Thanks. I have some of the EVE 40PL, Molicell P50B, and the JP40 cells. I also ordered the 40XP with the light. So I can play with them in the new A54U light (out for delivery today). Though, as alluded to in your linked post, without decent test gear, I doubt that I will be able to tell any differences. It seems like the Molicel P50B might be the best bet as the levels will probably be very close, but they have additional capacity.
A54U? Nice!
Capacity differences for just about any cell disappear at over a few amps though as the capacity numbers are set at 1A and lower (C/2 typically).
Even the 6000mAh cells will run for a shorter time than the good 4000mAh cells will when above a moderate current level…all thanks to voltage sag. 
3 Thanks
Yes, I agree. But likely the light will get most of its use at lower levels. So hopefully the capacity gains will have some impact on my run times. THX for the reply.
1 Thank
Very interesting points. Could you explain why “voltage sag” (meaning a brief drop in voltage?) would cause the 6000mAh battery to deplete faster than the good 4000mAh?
I could be wrong, but the analogy I have in my head is like sucking water out of a well with slow and finite recharge, if you pump really fast, the level of the water drops and you run out of water level (“voltage”) until you turn the pump off and let the water level rise again…
A high discharge cell is like a well that can keep the water level above the pump for longer, but often the resovoir (capacity) is less…
2 Thanks
The greater the temporary drop in voltage, the sag, the sooner a cell reaches whatever low voltage cutoff point was set.
A 6Ah energy cell has a much higher internal resistance than a 4Ah power cell, up to several times higher. This internal resistance is what causes the sag and as the current level increases the sag increases (Vsag = Current x Internal Resistance).
At 5A the 6A cell might only have about about 0.2V of sag. But at 20A it would have about 0.7V. But the 4Ah cell might only have about 0.1V of sag at 20A.
So the 6Ah cell would be running at about 0.6V lower right from the start than the 4Ah cell. This means the 6Ah cell will hit the low voltage cutoff point a lot sooner (less Ah delivered) than it would at a much lower current level.
At a certain point, depends on the cells and the current level, the 4Ah cell will run for longer than the 6Ah cell.
There’s still all that extra charge in the 6Ah cell but it’s not usable at higher discharge current levels.
It’s what we’re all forced to use, unless run-time tests were done, but nominal capacity really is a terrible way to choose a cell. It’s measured at about an amp or less discharge rates and can drop hugely at high discharge current levels.
7 Thanks
I certainly agree, if it is the only criteria used. One needs to know first what their current requirements will be. Then I always look at things like cycle life, IR and max charge current, as well as the continuous current rating. So when using a low drain device and everything else being good, the extra capacity is a good thing. So yes, you need to take your intended use into consideration, as well as the other factors when choosing a cell, not just the capacity rating.
The P50B is a bit unique, as it is a pretty low IR, high peak current cell with relatively high capacity. Which is why I think it may be a decent choice for the A54U… especially when 95+% of my usage will be using lower output levels.
Does that make sense?
BTW, thanks for the great explanation!
2 Thanks
I absolutely 100% agree!
My mention of capacity was only as part of all the preferences and priorities that are used to make the best choice in a cell. Whatever degree of importance the capacity plays in that, big or small, the utility of the nominal capacity spec is small IMO.
2 Thanks
The utility of the nominal capacity spec is not small if the degree of importance of the nominal capacity is large, and the applicable rate of discharge will be similar to the one used to establish the capacity spec. But that practically goes without saying.
2 Thanks
True, I shouldn’t automatically discount nominal capacity. I should have mentioned that just for most of the use cases I know the current level used to set the nominal capacity (perhaps 0.6A-1.2A) is lower than the typical load current and that reduces the usefulness of the nominal capacity spec.
Curious…and pardon my ignorance…do a lot of lights run at current levels that low in typical use? I know that many can, but are they often used that way? Thanks!
