Discharging IMR cells.

[SmurfTacular]

For my recent triple P7 mag mod, I used four 26500 IMR cells. It worked out great.

But when the light finally died, I used my multimeter to check the voltage of each IMR cell, and it read .8 volts each.

Afterwards, I tried to charge the cells with a home made charger consisting of four C battery holders wired in parallel to a lithium ion charger, after hours of charging, it wouldn't reach a charge of higher than 3.58 volts :shakehead. They still hold that charge, but not for long at all.

I knew at this point I fried $60 worth of batteries. Whats new, right?

My question is, how can a prevent these cells from discharging farther than 2.5 volts?

 

[TooManyGizmos]

But when the light finally died, the voltage of each IMR cell, read .8 volts each.

My question is, how can a prevent these cells from discharging farther than 2.5 volts?

Well .... to answer your question .......

Simply don't operate your light untill it dies.

Why did you have to run it that long continually ?

Keep an alternate set of batteries to swap to at mid point thru the burn.

~

 

[SmurfTacular]

How are you suppose to know when the batteries are at 2.5 volts? Am I suppose to carry a multimeter around with me when camping?

 

[^^Nova^^]

The light should get dimmer. The 26500 cells would not be able to supply much current at less than 3v.

Cheers,
Nova

 

[TooManyGizmos]

~
Ha ... I'm sorry ...... I just noticed your Sig line :

Infamously know for destroying batteries.
​

You must have been testing again

But yes ....... carrying you're DMM for testing would be a good idea.

Beyond that ..... experience and time will will give you a better feel for it .

~

 

[TooManyGizmos]

SMURF ,

I'm sorry you damaged your cells .

But I'm sure you'll agree ... that's a hard lesson you will not soon forget .(or repeat)

~

 

[SmurfTacular]

To be honest, I did not notice the drop in luminosity at all. When I was done with it, I put it away, and the next night it didn't turn on. When I got home and measured it, thats when I discovered its fatal voltage level.

 

[mrartillery]

Am I suppose to carry a multimeter around with me when camping?

Perhaps, or switch to protected batteries.

 

[Linger]

http://www.hobbyking.com/hobbyking/...roduct=6589&Product_Name=Battery_Monitor_2-6S

for $2.50 you could get a tiny battery monitor. The linked one needs at least 2s to opperate, and ranges up to 6s. For a field tester hack: plug a female JST lead into the tester, place two (or more) cells end-to-end, and put the leads on each end to complete the circuit through the battery monitor.
This little tester is damn simple and convenient, way more so than $60 for a new set of cells.

careful: get the tester and you may start migrating towards the high discharge hobby packs

 

[Justin Case]

How are you suppose to know when the batteries are at 2.5 volts? Am I suppose to carry a multimeter around with me when camping?

Son, you gotta be smarter than the thing you're working on.

There are many ways to avoid overdischarging cells. One way is to measure the tail current draw. Calculate an estimated run time. Then don't exceed say 50% of that (to give you a 2X safety margin) before changing out the cells for fresh ones.

What kind of driver, if any, are you using?

 

[SmurfTacular]

http://www.hobbyking.com/hobbyking/...roduct=6589&Product_Name=Battery_Monitor_2-6S

for $2.50 you could get a tiny battery monitor. The linked one needs at least 2s to opperate, and ranges up to 6s. For a field tester hack: plug a female JST lead into the tester, place two (or more) cells end-to-end, and put the leads on each end to complete the circuit through the battery monitor.
This little tester is damn simple and convenient, way more so than $60 for a new set of cells.

careful: get the tester and you may start migrating towards the high discharge hobby packs

Thats an awesome little invention. But how would a get a + and - lead to the tailcap?

Son, you gotta be smarter than the thing you're working on.

There are many ways to avoid overdischarging cells. One way is to measure the tail current draw. Calculate an estimated run time. Then don't exceed say 50% of that (to give you a 2X safety margin) before changing out the cells for fresh ones.

What kind of driver, if any, are you using?

Im using Der Wichtels A/B driver.

 

[TooManyGizmos]

Smurf ,

Is this the light you put up for sale this afternoon ?

I guess this was a bad experience for you .

~

 

[SmurfTacular]

It is, with a new Maglite host, and brand new never used 26500 cells.

