Hi Niconical!
I'm going to quote sections of your reply to respond to reach individually as best I can..
So. Protected. The charger won't overcharge it, and when it's in a light and it drops to a certain voltage it will cut off so as to not damage itself. It will however be at 3.7v or even more when fresh from a charger. I won't have access to the lower settings on a multilevel light.
I should clarify this further.
A protected cell prevents the voltage of the cell from being pushed dangerously out of bounds, but does not protect it from accelerated wear that can occur with minor overcharging or over-discharging. Ideally speaking, charging should be terminated at or below 4.20V, charging as high as 4.30V is usually *safe* on a new healthy cell, but it will DRAMATICALLY reduce the number of charge/discharge cycles the cell will last. Some cheap chargers are known to charge to above 4.20V if the cells are left on the charger too long. Even after the light goes "green" many of these chargers continue to trickle charge the cell, which is bad for li-ion. Ideally speaking, discharging should be terminated at ~3.0-3.5V under a load, depending on the severity of the load in question. The built in protection will allow the cell to be dragged down to ~2.5-2.75V in most cases, which, in a slow drain device, or a device set on "low," can result in open circuit voltages down around or below 3.0V, which isn't all that great for the cell. (but isn't really dangerous).
A bare unprotected cell, could be dragged down to well below 1V with the regulation circuitry found in many lights, this would almost certainly spell the death of the cell. If not on that cycle, within a few cycles. Trying to recharge a cell that has been repeatedly over-discharged can cause explosion. The protection is there to keep things safer, but not necessarily "perfect."
As far as having access to multi-levels on a multi-level light, it depends on the light itself. The P1D and P2D will loose those lower settings as the driver basically goes into "direct drive" mode when the voltage of the cell is higher than the voltage needed to accomplish the desired level of operation. There are a number of multi-level lights on the market that will operate from 0.9-4.2V and still run in low modes properly.
Regulated. It will operate at 3v, apart from a few milliseconds of first use while the regulation kicks in. However, if you let that voltage drop too much, it will damage or ruin the battery. I will have access to the lower settings on a multilevel light.
In this case, on P1D/P2Ds, the reduced voltage of the cell should give access to all modes, though the lowest settings may behave slightly different than they do on a CR123 as these 3.0V regulated cells still tend to run higher voltage than their primary brothers. Under a load, these often still deliver 3.2V give or take, while CR123s generally deliver ~2.5-2.75V under a load.
In your reply above you linked to some batteries that are both regulated and protected. You also mentioned further on that...
"the possibility of having a cell that you simultaneously don't have to worry about over-discharging, while at the same time, not have anything else go wrong, is an exercise in futility as you kind of have to decide to have one or the other."
I have to be honest and say that I still don't really understand that. If it's regulated I can't damage the light, and if it's protected I can't damage the battery, so what else is there to go wrong? I will freely admit though that there is no doubt a lot more to it, and I will hopefully learn more about the why and how in the future.
What I meant by this, is that, those extra components within the regulated/protected cell, are more possible failure points. So while they are supposed to "protect" you from anything "going wrong" if they themselves "go wrong" then something has gone wrong. Point being that the cell with the most complexity has the most opportunity to fail in some way or another.
There is also another aspect that I don't understand. I have read numerous references about single CR123A lights being a lot brighter when used with a 3.7v rechargeable instead of a 3v primary. My understanding is that the LED needs 3.6v (correct me if I'm wrong) to function. The circuit in a single battery light therefore boosts the 3v up to 3.6v to get the LED to work. Assuming that's true, why then is there such a big difference when using a 3.7v battery? If a light using a primary is running the LED at 3.6v, how much difference can an extra 0.1v make, and why do I see these claims of them being much brighter with rechargables?
Every LED is different. to say they they need 3.6V to run is a gross generalization. For years around here we have talked about the "LED Lottery," we used to call it the "Luxeon Lottery" when Luxeon was the only big game in town for high power LEDs.
What I mean by lottery here, is that not every LED requires the same voltage. If you had 10 Q5 emitters in front of you, and you wanted to run 700mA across each of them in a different device, every one of them would require a different voltage to achieve this current.
Regulated lights, like the P1D, do not regulate based on voltage at all, they shoot for a particular forward current across the diode. They automatically, instantly, ramp voltage (up or down, depending on the light and power source available), to whatever voltage is needed to maintain that set current. As the LED warms up and cools down that voltage requirement changes some. As the LED ages, it changes some, but the driver is always shooting to meet a certain current goal, and just modifies voltage to achieve this goal.
The voltage required to achieve a particular common forward current, is called the "Vf" of the LED. LEDs within certain ranges of Vf are grouped together and sold as a particular part number. This is known as the "binning" of the LEDs. You'll find long part numbers used to describe an LED, one section of that part number describes it's color tint, one part describes it's Vf, another part might describe what family of LEDs it belongs to, while another part describes it's luminous intensity range when driven to specified forward current.