1 Thank
I have the impression that most people primarily use their lights at rather low output levels, like 2 Watts or less. If my eyes are night adjusted I am usually using the lowest mode possible, in the mA range. If they aren’t night adjusted, probably around 100 lumens / ~300mA current draw, or less. If I am working indoors in a lit up environment and need a task light, I’ll want several hundred lumens, which might be around a 1 Amp draw from the cell. If I am changing a light fixture or doing indoor work in the dark, I will usually run high or turbo continuously, tailstanding the light to ceiling bounce. Outdoors I tend to use turbo intermittently as needed to see things at a distance, or for fun.
4 Thanks
My most common use of my lights is at lower levels. The current draw is typically (I am pretty sure) well below one amp. With Andruil lights this might be 20/150. I try to have all of my lights come on at the lowest level, then only use more when required. Actually, I wish my new A54U had a much lower low setting.
But admittedly, this is for indoor use. I don’t do any trail running or cycling at those low levels. If requiring higher levels, I would definitely opt for a high current cell over just picking one with a high advertised capacity.
3 Thanks
Thank you! And @Mandrake50 too.
I learned something and that makes it a good day.
3 Thanks
Be careful with the USB charging on this light… Bad News…
Got my light in yesterday with a Molicel P50B battery.
Charged the battery up and used the light for 5 minutes or so. Very bright on high and as expected gets hot!
I decided to use the USB port to charge the Battery back up. The light switch was Red while charging and turned Blue when done. The light stayed on the USB charger for a little while after turning Blue.
Turned the light on for a little bit and noticed it flickered some after going to turbo for a few seconds. Noticed a dark spot on one of the LED’s.
I checked the voltage of the battery and it was at 4.65 volts. The USB charging had over-charged the battery… I then tested the USB charging and in fact it continued to charge the battery after the switch indicated blue and fully charged.
No word back from Neals gadgets yet…
.
6 Thanks
Hello @Mooch, I am having a problem with I believe the charging system of this flashlight that might have killed 2 of my AM04 batteries.
How I charged the batteries during output testing of this light:
- Quick charge adapter 3.0, 5v 3A/9v 2A/12v 1.5A (I have Opus but mostly charge using flashlight on board charging - never had any problem like this before.).
- USB-C to the A54U to charge the battery
Yesterday I noticed one AM04 suddenly dead, as in 0.2V with DMM and not charging anymore. I thought it was just the battery. This morning the second AM04 charged with the flashlight is dead. I just now noticed the voltage of the other batteries that have been charged with this same light is: 4.63v in Molicel P45B and 4.65v in Nightwatch 45XP. Numbers that I have never seen before.
EDIT: Just now see post above. I was going to ask Mooch why, but it’s obvious it is the ~!@#$ light. Thank goodness the LED is ok.
3 Thanks
Whoa…yea, definitely NEVER use that light to charge! That’s getting to a dangerously high voltage and, as you’ve seen, can easily damage or destroy a cell.
This is absolutely inexcusable. Charging is the most importantly safety issue and can’t be cheaped out on or insufficiently tested.
Do we have any idea what charging controller they’re using? Or have there been any hi-res teardown photos of the circuit board?
You mentioned the emitters were okay in your last post but one is black? That can’t be good.
1 Thank
I don’t believe anyone has done a teardown of this light.
Two of us discovered this problem simultaneously, forum member “Lips” and me. His light has one dead LED but mine is ok, no dead LED. Only casualty for me is two of my AM04’s are kaput.
I had some pre-conceived opinion of AM04 and was all ready to blame it for quality problem with self-destruction. 
Relieved to identify the source of the issue.
4 Thanks
Thanks for telling us this! I seldom use USB charging on any light. Too many good chargers that I trust. But I have been known to do a quick top off using USB. I will definitely avoid that with the A54U , Especially since @cannga has reported the same thing.
4 Thanks
I had a defective light ordered from Nealsgadgets and based on that experience, the best (and only?) way to get money back IMHO is to file dispute with paypal. I explained to paypal there was no response from seller, and sent picture and explained why the item was defective. Full refund was quick.
3 Thanks
Too bad because I like the light. It is very bright and a fun toy. My solution (though I agree a light that kills cells on charging IS broken) will be to simply not charge cells in the light.
3 Thanks