 

[Justin Case]

Did you wire the 3 LEDs in series or parallel? In series, you could have a total Vf of about 10V-11V, while Vbatt could be about 14.8V. When the cells drain down below ~3V per cell, the driver probably will fall out of regulation. Definitely at 2.5V per cell, you'd be out of regulation and should notice a rapid decrease in output. If you didn't notice any decrease in output, my guess is that you didn't wire the LEDs in series.

If you wired the LEDs in parallel, the total Vf is probably around 3.5V. With 4 IMR26500s in series, Vbatt is essentially going to be well about Vload "forever" and the der Wichtel buck driver is going to keep on drawing from the cells and you can basically drain the cells down to nothing and kill them.

 

[Justin Case]

Duplicate post deleted

 

[TooManyGizmos]

~

Well .... it must not be the exact same light ................

Cause that sales thread says ....... "Brand new, never used."

must not be the one you used to deplete those cells ?

(but a little confusing)

 

[SmurfTacular]

Did you wire the 3 LEDs in series or parallel? In series, you could have a total Vf of about 10V-11V, while Vbatt could be about 14.8V. When the cells drain down below ~3V per cell, the driver probably will fall out of regulation. Definitely at 2.5V per cell, you'd be out of regulation and should notice a rapid decrease in output. If you didn't notice any decrease in output, my guess is that you didn't wire the LEDs in series.

If you wired the LEDs in parallel, the total Vf is probably around 3.5V. With 4 IMR26500s in series, Vbatt is essentially going to be well about Vload "forever" and the der Wichtel buck driver is going to keep on drawing from the cells and you can basically drain the cells down to nothing and kill them.

For sure in series. If they where parallel, wouldn't each LED get 10 volts each?

~

Well .... it must not be the exact same light ................

Cause that sales thread says ....... "Brand new, never used."

must not be the one you used to deplete those cells ?

(but a little confusing)

Ya dude, I would consider the Maglite to still be new, although it is debatable. Brand new digital camo host, brand new fresh un-used 26500 cells. The only thing that would be considered "used" would be the switch, heatsink, driver, and LED's. When I removed everything, I also rewired everything, so new wire too.

BTW, I now renamed my sales thread title to avoid accusations.

 

[TooManyGizmos]

~

Well ..... Smurf ..................................... it IS what it IS


~

 

[Linger]

Thats an awesome little invention. But how would a get a + and - lead to the tailcap?

Easiest way to check voltage is to unscrew the tailcap and take out the cells.
The pack monitor is best installed in a lantern-type unit as it needs a disconnect switch (lest it provide constant read-outs and drain the cells down). Stuffing said components into a m@ghousing is beyond the scope of this post.

 

[Justin Case]

For sure in series. If they where parallel, wouldn't each LED get 10 volts each?

No, the der Wichtel constant current buck driver would deliver the appropriate Vout to drive the parallel 3xP7. What would happen is that each P7 would get 2.8A/3~900mA drive, assuming identical LEDs.

If you ran your der Wichtel mod with 3 P7s in series, all I can say is that you either totally ignored the warning signs of severely fading output or else you left the light on. Either way, you overdischarged your cells.

If we assume that each P7 has a Vf of 3.5V at 2.8A drive, then total Vf is about 10.5V. Your four 26500 cells nominally deliver 4*3.7V=14.8V.

If we make the simplifying assumption that the der Wichtel buck driver can run in full regulation as long as Vbatt>Vload, then when Vbatt finally falls to 10.5V, the light falls out of regulation and goes into direct drive. That translates to 2.625V per cell.

I can believe that it is quite possible to miss the fading output signs at this point, since even at say 10.0V in direct drive, each LED is getting about 3.3V, which still can give you 400 emitter lumens per P7, based on the measured data from jtr1962.

But overdischarging the cells to 9.0V total (2.25V per cell) puts you down in the 140 lumen range per P7 and that should be clearly noticeable vs 900 lumens per P7.

When you are down to 8.1V total Vbatt (2.025V per cell), you are delivering only about 2.7V to each P7. I've found that this forward voltage is approx where white power LEDs typically turn on and start producing noticeable output (moonglow can sometimes occur at a lower Vf such as 2.4V). At ~2V per cell, you've already seriously overdischarged the IMR26500s, but that's a far cry from driving them below 1.0V. At 1.0V per cell, the light would have long turned off. That fact alone suggests to me that you left the light on, killing your cells.

The simplest method to avoid overdischarge is to pay attention and stop using the light when output drops noticeably. Another method is to pay attention and use the light for say 50% of the calculated run time before shutting down and swapping to fresh cells, giving yourself a 2X safety margin. The common thread here is to pay attention.