Olight S1R battery capacity vs resting voltage, ORB-163C05 IMR RCR123A

i am trying to find out if there is any data available on the capacity of high discharge IMR cells, specifically the Olight ORB-163C05 IMR RCR123A. In another thread I was pointed to some work that HKJ did here: http://lygte-info.dk/info/BatteryChargePercent UK.html

This was very helpful, and I can see the performance of a couple different types of cells. However, one cell type has ~10% capacity remaining at 3.7V, while others have ~50% capacity remaining at 3.7V. I am not very fluent in rechargeable lithium, so I am not sure if my IMR cell is similar to any of these types.

Hoping someone can point me to data on IMR cells.

thanks, Matt

sbslider said:

i am trying to find out if there is any data available on the capacity of high discharge IMR cells, specifically the Olight ORB-163C05 IMR RCR123A. In another thread I was pointed to some work that HKJ did here: http://lygte-info.dk/info/BatteryChargePercent UK.html

This was very helpful, and I can see the performance of a couple different types of cells. However, one cell type has ~10% capacity remaining at 3.7V, while others have ~50% capacity remaining at 3.7V. I am not very fluent in rechargeable lithium, so I am not sure if my IMR cell is similar to any of these types.

Hoping someone can point me to data on IMR cells.

thanks, Matt

Click to expand...

Since all of these smaller li-ion cells like the 16340s are generally sourced from China, they can be all over the map quality wise.

Capacity varies with the current load of any particular output, so it's not linear.

Right now, all of the cool kids are running Aspire high drain 16340s and 18350s. They have good capacity for the form factors and they can handle decent/high loads.

AWT would be another, but they can be sometimes hard to find.

My past favorite is AW IMRs, at about 500mAh at 1A drain, but the factory recently got flooded and well...next man up.

Figure that you'll get ~500mAh out of most IMR 16340s at 1A-2A. The higher capacity ICR chemistry is between 600mAh and 800mAh, for the older fair, but these newer generation cells are pretty wicked.

Chris

Thanks for the inputs. I am not really in the market for a new or different battery, just trying to find out about the behavior of the battery I have. This is my first and only rechargeable lithium battery.

The S1R is designed to charge the battery inside the light with a magnetically attached charger. It is a very simple arrangement, red is charging, green is done. The tail of the light has exposed electrical contacts which allow me to measure the battery voltage without removing it. I would like to be able to measure the voltage and estimate the remaining capacity. In the link above to HKJ's site, the difference in remaining capacity when a battery is delivering 1 or 3A was rather small a few percent. I am just hoping to be able to see my battery is at say, 3.7V, and know that there is 10%, or 30%, or 50% left.

Perhaps over the Christmas break I will run my own testing, but was hoping that there is general info regarding IMR style batteries and capacity remaining versus voltage.

But if/when I decide to branch out into some new cells and a new charger, it is nice to have a recommendation.

thanks, Matt

3.6v/3.7v is the nominal voltage for these cells, but they can be anywhere from 40%-60% at that level and if you're driving a light at 2A, that remaining 40%-60% is going to disappear much quicker than if you're running a light in moonlight mode.

Most of us don't think in terms of where a cell is at on the voltage scale, but how long a runtime will we have if the light is used at 3.85v and on a 200mA low setting.

CottonPickers' belief

4.20v 100%

4.10v 87%

4.00v 75%

3.90v 55%

3.80v 30%

3.50v Empty

I don't believe that 3.50v is empty, but as the voltage drops towards 3.00v-3.30v, you're getting to the level where you're not going to run higher drain loads, even though you might have a couple/few hundred mAh of capacity left. Most LEDs we deal with need ~3.00v to be lit up, so that's another problem.

The 'vampire light' guys can take an almost dead alkaleak and suck that last few electrons out of it, but they're not putting out tons of light.

Most of us consider 3.6v-3.7v the point where recharging is needed. Remember, that HKJ is/was testing down to 2.50v and/or 2.85 (I now think), but we try not to go that low for the cell's sake.

Chris

What I believe I learned from HKJ's artilcle I linked to in the OP is that the remaining capacity for a cell vs resting voltage depends on the type of cell it is. This table may explain it better than I have been able to do.