Anyways, the point being, that if you buy a bunch of P1Ds all at different times, you are likely to have a wide variety of LED Vfs in the batch. If one of them requires 3.5V to achieve 700mA (or whatever the "high" setting on a P1D is), and another requires 3.7V, then the amount of "overdrive" would be much different for these two lights when driven by the same 3.7V cell.
Keep in mind, that a 3.7V li-ion cell, actually comes fresh off the charger at 4.20V, under the load of direct driving a LED, the voltage of the cell will sag and find equilibrium with the LED at some forward current. For the 3.7V Vf LED, that might happen at 3.85V with a overdriven forward current of 1100mA. For the 3.5V Vf LED, that might happen at 3.75V with a forward current of 1800mA. In the case of the 3.7V Vf LED, the overdrive would probably be pretty tolerable provided it wasn't run for excessive lengths of time, while the 3.5V Vf LED would probably suffer more substantial damage from heat even in the short term. Knowing which Vf LED you have is only possibly by running your own tests on the LED with a good bench power supply and a good amp meter.
So, back to the original question, how much could that 0.1V difference really have on the LED?
Sometimes, a LOT. Diodes are not like regular resistive loads. Increasing the voltage even SLIGHTly on a semiconductor can have dramatic effects on current flow. For most popular high current LEDs like the Crees, 2.5V is not enough to get hardly any electrons to flow across the diode, while increasing to 3V will start to get a few flowing, increasing to 3.5V will get a LOT flowing, and 4V+ will often allow so much current to flow (several amps) that it will fry the LED. The behavior of semi-conductors as it is related to voltage input is measured on a logarithmic scale, not linear.
Moving on to my situation. For reasons of who will use the lights, where and how, it would be good for me to have regulated 3v batteries for the P2D, and unregulated but protected batteries for the Romisen. This is because I don't want to lose the lower settings on the Fenix, and the Romisen only has 1 setting anyway so nothing to lose. Based on what I wrote above about the difference between a 3v primary and a 3.7v rechargeable, I assume the Romisen will therefore be brighter, but as mentioned, I still don't understand why.
Although I've written a lot here (sorry about that BTW, I'm sure you all have better things to do, but if you're reading it in daylight hours, at least I haven't wasted good flashlighting time), I only really have 3 main questions, and here they are...
I'm not totally familiar with the Romensen, but keep in mind that just because the P1D/P2D tend to be a bit brighter on a 3.7V cell, does not mean this will hold true for all lights. Since some can buck or boost voltage as necessary to accommodate a wider range of voltage input.
1. P2D with regulated batteries, not protected. Good for me, 3v, keeping my lower settings, but be careful not to over discharge. However, other than me needing to be careful not to over discharge the battery, what is the downside? I know there must be one, I'm just not really sure what it is.
actually, as it stand currently, I am not aware of any voltage regulated 3.0V cell that is also not protected. I don't think such a cell is available. For any 3.7V cell, you want to have a PCB (protection circuit) onboard. The regulated 3.0V cells are just 3.7V cells under the skin, so having the protection is the most important part. The reason I lean towards LiFeP04 cells is that they remain safe regardless of how you abuse them. While they can be abused to the point of not recovering (no longer taking a charge), they will not become unsafe when this happens.
2. Romisen with protected batteries, not regulated. Good for me, don't have to worry about over discharging, and there are no lower settings to lose anyway. Again, same question. What's the downside?
Provided the rominsen support the 3.7V cells, then the protected version of these cells is the best option as you will get the most available capacity. The protection keeps the cell in check. The protection can fail, but I'd say it's worth the better capacity in this case.
but can I not just get batteries that are LiFeP04, AND regulated, AND protected, even if they do have a low capacity?
I have not seen any such cell. There isn't really much good reason for them to put a regulator on a cell that is already naturally ~3V operation, it would just be a wasteful component. Adding protection to a LiFeP04 cell would be similar to putting it on a NIMH cell, it's just not needed for safety purposes as the cell will not become volatile if it is overcharged or over-discharged.
Sorry, 1 more edit: In the other thread you replied to I looked at the LiFePO4 you linked to. They state "Discharge cut-off voltage: 2.2V". They also state "Please never overdischarge battery below 2.2V/cell".
What am I missing? "cut off voltage at 2.2V". How can you discharge it lower than 2.2V if it cuts off at 2.2V? I know I'm missing something, but I just don't know what.
The LiFeP04 cells do not do this themselves, it's just a "warning" to the user that discharging below 2.2V can cause advanced/premature/accelerated wear on the cell. Discharging to below 2.2V and holding the voltage down with a slow drain for an extended period of time would probably "kill" the cell, making it impossible to take a useful charge again. But the cell would still be safe.
. So the best would be the third option with the LiFeP04?
