Practical results do not lie. I also noted that I do not know if/how this might apply to a MS battery. I will be doing an experiment to check.

I also don't think that the phone is what makes the difference, but the battery. I have used this method on several phones over the span of five years. My only reason for bringing it up was to address the issue of a red light on the MS after a short period of use.

I've just received my a replacement Geoman Gear battery for the defective new battery when I bought a new MJ808 light set two weeks ago. This time the new battery did not arrived DOA and I was able to charge it for an hour when the green light came on the charger. I cannot say if this is the same pack that will be the replacement pack for the thousand of recall packs but I believe it will be. Here are some data of the battery pack running with an older (1.5 year old) MS900 light.

The battery stopped charging at 8.46 volts.
The new battery OCV was 8.26 volt after disconnecting from the charer


The test is done with the light on HI mode and volt was check under load with the light on Hi.
after 15min 7.95v indicator light was green
after 30min 7.52v indicator light was green
after 60min 7.24v indicator light was green
after 90min 6.99v indicator light was green
after 105min 6.91v indicator light turned red
after 120min 6.83v indicator light was red
after 150min 6.70v indicator light was red
after 180min 6.47v indicator light was red

I stop the test at 3hrs. The light was still on and the battery have not shut down by itself. I stop since it was only rated for the three hours and no sense in me running it down.

In comparison with the old stlye battery, the indicator light on the light head of the old lamp with the old battery does not turn red until the last 10 minutes when the old battery were new. It seem these new battery run on a lower voltage for a longer duration but has the same Amp hr capacity. The newer light head has three colors for the indicator: Green, Orange and Red. The indicator will remain in the Orange during the time when the old light head will be showing red. I have to agree that the indicator light will have to be ignore if you are using the old light head unit with the new battery.

Thanks for the results. Did the light flash when you reached the end of the test? If the replacement battery is the same you bought, it looks like we'll just have to time the use of the light.

I assume the voltage at 105 min was 6.91v, not 9.91v.

Yikes, thanks. Yeah that was a typo error. I just edit that. I thought I double checked all that before submitting. No wonder why my checking account never balance each each month.

AFAIK the original lightheads do not flash. They just shut off.

In other words, the indicator light on my light head has just become pretty much worthless. Bummer, but not a deal breaker.

It's actually an 0.3 Ah increase in capacity, not 3 Ah (as the pack is 2s2p.)

Put in more useful terms, if the capacity is 7% higher, you should get 7% longer runtimes -- so if you got 3 hours before, you should get around 3 hours and 12 minutes now. 4h before? Then 4h 17m now.

(Basically, the improvement is small enough that you probably won't even notice.)

In any event, unless your old battery is quite new, it's probably lost some capacity just due to it's age, so your new battery will probably give you more than 7% more runtime. And this assumes that all the batteries involved are exactly what they're rated at -- which is rare, usually there's some variations.

I thought it flashing before it cutoff, but you're probably right. It's been six months since I even hooked the light head to the battery.

Good question. I don't recall myself. Perhaps I should tested it to the last seconds. I do have one old pack I am running it down before sending that out for the recall replacement. I'll try to plug my old MS light head into that pack when it's almost dead.

The gain from the much improve housing is worth the lost of the low battery indicator. The wires are more HD and the mounting design makes more sense than the old bag.

I cannot say for sure if your lighthead and the battery you are getting will have the indicator issue. It seem to vary. Mine took 2 hrs while Stringbreaker was much faster and a member at MTB forum was also less than a hour before the indicator tripped. Hard to say but indication points to it being useless at this point.

Yeah I totally botched that post, I meant to say it was a three "watt hour" increase.

I don't get that math either, amp hours and watt hours are apples and oranges.

