# Thread: Li-ion discharge - How far can I go?

1. ## Li-ion discharge - How far can I go?

I have a few different lights run off Li-Ion cells (3.7v).

One is an MC-E at 800mA (14.8v); one is a Triple XPG -R4 + 1 x R2 combo driven at 1000mA (14.8v).
The triple XPG has low beam also with a 750ohm resistor.

The batteries charge to ~ 4.2 - 4.3 volts and I know they discharge (supposedly) to 2.75 volts before their low voltage cut off operates.

We used the lights last night for 3 hrs. The MC-E only has high beam or off; the XP-G has low, high, off.

This morning the batteries have 3.95 v for one pack and 3.75v for the other.
They didn't seem to dim during the ride.

My questions are:-

At what voltage do they dim?
What voltage do they drop out of regulation (Is it 2.75 x 4)?
How much weaker/dimmer do they get as voltage drops?

2. I don't think you'll find any hard and fast rules as to how low you can/should discharge a typical 3.7v rechargeable lithium ion battery, but 3.0v is probably a good point to consider the cell discharged. Lithium cells drop off very quickly towards the end of their charge so the difference in run time between 3.0v and 2.75v is actually very short. So for a 4x pack that has a fully charged voltage of 14.8v I use 12.0v as the fully discharged point and set the cutoff of my Maxflex drivers at this voltage.

3. As Kwarwick says 3v per cell is a good point not to go below if you want them to last for a few years.

Remember that the driver(you don't say what one) will be measuring the voltage under load but the voltage that you measure with the light off will not be the same.

Try not to go lower than 3v per cell under load.

If you have the correct driver, LED and battery combo you shouldn't get any dimming..... full power untill low voltage protection kicks in be it from the driver or battery PCB.

4. Driver for the MCE is this one :- Recom RCD 24-70W 700 mA (not 800 as I stated).
Driver for the XPG is this one :- Taskled CC5W - 700/350/80mA.

Both are on Cutter's drivers.
I set the Taskled driver at 700/350.

I'll hook up the lights again and test them under full load.

Is there any way of guessing/knowing how much more runtime I get from 3.7 down to 3.0 volts with these units?

Just measured them both:-

MC-E has 15.60 v no load; 15.29 v under load (high beam only).

XP-G has 14.93 v no load; 14.84 v on high beam.

I measured them both at the battery where the lead goes to the lights.

5. Originally Posted by ireland57

Is there any way of guessing/knowing how much more runtime I get from 3.7 down to 3.0 volts with these units?
Yes, look at the discharge graphs in the battery data sheet. If you can't find the specific datasheets for your batteries, here are the panasonic and sanyo datasheets. Just keep in mind that both of these are very high quality batteries. If you're using chinese made cells, they probably won't do as well.

http://battery.sanyo.com/en/spec/ion/UR18650F.pdf

http://industrial.panasonic.com/www-...A4000CE216.pdf

6. I see dimming and its time to switch my torches off, 10mins later being lazy and the cell won't charge again.

7. Thanks fellas.

I did reply this am but forgot to hit the button.

8. I hooked up both these lights the other night (I need more light on rocky tracks) to one battery pack (14.8v i.e. 4 x 3.7 v Li-Ion).

So MCE (700mA) + [XPG triple + R2] 700mA high beam run in parallel off the one battery pack.

Battery pack starting voltage was 16.84 volts (no load).

The calculator says 1.5 hrs using 1400 mA, 14.8v Vf, 14.8v input voltage, 2500mAh, 90%efficiency.

I ran the lights the other night for 95 minutes actual riding time and still have 14.74 volts under full load. This suggests much more run time left.

There were minutes of low beam and one light but very limited due to me needing as much light as they could provide.

What have I got wrong here?

9. Originally Posted by ireland57
I hooked up both these lights the other night (I need more light on rocky tracks) to one battery pack (14.8v i.e. 4 x 3.7 v Li-Ion).

So MCE (700mA) + [XPG triple + R2] 700mA high beam run in parallel off the one battery pack.

Battery pack starting voltage was 16.84 volts (no load).