Summary of all cells he tested in this article at 1A



Summary of all cells he tested in this article at 3A



Take for example 3.7V. For both the 1A and 3A tables, note how he measured a different remaining capacity for the cells at this voltage. There is a 40% difference in capacity measured at 3.7V for both 1 anf 3A. I don't think one can take a generic capacity versus voltage table and apply it to any cell. The Panasonic cells have siginficant capacity remaining at 3.7V, while the Sanyo cell does not. It is not the manufacturer that causes this variation, but the cell type. The Panasonic cells are NCR and CGR, while the Sanyo is not specified in the article. None of these are IMR cells, which is what I am interested in finding out about.

I doubt IMR cells exist anymore, but you need to find what chemistry the cell uses and check if one of my tests covers it.
The easiest way to get an idea of similar chemistry is to compare discharge and charge curves on my website, that will only work if I have tested your cell. If the low current discharge and the charge have similar curves between the cells, you can use that voltage table.

Maybe I ought to dig that old test script up again and run a few newer cells.

Cells have gotten better and we now have hybrid chemistries, so one needs to either focus on a specific cell, as in HKJ's graphs, or just speak loosely in general terms, which is about the best most of us can hope for.

The Pannie CH cell in the graph 'was/is' a typical low capacity, high drain cell from about the time that I started here. Back then, you could have higher current handling (IMR-ish), or your could have higher capacity (ICR, hybrids), but you really couldn't have both.

Now, we have the 3000mAh Sony VTC6, the 3500mAh Sanyo-Panasonic NCR-GA and a plethora of others, that give us both high current handling and higher capacity.

The smaller cells are still nominally 3.6v/3.7v, so the general rule of thumb is still a ~50% state of charge, plus or minus a few.

You're not into this stuff, so buying an analyzing charger and doing your own discharge tests aren't in the cards, but you can buy an Opus BT-3100/3400 for $35 if you have to know the exact amounts left in a cell at X, Y or Z voltage.

Happy holidays!

Chris

I see that what I thought was a fairly simple request is not so. I thought the cell I had was an IMR cell, due to the part number of it, but I guess that is not the case either. I am not sure how to find out what chemistry this cell is, as it is not stated on the Olight website.

Thank you for the suggestions to help answer my question, they are appreciated. If/when I do some testing or learn more, I will post it here.

thanks, Matt

Hi,

Well, the ICR (cobalt), IMR (manganese) and IFR (iron phosphate) designators are now mostly a thing of the past, with all of these aluminum, nickel and carbon components comprising 'hybrid' chemistries, that they're almost superfluous.

Yeah, you're kind of barking up a tree that can't be precisely answered. 3.6v/3.7v depending on the manufacturer's testing protocols are considered a halfway point, so that's something you really want to keep in mind. Those levels are good charging points, if you want some wiggle room out in the field.

Add to this, the non-linear drains based on varying loads found in the zillions of output levels and 'worrying' about whether 3.35v is actually 73% depleted from 100%, while out in the field, is a fool's errand, IMO.

Here's how I look at it: with lipstick sized lights, using the 16340 3.6v/3.7v cells (these are what I carry everyday), I get about 18-20 minutes of runtime on 'high/turbo' before the cell is toast. It's much easier to keep a mental note of time than it is to carry a DMV and measure the voltage, then try to figure out what's left in the tank.

Modern lights will start to step down from high to medium and then medium to low, so we have a visual cue telling us that it's time for a recharge.

Honestly, there are probably more important things to worry about, is my thought on the matter, but as HKJ shows us daily, one can do testing and get a reasonable idea of what's left at each voltage step, for any particular cell, but he's an odd bird that enjoys doing this stuff for all of our benefit.

Chris

HKJ said:

Maybe I ought to dig that old test script up again and run a few newer cells.

Click to expand...

That could prove immensely useful - especially if you discover in common use cells that have very different curves than the two (classes) you currently have.

So I ran some testing on the subject battery in my S1R. I ran measured the resting voltage at start, not quite fully charged. I ran the light for 15 minutes, and measured the battery voltage under load at the end of the 15 minutes, then turned the light off. After it rested 30 minutes to an hour or more, I recorded that voltage, and repeated. The battery is specified to run 4 hours on medium mode. That is basically what I got, I did end the test after 4 hours as the red light was on, and the battery was down at 2.8V under load, plenty low for me. My meter reads ~0.05V high compared to a calibrated meter at my work FWIW.

Subtracting said 0.05V yields results very close to HKJ's results for Panasonic NCR B chemistry - see below

Yes, thanks for pointing that out. Both the ORB and NCR are very similar in that table. The rate HKJ tested in the chart was near 1C (a bit above). I am thinking of testing my cell at near 1C when I return home from Christmas. Likely just run it straight out instead of measuring the resting voltage, we will see.