# Supercapacitor electric model

I am doing a dissertation in Electrical and Computer Engineer which involves supercapacitors.

I read the norm IEC/EN 62391–1 and the 60384-1 but still I don't have a clear answer for what's the electric model to be used to simulate a ultracapacitor.

It seems to be a Series-equivalent circuit. (R_leakage || C ) in series with (R_ESR-L_ESL). But if in case I intend to use this model, I don't seem to find these parameters in the supercapacitor's datasheet, only C and R_ESR.

What should I be looking for?

• A way to measure these. Every model is just an approximation, you have to decide how far you want to go. There are even people that have lumped elements for every possible component of a supercap, and some that just use a capacitance, both work for their needs. – PlasmaHH Aug 31 '18 at 11:32
• A lot of datasheets give you a number for the leakage current. So maybe a combination of C, R_ESR and a constant current source (I_LEAKAGE) would be good enough for you? It really depends on what effects you are trying to simulate. If your topic is only super caps, you better come up with a good model and a way to measure it, if you just need it for a runtime analysis for a system, the small model might be enough already. – Arsenal Aug 31 '18 at 11:39
• @Arsenal The professor still not sure about my thesis but it is certain it will be with supercapacitors and I know I will have to do the BMS (battery management system) and all in simulink/matl. At this point I already did a boost-converter in simulink. The idea I have is: a renewable system, charging the supercapacitors, and in case it is needed (or the cost is favorable) it is charged from the grid. In case of a grid instability the supercapacitors would assist in stabilizing the grid (reactive power). Would it have a big impact in the "project" if the model is a simple one, C and R_ESR..!? – user3397159 Aug 31 '18 at 16:17

## 1 Answer

Here is a model and a paper you can use:

If you don't have the ESL, it could be measured, or estimated. If you know something about the physical construction of the capacitor for example (like the terminals or metal or even take the capacitor apart (not recommended, but I have done similar things to model other components)). By knowing the physical construction, you could estimate the inductance of the metal in the system. Again I'm not saying do this, but it can be done.

A better way would be to run tests with the model below doing a frequency sweep and then fitting the modeled data to the parameters of a super capacitor

• The dissertation would be all in simulation. Plz see the comment I wrote for Arsenal where I say what the thesis will +- about. – user3397159 Aug 31 '18 at 16:35
• First off, if your in the business of writing a dissertation, it would be in your best interest to throw your text speech with "plz" , shorting words and improper punctuation out the window when writing other professionals. I can't tell the difference between your speech and a 13 year old. I'm not mad, I'm offering you a tip. Second thing, you will need to simulate the ESL if you want to find out how fast you can pull energy in and out of the capacitor. If you don't need to know how fast, then it probably won't matter. – laptop2d Aug 31 '18 at 16:41
• You are right about the text speech, well said, and thank you for pointing out. I will look into the model you described and see how it goes, with the ESL for a better time precision. – user3397159 Aug 31 '18 at 17:00