View previous topic :: View next topic |
Author |
Message |
Christophe
Joined: 10 May 2005 Posts: 323 Location: Belgium
|
Measuring charge of NiMh batteries with PIC (adc) |
Posted: Fri Jun 23, 2006 6:36 am |
|
|
Hello,
We are charing 4X niMh 1600mAh battery pack at 600 mA. The nominal voltage is 5V. They seem to be charged at 6.1V
We are using the LTC4060. Review the schematic:
http://users.[spam].be/chs/Werk/Laadcircuit.jpg
Battery voltage is measured by a P�C processor ADC via a voltage divider
Now I am searching for a way to know how much % the batteries are charged? I can detect the start and the end of a charge cycle.
Is the max. voltage always the same when batteries are full?
If the max battery voltage gets higher, does the charge curve gets higher linearly? |
|
|
Humberto
Joined: 08 Sep 2003 Posts: 1215 Location: Buenos Aires, La Reina del Plata
|
|
Posted: Fri Jun 23, 2006 7:20 am |
|
|
Christophe,
Microchip has a very good AN: Intelligent Battery Charger Reference Design that you would read it before to start, it has some algorithms and reference values that support NiMh technology.
http://ww1.microchip.com/downloads/en/AppNotes/30451c.pdf
Humberto |
|
|
Christophe
Joined: 10 May 2005 Posts: 323 Location: Belgium
|
|
Posted: Fri Jun 23, 2006 8:32 am |
|
|
Tks for the input but all that is done by the LTC4060. I only want to know the %charge when charging. |
|
|
rwyoung
Joined: 12 Nov 2003 Posts: 563 Location: Lawrence, KS USA
|
|
Posted: Fri Jun 23, 2006 8:57 am |
|
|
Read the application notes anyway. One thing that is not easy to estimate is % charge based only on the pack voltage. You really need to be counting coulombs. Count them going in and count them coming back out again.
Look around on the TI web site, they have some ICs that do this and probably enough information to figure out how to make a PIC plus a few other parts emulate the same function. _________________ Rob Young
The Screw-Up Fairy may just visit you but he has crashed on my couch for the last month! |
|
|
Ttelmah Guest
|
|
Posted: Fri Jun 23, 2006 10:16 am |
|
|
The normal key, is Dv/Dt.
You cannot really rely on voltage with almost any battery. However when they become fully charged, the cell voltage will stabilise, then drop, and this is the indicator of fully charged, normally used for NiCd batteries.
Unfortunately, NiMh, batteries only show a tiny drop, so this is not a common method on better chargers. The rate of change of temperature, rises sharply at the full charge, so heat is a good indicator of NiMh charge completion. However some chargers do just look for the levelling of the voltage characteristic. 'Belt and braces', with a limit on heat, combined with voltage change detection, is also effective.
You can use the cell voltage as a _very_ approximate indicator of charge state, on the way to this point.
Best Wishes |
|
|
kender
Joined: 09 Aug 2004 Posts: 768 Location: Silicon Valley
|
on fuel gauges |
Posted: Fri Jun 23, 2006 2:41 pm |
|
|
rwyoung wrote: | One thing that is not easy to estimate is % charge based only on the pack voltage. You really need to be counting coulombs. Count them going in and count them coming back out again. |
I would like to second that oppinion. Also, there is s number of fuel-gauge chips that combine the current-sense amplifier, A/D, digital integrator and sometimes EEPROM. Most of them are designed for the Li chemistries, but theer are some for the Ni chemistries as well. Usually their digital interface is I2C. Some are designed to work with the miscrocontroller inside the battery pack (like my favorite MAX1660), while others log the counted coulombs into the EEPROM, and don't require a uC in the battery pack.
Obviously, you could roll your own PIC-based fuel gauge. |
|
|
Christophe
Joined: 10 May 2005 Posts: 323 Location: Belgium
|
|
Posted: Mon Jun 26, 2006 12:46 am |
|
|
Thks 4 the input. So I've learned that the only accurate way to measure battery charge is a fuel gauge IC counting coulombs in and out. Now as hardware is allready finished I cannot implement this.
With the LTC4060 I've got -DV/Dt; DT/Dt and a time limit. So that should be safe.
As for battery charge I've only got the battery voltage.
So
6.1V = 100%
5.9 = 90%
..
5.5 = 10%
using the curve and a lookup table.
Q: Is the maximum battery voltage always _approximately_ the same at full charge? |
|
|
asmallri
Joined: 12 Aug 2004 Posts: 1635 Location: Perth, Australia
|
|
Posted: Mon Jun 26, 2006 12:56 am |
|
|
No. It depends on the temperature of the battery pack. _________________ Regards, Andrew
http://www.brushelectronics.com/software
Home of Ethernet, SD card and Encrypted Serial Bootloaders for PICs!! |
|
|
JFK Guest
|
|
Posted: Mon Jun 26, 2006 2:45 am |
|
|
asmallri wrote: | No. It depends on the temperature of the battery pack. |
Absolutely spot on, as the temperature of the batterys vary so will the their voltage so this is not a very good indication of a fully charged battery.
As you have access to the battery voltage I would suggest that the best method that you can implement to detect full charge is to look for the peak voltage, this can be easily implemented using the an A/D input on the PIC.
JFK |
|
|
|