Amp hours rating only accounts for current during discharge. Watt hours, at least -unless you want educate me in another way - actually reflects "power" coming out of a light. (current x voltage)

So I compute the rating of a battery in amp hours - only in conjunction with known current flowing during light operation, and arrive at the theoretical discharge capacity in watt-hours. if my previous testing is repeated with a newer battery, I will expect to see 10-11 watts delivered to the light head for about .3 of an hours - like my previous botched post says - 3 "watt hours." (10-11 watts x .3) - ~3-3.1

However, i do continue to be full of crap - so i could have all this wrong, but basically my assumptions still should bear out accurately in reality.

As I've said, the only "real way" is to measure is recording the current and voltage over the life of a battery cycle (time).

The bottom line, for the batteries I've been using, I'll expect about 15 minutes on high, or maybe an hour longer runt time on low. (1 hour at 3watt consumption) in other words - 3 watt hours.

But i digress.

But even that's not quite right. Assuming an average voltage of 3.9 volts (the midpoint of 4.2 volts (full) and 3.6 volts (time to stop discharging) -- simplistic, but close), 4 cells * 3.9 volts * 150 mAh (difference) = 2.34 watt*hours.

Ultimately, it's hard to give an accurate "watt hour" rating on a battery because the voltage decreases as it's discharged -- so the true watt hour rating depends on the discharge curve of the battery, and it even depends on how fast you discharge it -- the faster you discharge it, the more energy wasted to it's internal resistance and so the lower the watt hour rating.

Yes, watt hours are the best value to look at. But as long as you're looking at batteries with the same chemistries (and assume the rest is unchanged, such as discharge curves and internal resistances), you can generally just compare Ah ratings rather than working out watt hours.

Either way, if the Ah rating goes up 7% and everything else is the same, you can generally just assume 7% longer run times with pretty good accuracy. No need to get more complicated than that. 7% is pretty small -- most people probably wouldn't notice. A much bigger deal will be that the new battery is fresh (they do lose capacity over time), won't become unbalanced over time (I hope it has something to keep the cells balanced) and hopefully won't burn your house down.

Actually, that's why I measured the discharge (in wattage) in real time. And that's why I know that the light heads I used during measurement really do consume ~2.5 low, ~5-med, ~10-high watts during testing. Discharge curve means little in a regulated circuit -assuming it regulates.

Whether its 1.25A at 8V or 1.66A at 6V - it's all ten watts to me. No you can measure watt-hours - accurately - as i already have.

Same here.

Well, not hard, just expensive. High end voltmeters with data logging will tell you the watt hours. You just need a $350 voltmeter.

Honestly, two cheap voltmeters and taking volt/amp readings every 5 minutes would do the job adequately.

I used my recall battery yesterday after fully charging it with the original battery charger the night before, and the red light came on way too soon. I rode to work with it on flash (about 30 minutes) and then another 5 minutes to the GF's house (another 5) without problems. Coming home, I rode for a few minutes to stop and put on the kneewarmers and the red light came on when I turned the light back on. I unplugged it, ran it throught a few cycles and the green light would come on briefly and then change to red as I attempted to use the light while I rode home until it stayed green while on high. That lasted most of the remaining 20 minutes of my ride, and I recharged it when I got home. I hope I get longer run times than that after the battery breaks in!

Sorry man, but reading this just confused the hell out of me (not that confusing me is hard to do).

Has anyone compared the actual run time...ignoring the darned indicator lights?

See my post #27.

My battery is not a replacement of a recall but rather a new Geoman style battery included in the new lightset I purchase. I believe it is the same battery as the recall replacement battery. The post show runtime and indicator light.

Maybe I missed it but I didn't see in your post anywhere that indicated that the light actually shut off.

My MagicShine green/red indicator is pretty much random. I can have it come on with 2 hours of runtime left (out of 3) or with 5 minutes left. I just ignore it. It's possible that the protection circuitry in the new battery interacts differently with the light indicator than the old one did. What matters is how long the actual light runs for.

Thanks, I saw that, but I'd like to see one with an actual replacement battery and original head....and just a simple runtime comparison. Despite my electrical engineering degree, I sometimes just like to know the bottom line in my old age.

