it's not the voltage as the SB-800 flash can take either 4 or 5 batteries. So 5 alkalines will have more voltage than 4 lithiums, yet in every case the lithiums charge slower. It's almost like the lithiums have some current limit in them for very high drain while alkaline will still dump current even though it drops so much voltage.

Here is some of the recycle specs for the sb-800, notice how strange the lithiums perform.

                4batts         5batts
alkaline        6.0sec         5.0sec
lithium         7.5sec         7.5sec
Nimh            4.0sec         2.9sec

 

[VidPro]

looks like you pretty much confirmed that its the battery itself.

so now just do a hard load test, on the batteries, (somewhere safe) and see if the amperage takes a dive on the lithium when the pct heats up.
flashes are what about 2-5 amps demand for very short periods of time? so slam it with a 3 amp draw and see if it starts all fine then chokes in current output sorta suddenly.

use a simple resistive load, and then you can check if the current droops or if you use a Constant Current load, then see if the voltage droops, as a constant current load would maintain the current.

i dont recall any discharge Graph on E lithiums that showed lithiums choking over certain amperages. but they are known to be made with various safety in mind.

 

[Bullzeyebill]

L91's have more to offer than high discharge rates, such as their long shelf life, low self discharge, and tolerance to extreme temperatures low and high. You want high discharge rates, quicker flash recovery, and low self discharge, then go with the Eneloops.

Bill

 

[GarageBoy]

So how come the Quantum flash batteries are always sealed lead acid?

 

[VidPro]

because lead acid is cheap? and you dont need sandbags to hold the tripods down with them and they take to cheap float charging well.

really there are some great rolled lead acid batteries, like the optimus and hawking, even though they weigh a ton.

 

[bcwang]

I found an interesting document on the L91 battery
http://www.humanedgetech.com/manuals/power/energizerspecs.pdf

This document is from 2001 so the L91 today seems to be different than before, capacity already differs as this shows 2400mah.

Basically it says L91 (lithium) has higher internal resistance than E91 (alkaline) with 180 milliohm instead of 170 milliohm.

It also mentions a 1.4amp max continuous discharge current, with a 2amp max current specification.

It also talks about a current limiting device in the batteries that basically increases resistance if the current draw is too high. It sounds like the batteries do heat up under high current draws. This is in line with photographers saying their lithium batteries heat up tremendously when used in a flash.

I also found the current E91 and L91 datasheets on the energizer website, and it seems they updated the performance a bit since 2001. Now the L91 can do 2 amp continous with a 3 amp max draw and have 3000mah capacity. The E91 alkaline shows no such limitations on current draw so likely it can deliver more current when new than a lithium.

So in summary, I'd say the current L91 outperform alkalines in most applications easily. But for things that require extremely high currents like charging a flash, or anything else pulling more than 3 amp pulses, lithiums are a bad idea since they get hot and have self current limiting built-in. No wonder a flash that can charge in 2.9 sec on nimh takes 7.5 sec on lithium AA.

I wonder if cr123 batteries have such low current limitations.

 

[wapkil]

I have tested an L91 some time ago close to its maximum current. It was able to sustain ~1.9A at ~1.2V. I used a fixed resistance and the resulting current and voltage were almost identical to an Eneloop after a few minutes of usage. I tested this L91 keeping it under load only for a few minutes. I think it got maybe a bit warmer but definitely didn't heat up.

EDIT: The 1.9A current came from the used resistance. Most probably the L91 could as well provide 2A and more (with lower resistance) but I haven't checked it. I only wanted to see how it compares to an Eneloop at ~2A.

 

[bcwang]

I have tested an L91 some time ago close to its maximum current. It was able to sustain ~1.9A at ~1.2V. I used a fixed resistance and the resulting current and voltage were almost identical to an Eneloop after a few minutes of usage. I tested this L91 keeping it under load only for a few minutes. I think it got maybe a bit warmer but definitely didn't heat up.

EDIT: The 1.9A current came from the used resistance. Most probably the L91 could as well provide 2A and more (with lower resistance) but I haven't checked it. I only wanted to see how it compares to an Eneloop at ~2A.

In a flash it is probably acting as almost a short on the cells which will probably draw the max of 3A. I'd say if you are running below the rated continuous current of 2A, it probably won't be getting hot as it should be able to sustain that for the full discharge. But jump to 3A and it'll probably be over.

Kind of dissappointing that under just a 2 amp load the lithium is running at eneloop level voltages.

 

[SilverFox]

Hello Bcwang,

I dug up some L91 cells and explored this a little further...

At 5 amps, the cell will only run for about 30 seconds, then it heats up to a point where the voltage drops below 0.9 volts. It will run a short time after that, but the voltage is continually ramping down until it drastically drops off.

At 4 amps, it runs a little longer, but the cell quickly heats up and the performance quickly drops off.

At 3 amps, it did pretty well, but the voltage was dropping off as the cell heated up. You can probably run a burst current of 3 amps, but longer runs will trigger the PTC resulting in voltage drop.

At 2 amps, it does a normal run, but it does heat up. The amount of heat produced does not seem to be enough to trigger the PTC, but the cell ends up warm at the end of the discharge.