The calculator says 1.5 hrs using 1400 mA, 14.8v Vf, 14.8v input voltage, 2500mAh, 90%efficiency.

I ran the lights the other night for 95 minutes actual riding time and still have 14.74 volts under full load. This suggests much more run time left.

There were minutes of low beam and one light but very limited due to me needing as much light as they could provide.

What have I got wrong here?
Not sure you've got anything wrong at this point. You really need to run the lights until your battery pack gets down to around 12 volts (3 volts per cell) to really be able to determine a full reasonable run time. I don't know of any easy way to calculate percentage of remaining battery power from a particular battery voltage since Lithium Ion batteries have a non-linear voltage curve as they drain.

10. Originally Posted by ireland57
I hooked up both these lights the other night (I need more light on rocky tracks) to one battery pack (14.8v i.e. 4 x 3.7 v Li-Ion).

So MCE (700mA) + [XPG triple + R2] 700mA high beam run in parallel off the one battery pack.

Battery pack starting voltage was 16.84 volts (no load).

The calculator says 1.5 hrs using 1400 mA, 14.8v Vf, 14.8v input voltage, 2500mAh, 90%efficiency.

I ran the lights the other night for 95 minutes actual riding time and still have 14.74 volts under full load. This suggests much more run time left.

There were minutes of low beam and one light but very limited due to me needing as much light as they could provide.

What have I got wrong here?
Why do you say 14.74 under load suggests you have a lot of runtime left? If you reference that voltage into the discharge curves in the datasheets I linked above for the 1.5amp discharge curve. It comes out to about 90% discharged. The voltage of a li-ion battery is very non-linear. The drop from 3.68 v/cell to 3 v/cell is very rapid.

Also, I don't think your 1.5 hour runtime calculation is right. The triple XPG will have a Vf at 700ma of 3.2 * 3 = 9.6v. So at 700ma the LEDs are using 6.72 watts of power. If the driver is 90% efficient, then that's 7.56 watts from the battery pack.

If I assume your MCE is wired in series, then it'll have a Vf of 3.5 * 4 = 14v. So at 700ma the LEDs are using 9.8 watts. 10.8 watts from the battery with 90% driver.

That gives you a total power consumption of 18.3 watts. A 14.8V 2.5ah battery pack is 37 watt-hour. So that gives a runtime of 37 / 18.3 = 2 hours. That's a theoretical number that assumes you'll really get 2.5ah out of that battery pack. You'll never will get that much in real life and experience tells me to expect a runtime of around 1:45 minutes.

11. I didn't write clearly enough. The XPG has an XRE (R2) with it so it's Vf is 4 x 3.2 volt.

I'll go through the graphs again later today; I only partly understand them to be honest.

The next version of this light will be XPG Triple (no XRE) + MCE. I discovered the other night they make a decent light between them.

The XPG will be at 1000mA and the MCE still at 700mA (due to heat if nothing else).
A swap of drivers with other lights I made will give low beam on the MCE as well.

Originally Posted by MtbMacgyver
Why do you say 14.74 under load suggests you have a lot of runtime left? If you reference that voltage into the discharge curves in the datasheets I linked above for the 1.5amp discharge curve. It comes out to about 90% discharged. The voltage of a li-ion battery is very non-linear. The drop from 3.68 v/cell to 3 v/cell is very rapid.

Also, I don't think your 1.5 hour runtime calculation is right. The triple XPG will have a Vf at 700ma of 3.2 * 3 = 9.6v. So at 700ma the LEDs are using 6.72 watts of power. If the driver is 90% efficient, then that's 7.56 watts from the battery pack.

If I assume your MCE is wired in series, then it'll have a Vf of 3.5 * 4 = 14v. So at 700ma the LEDs are using 9.8 watts. 10.8 watts from the battery with 90% driver.

That gives you a total power consumption of 18.3 watts. A 14.8V 2.5ah battery pack is 37 watt-hour. So that gives a runtime of 37 / 18.3 = 2 hours. That's a theoretical number that assumes you'll really get 2.5ah out of that battery pack. You'll never will get that much in real life and experience tells me to expect a runtime of around 1:45 minutes.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•