Im currently involved in Li-ion cell balancing methods. I have come to a time where i need a model of a li-ion cell (to have a voltage profile with State of charge). Im aware of a schematic to do this(other suggestions are welcome). I now need to make a model in spice. Can someone briefly describe to me the process for doing so?
Thanks in advance,
you can try it by using a lookup table. Therefore you only need a graph of the open circuit voltage over SOC. The cell type should match with your "planned" Li-ion cell in capacity and cell chemistry.
If its allowed to tell, I'm interested in which kind of cell balancing do you use (active or passive)?
Funkensprung said: Hi Bruce, you can try it by using a lookup table. Therefore you only need a graph of the open circuit voltage over SOC. The cell type should match with your "planned" Li-ion cell in capacity and cell chemistry.If its allowed to tell, I'm interested in which kind of cell balancing do you use (active or passive)? Funkensprung
A possible hint for the look-up table would be:
How to change the SOC in SPICE is a good question. Maybe you could do it by converting the cells capacity from SOC into Coulomb.What I mean: A fully charged cell (with 10Ah capacity for example) has 36000As . For estimation of the "SOC" - and the according cell voltage - you could integrate (coulomb counting) the current going through the cell. => But this is be only an idea. I never did it in SPICE.
In a former work I modeled a Li-ion cell in MATLAB, together with an idealized model of a transformer for cell balancing. The major focus was on the performance of cell balancing.
I think it mainly depends on what you want to get out. Would it be only the general behaviour/performance of the switched capacitor (then MATLAB), or detailed switching behaviour of the flying capacitor (then SPICE).
I think some thing to do with coulomb counting for SOC and then a look up table for OCV. The model i found in Intusoft ICAPS uses IsSpice4 and has a look up table for the OCV. However im unsure how they get the SOC. I know there are a few RC combinations to get time constants for behaviour but the actual SOC change is a mystery to me. Maybe you can enlighten me :-).
The model: http://www.intusoft.com/nlhtm/nl78.htm#Modeling_Lithium_Ion_Batteries
The model is based on a paper : http://www.reapsystems.co.uk/_publications/Published%20Paper.pdf
Im mainly involved in investigating this method for balancing; after finding other methods are either inefficient of bulky and expensive. Meaning: carrying our simulations to see the lengths of time to balance and the minimum SOC difference that can be resolved etc; then building a test setup for a specified battery.
Im a student :-)