So, if the flash application involves a surge of 3+ amps, the L91 cells would heat up, trigger the PTC, and show reduced performance. This seems to be what was going on.

Another interesting observation is that the PTC kicks in when the internal temperature of the cell reaches around 194 F and when that point was reached, the external cell temperature was at around 140 F. I usually set my cut off temperature at 140 F, but for this testing I raised it to 160 F. Obviously, there is some thermal lag involved in this type of testing, but I didn't realize the magnitude of the lag would be this great. Interesting...

Tom

 

[bcwang]

Thanks for doing some testing Silverfox. This prompted me to look at the sb900 spec sheet more closely and did some math.

For eneloop it specs:
Ni-MH (eneloop) 2.3 sec. 165/2.3 – 30 sec.
So with 2000mah eneloop, 165 flashes at 2.3 sec each = 2000/(2.3*165) = 5.27 amps

For lithium it specs"
Lithium (1.5V) 4.5 sec. 230/4.5 – 120 sec.
So assuming 3000mah which is quite a stretch at the amps it's pulling, at best it is getting 3000/(4.5*230) = 2.89 amps

And another note, of all the battery types listed in the spec, here is a footnote about the testing

"* When firing the Speedlight at full output once every 30 seconds (120 seconds with lithium batteries)."

So they even have to give the lithiums a break in testing or it'll probably completely overheat.

 

[wapkil]

Kind of dissappointing that under just a 2 amp load the lithium is running at eneloop level voltages.

I wasn't expecting the L91 to behave better than it did. Although an L91 has higher nominal voltage, 2A is a maximum (continuous) current for it while an Eneloop is still working well at 10A. I was rather satisfied with this test result.

 

[45/70]

I wasn't expecting the L91 to behave better than it did. Although an L91 has higher nominal voltage, 2A is a maximum (continuous) current for it while an Eneloop is still working well at 10A. I was rather satisfied with this test result.

wapkil, I'm curious. Where have you found that AA eneloops work well at 10A rate? I keep seeing people say this on the forums. It seems like some sort of myth that keeps spreading around. SilverFox's graph for instance, shows that the AA eneloop falls under 1.2 Volts at 3A. Granted, the eneloop will do better than most standard NiMH cells, probably better than lithium aa's, and certainly better than alkaline's under high current drain, but I wouldn't say that eneloop's perform well at 10A, they all but fall flat on their face!

Maybe it's just my idea of performing well. I not only use batteries in lights, but handheld transceivers and other higher current drain devices as well. The ability to maintain 1.2 Volts under load is a requirement, for many of my needs. To put it another way, if I needed a single cell, or series of cells, to supply 10Amps, I sure wouldn't count on eneloops! I don't even use them for the 4+ Amp load of my ROP.

I'm in no way bashing eneloops, they're my main battery contingent, but I'm wondering if I'm missing something?

Dave

EDIT: I apologize for going off topic. I won't do it again...... 'til next time.

 

[wapkil]

wapkil, I'm curious. Where have you found that AA eneloops work well at 10A rate? I keep seeing people say this on the forums. It seems like some sort of myth that keeps spreading around. SilverFox's graph for instance, shows that the AA eneloop falls under 1.2 Volts at 3A.

[...]

Maybe it's just my idea of performing well.

You are right that I wasn't specific and that it depends on the requirements. At 10A Eneloops are still able to stay above 90% of their nominal voltage. I'd have to check but I think they behave in a way similar to other batteries at their maximum rated current so I wrote that they work "well". I won't claim that this a good metric of what a "good behavior" is. Maybe the standard says how the maximum current should be defined?

 

[45/70]

OK, thanks wapkil. I guess it's probably just me and my own "standard". I'm not familiar with the actual standard or the 90% figure. I just came to my own conclusion that 1.2 Volts was a good point, as that's the voltage nickel cells are rated at, and what seemed to work for me.

Direct drive applications are where I see the voltage under load makes the most noticeable difference. Regulated circuits, of course, are much less affected.

Again, the eneloops do perform better than 90% (or more) of other NiMH cells at this load, so that's not bad at all. Most others won't work @ 10 Amps period!

Back OT!

Dave

 

[Mr Happy]

So with 2000mah eneloop, 165 flashes at 2.3 sec each = 2000/(2.3*165) = 5.27 amps

Gotta get your Units of Measure right!

2000 mAh / 1000 mA/A = 2.3 sec / 3600 sec/h x 165 x I

=> I = 2000 / 1000 / 2.3 x 3600 / 165 = 18.97 A

Either we think NiMH cells can supply 19 A (they can't) or something is not right with that calculation.

We have to look at this a different way. If you can get 165 flashes from 2000 mAh, then it is 2000 / 165 = 12 mAh per flash (we should work in mWh, but let's assume NiMH have a flat discharge curve and nearly constant voltage).

Now 12 mAh is 12 x 3600 = 43,200 mAs = 43 As.

To cycle the flash in 2.3 sec the current would be 43 / 2.3 = 19 A. Oops, same answer.

Something is wrong. The SB-900 manual says to replace the batteries when the cycle time exceeds 30 seconds. I think we have to conclude that the flash does not come close to draining 2000 mAh out of the eneloops. Probably less than half of that.