Forums

Li-Ion battery model

Started by Yzordderrex January 19, 2012
Greetings.

I am involved with a Lithium Ion Phosphate battery charger design.  I
am trying to get a handle on the behavior of the battery in order to
figure out how to get it to play nice with the charger.

I think the battery is going to act like a big capacitor.  It has a
0.25v drop after about 2 amp*hours, and a internal impedance of <0.009
ohms measured @ 1kHz AC.

Does it make sense to use this data to create a spice model and then
'try' to get the thing to run?  It seems to me that the standard
methods that we have all used to get current mode supplies to run
might fall a little short when it comes to this application.

Any input will be greatly appreciated.

regards,
Bob
N9NEO
On 1/19/2012 10:51 AM, Yzordderrex wrote:
> Greetings. > > I am involved with a Lithium Ion Phosphate battery charger design. I > am trying to get a handle on the behavior of the battery in order to > figure out how to get it to play nice with the charger. > > I think the battery is going to act like a big capacitor. It has a > 0.25v drop after about 2 amp*hours, and a internal impedance of<0.009 > ohms measured @ 1kHz AC. > > Does it make sense to use this data to create a spice model and then > 'try' to get the thing to run? It seems to me that the standard > methods that we have all used to get current mode supplies to run > might fall a little short when it comes to this application. > > Any input will be greatly appreciated. > > regards, > Bob > N9NEO
Try here, http://batteryuniversity.com/learn/article/types_of_lithium_ion and here, http://batteryuniversity.com/learn/article/charging_lithium_ion_batteries This DIY electric car forum is very active with a category specific to batteries and charging. The LiFePo4 is their high end battery. http://www.diyelectriccar.com/forums/ Mikek
On Thu, 19 Jan 2012 08:51:36 -0800, Yzordderrex wrote:

> Greetings. > > I am involved with a Lithium Ion Phosphate battery charger design. I am > trying to get a handle on the behavior of the battery in order to figure > out how to get it to play nice with the charger. > > I think the battery is going to act like a big capacitor. It has a > 0.25v drop after about 2 amp*hours, and a internal impedance of <0.009 > ohms measured @ 1kHz AC. > > Does it make sense to use this data to create a spice model and then > 'try' to get the thing to run? It seems to me that the standard methods > that we have all used to get current mode supplies to run might fall a > little short when it comes to this application. > > Any input will be greatly appreciated.
I think you want to treat the problem as one of getting the charger to play nice with the battery. The battery isn't really going to act like a big capacitor, at least not a linear one. The closest model is probably a resistor in series with a voltage source, or possibly a bunch of RC sections leading up to either a voltage source or a honkin' big nonlinear capacitor. I'd just make sure that you can make the charger run correctly when working into any of the expected short-term impedances of the battery, make sure it regulates well for maximum voltage, and make sure that it does a good job in the transition from current mode to voltage mode. -- My liberal friends think I'm a conservative kook. My conservative friends think I'm a liberal kook. Why am I not happy that they have found common ground? Tim Wescott, Communications, Control, Circuits & Software http://www.wescottdesign.com
"Yzordderrex" <powersupplyguy@netzero.net> wrote in message 
news:3beacab7-adda-420c-842a-a425a68ca676@q8g2000yqa.googlegroups.com...
> Greetings. > > I am involved with a Lithium Ion Phosphate battery charger design. I > am trying to get a handle on the behavior of the battery in order to > figure out how to get it to play nice with the charger. > > I think the battery is going to act like a big capacitor. It has a > 0.25v drop after about 2 amp*hours, and a internal impedance of <0.009 > ohms measured @ 1kHz AC. > > Does it make sense to use this data to create a spice model and then > 'try' to get the thing to run? It seems to me that the standard > methods that we have all used to get current mode supplies to run > might fall a little short when it comes to this application. > > Any input will be greatly appreciated. > > regards, > Bob > N9NEO
Check with the manufacturer. This type of battery is relatively new and I have seen a post like this not too long ago. I have a hunch the cell characteristics have been well investigated by now and the manufacturer would have all the details. Cheers
On Jan 19, 1:28=A0pm, "Martin Riddle" <martin_...@verizon.net> wrote:
> "Yzordderrex" <powersupply...@netzero.net> wrote in message > > news:3beacab7-adda-420c-842a-a425a68ca676@q8g2000yqa.googlegroups.com... > > > > > > > Greetings. > > > I am involved with a Lithium Ion Phosphate battery charger design. =A0I > > am trying to get a handle on the behavior of the battery in order to > > figure out how to get it to play nice with the charger. > > > I think the battery is going to act like a big capacitor. =A0It has a > > 0.25v drop after about 2 amp*hours, and a internal impedance of <0.009 > > ohms measured @ 1kHz AC. > > > Does it make sense to use this data to create a spice model and then > > 'try' to get the thing to run? =A0It seems to me that the standard > > methods that we have all used to get current mode supplies to run > > might fall a little short when it comes to this application. > > > Any input will be greatly appreciated. > > > regards, > > Bob > > N9NEO > > Check with the manufacturer. This type of battery is relatively new and > I have seen a post like this not too long ago. > I have a hunch the cell characteristics have been well investigated by > now and the manufacturer would have all the details. > > Cheers- Hide quoted text - > > - Show quoted text -
I asked manufacturer for a spice model and he said, "What do you mean spice? What is Spice?" I'm going with a 890,000F (yes, FARADS) capacitance with a 0.0008 ohm ESR. There's 20 of these babies in parallel and stacked 8 high.
On Thu, 19 Jan 2012 13:22:06 -0800, Yzordderrex wrote:

> On Jan 19, 1:28&nbsp;pm, "Martin Riddle" <martin_...@verizon.net> wrote: >> "Yzordderrex" <powersupply...@netzero.net> wrote in message >> >> news:3beacab7-adda-420c-842a-
a425a68ca676@q8g2000yqa.googlegroups.com...
>> >> >> >> >> >> > Greetings. >> >> > I am involved with a Lithium Ion Phosphate battery charger design. &nbsp;I >> > am trying to get a handle on the behavior of the battery in order to >> > figure out how to get it to play nice with the charger. >> >> > I think the battery is going to act like a big capacitor. &nbsp;It has a >> > 0.25v drop after about 2 amp*hours, and a internal impedance of >> > <0.009 ohms measured @ 1kHz AC. >> >> > Does it make sense to use this data to create a spice model and then >> > 'try' to get the thing to run? &nbsp;It seems to me that the standard >> > methods that we have all used to get current mode supplies to run >> > might fall a little short when it comes to this application. >> >> > Any input will be greatly appreciated. >> >> > regards, >> > Bob >> > N9NEO >> >> Check with the manufacturer. This type of battery is relatively new and >> I have seen a post like this not too long ago. I have a hunch the cell >> characteristics have been well investigated by now and the manufacturer >> would have all the details. >> >> Cheers- Hide quoted text - >> >> - Show quoted text - > > I asked manufacturer for a spice model and he said, "What do you mean > spice? What is Spice?"
Which tells that the possibilities are two: One, that the manufacturer is foolish for not having a SPICE model of their batteries; or, two, that your are foolish for trying to wedge the battery behavior into a SPICE model.
> I'm going with a 890,000F (yes, FARADS) > capacitance with a 0.0008 ohm ESR. There's 20 of these babies in > parallel and stacked 8 high.
Well, feel free to ignore years of wisdom in battery charging that has made the designs work without a mindless reliance on SPICE, and have fun. -- My liberal friends think I'm a conservative kook. My conservative friends think I'm a liberal kook. Why am I not happy that they have found common ground? Tim Wescott, Communications, Control, Circuits & Software http://www.wescottdesign.com
Tim Wescott wrote:


> Which tells that the possibilities are two: One, that the manufacturer > is foolish for not having a SPICE model of their batteries; or, two, > that your are foolish for trying to wedge the battery behavior into a > SPICE model.
Yes, and assume that the full charge cycle takes 8 hours of real-time, do you want to run a SPICE simulation that covers the full 8 hour charge? Do check in with us next century and tell us how that simulation run turned out. Hopefully this is a switching converter with 1 MHz cycle and 1 ns edge rates, should get you a real "snappy" result. Jon
Yzordderrex wrote:
> On Jan 19, 1:28 pm, "Martin Riddle" <martin_...@verizon.net> wrote: >> "Yzordderrex" <powersupply...@netzero.net> wrote in message >> >> news:3beacab7-adda-420c-842a-a425a68ca676@q8g2000yqa.googlegroups.com... >> >> >> >> >> >>> Greetings. >>> I am involved with a Lithium Ion Phosphate battery charger design. I >>> am trying to get a handle on the behavior of the battery in order to >>> figure out how to get it to play nice with the charger. >>> I think the battery is going to act like a big capacitor. It has a >>> 0.25v drop after about 2 amp*hours, and a internal impedance of <0.009 >>> ohms measured @ 1kHz AC. >>> Does it make sense to use this data to create a spice model and then >>> 'try' to get the thing to run? It seems to me that the standard >>> methods that we have all used to get current mode supplies to run >>> might fall a little short when it comes to this application. >>> Any input will be greatly appreciated. >>> regards, >>> Bob >>> N9NEO >> Check with the manufacturer. This type of battery is relatively new and >> I have seen a post like this not too long ago. >> I have a hunch the cell characteristics have been well investigated by >> now and the manufacturer would have all the details. >> >> Cheers- Hide quoted text - >> >> - Show quoted text - > > I asked manufacturer for a spice model and he said, "What do you mean > spice? What is Spice?" I'm going with a 890,000F (yes, FARADS) > capacitance with a 0.0008 ohm ESR. There's 20 of these babies in > parallel and stacked 8 high.
I suggest that if you model (or better yet, _attempt_ to model then do ONE CELL. Also, as points of reference, get charge and discharge curve families for ONE CELL and also buy ONE CELL and do your own curves.
On Jan 19, 2:52=A0pm, Tim Wescott <t...@seemywebsite.com> wrote:
> On Thu, 19 Jan 2012 13:22:06 -0800, Yzordderrex wrote: > > On Jan 19, 1:28=A0pm, "Martin Riddle" <martin_...@verizon.net> wrote: > >> "Yzordderrex" <powersupply...@netzero.net> wrote in message > > >> news:3beacab7-adda-420c-842a- > > a425a68ca...@q8g2000yqa.googlegroups.com... > > > > > > > > >> > Greetings. > > >> > I am involved with a Lithium Ion Phosphate battery charger design. =
=A0I
> >> > am trying to get a handle on the behavior of the battery in order to > >> > figure out how to get it to play nice with the charger. > > >> > I think the battery is going to act like a big capacitor. =A0It has =
a
> >> > 0.25v drop after about 2 amp*hours, and a internal impedance of > >> > <0.009 ohms measured @ 1kHz AC. > > >> > Does it make sense to use this data to create a spice model and then > >> > 'try' to get the thing to run? =A0It seems to me that the standard > >> > methods that we have all used to get current mode supplies to run > >> > might fall a little short when it comes to this application. > > >> > Any input will be greatly appreciated. > > >> > regards, > >> > Bob > >> > N9NEO > > >> Check with the manufacturer. This type of battery is relatively new an=
d
> >> I have seen a post like this not too long ago. I have a hunch the cell > >> characteristics have been well investigated by now and the manufacture=
r
> >> would have all the details. > > >> Cheers- Hide quoted text - > > >> - Show quoted text - > > > I asked manufacturer for a spice model and he said, "What do you mean > > spice? =A0What is Spice?" > > Which tells that the possibilities are two: =A0One, that the manufacturer > is foolish for not having a SPICE model of their batteries; =A0or, two, > that your are foolish for trying to wedge the battery behavior into a > SPICE model. > > > =A0 I'm going with a 890,000F (yes, FARADS) > > capacitance with a 0.0008 ohm ESR. =A0There's 20 of these babies in > > parallel and stacked 8 high. > > Well, feel free to ignore years of wisdom in battery charging that has > made the designs work without a mindless reliance on SPICE, and have fun. > > -- > My liberal friends think I'm a conservative kook. > My conservative friends think I'm a liberal kook. > Why am I not happy that they have found common ground? > > Tim Wescott, Communications, Control, Circuits & Softwarehttp://www.wesco=
ttdesign.com- Hide quoted text -
> > - Show quoted text -
Thanks for the good input Tim. I'm just trying to get the charger not to oscillate. It's an existing design that the company is trying to get into production. Based upon an analysis I just finished I should be able to roll off the FB amp with a big cap and thing should run no prob. I'll let you know. regards, Bob P.S. As long as I'm ignoring years of wisdom when it comes to battery charging, do you think it might do me some good to cross my fingers?