Actually -that's pretty much what I do, I workout while watching the battery run down

yes- I am indeed the grand pooh-bah of all things 18650ish.........

I've been testing Li Ion batteries for 18 months. I have tested two and four cell packs from four different sources.

The reason there is so much confusion lie with the fact that semi-conductors and circuits vary in accuracy and respond differently when operated at different temperatures and combination.

First of all - most of you did not know that not all battery chargers are equal. I have seen chargers produce as little as 8.28 and as much as 8.45 volts. This variance alone, could account for .5Ah difference in charge to a given battery pack. (this means a battery might run on high 2:30 off one charger and 2:40-45 off another.)

Second of all - not all light heads regulate equally. I've got a first generation MS from dealX that draws almost 200mA more than Geoman's light heads. And the new light head has yet another difference in its current consumption slope. (it changes with temp)

Obviously the different light heads will draw a battery down at different run times AND different voltages.
(this means warning lights/voltages couldl change with every attempt to measure them)

One last very important issue. To do any of the tests I perform - one would have to create testing cables that allow you to connect testing devices to a circuit without causing damage to the light and battery. This is the only way to verify that you have valid connections producing valid measurements. It is impossible to measure properly without using the correct connectors and verifying they are fully contacting the circuit.

So there you have it - different chargers and different light head make posts about "battery packs" by themselves pretty meaningless.

If anyone has a specific question PM me, but be brief and to the point.

Thanks for shedding more light on this, Richard. (No pun intended!)

I purchased a couple of the MagicShine heads less batteries (being an R/C flyer and ebike hobbyist I just don't need any more batteries!) and I also ordered a light kit with the CREE XM-LT60 in the lighthead from DX. Are the battery packs from Geoman for these lights the same as the ones from DX (Deal Extreme)? I'll compare current draws and post some discharge curves when I receive the stuff in a month or two.

Sorry for the confusion Chip, I think I resemble that remark myself somedays. The light never shut off on it's own. I shut the light off several times, either because I didn't need it anymore or because I was cycling through the settings to see if that would change the red light back to green and wanted to avoid completely discharging the battery. Seeing the red light come on after less than an hour of use made me nervous, and I remember reading that the battery be damaged by using the light for too long after the red indicator light came on. Do I need to worry about damaging the battery by running it until the battery voltage is so low that it shuts off? Or can I just run it until it doesn't work anymore and quit worrying about the green indicator turning red? Either way I'm hoping for a longer runtime than 35 minutes before the warning light comes on.

Running the light until either the light or the low voltage circuit on the Protection board cutoff will not be good for the battery, However that does not mean that you cannot run your light while that indicator is showing red. The indicator supposed to be calibrated for the older packs and even at that, many MS owners have experience with those lights as being very inaccurate. The new pack might have some variation in comparison with the old pack and that alone might no longer in sync with those indicator light. One thing I noticed is the voltage rating written on my pack as 7.2volt instead of 7.4 volts. Does that mean anything? Perhaps just lower nominal voltage but with same capacity? PCB higher internal resistant? Different battery chemistry? Or does it really matter? Dunno.

Instead of relying on that indicator light, you better off relying on the runtime you used on the pack. The pack is rated for 3hrs. So running the light while it is red for the first hour or two is probably not harmful. During the last hour, I think it will be a good idea to get it recharge ASAP well before getting to close to the full 3hrs usage.

Edit: I just looked at Geoman site and it does list the Panasonic battery at nominal 3.6 volts. So yeah, that may trigger the indicator light to red sooner than what it used to be assuming that the older style pack uses nominal 3.7volts cell.

I ran it for under 2 hours last night, and the red light came on again after about 45 minutes of use. And then the green light came back on later, and it switched back and forth for the rest of the ride. It was green this morning (uncharged from last night), I'm going to email geoman and see what they have to say about this.