DIY Electric Car Forums banner

Capactiors in parralel

5K views 10 replies 7 participants last post by  Sparrow159 
#1 · (Edited)
So I believe I've seen some threads around here about using capacitors to help ease the peak load stresses on a pack as well as potentially smoothing out the 'pain' of the pulses from PWM controllers.

That was before it really applied to me... now that it does I wonder how it can work. All I really remember was it seemed it probably wasn't worth it because for the money you could simply get more cells and be better off.

I just want to verify this is true -- because if I can reduce the loads on my cells a bit, I'd be very happy.

Reading this page: http://electronics.howstuffworks.com/capacitor2.htm it's clear that using the capacitors for any significant storage of energy is simply not going to happen. But it does mention:
Capacitors can also eliminate ripples. If a line carrying DC voltage has ripples or spikes in it, a big capacitor can even out the voltage by absorbing the peaks and filling in the valleys.
That sounds great.

With the goal of taking some of the burden out of high-discharge events (ie: quick acceleration) I did a little mathing. I figured if I can take some of the load for 3 seconds that would cut almost all the high-level loads from my batteries. My pack is 150V and I have it limited via the Soliton Jr to only pull 400A max.

3 seconds at 400A is .33 amp-hours.

.33 amp-hours at 150V is roughly 8 farads.

I found these 9 Farad 2.5V capacitorswhich can be purchased for $8.62 per. At 2.5V I'd need a set of 63 or so to match my pack voltage. That's 541 dollars. Not bad, really.

Now my first question: Is the above right? I think I did the calculations correctly, but I'm not 100% positive.

And the second question: How exactly would this work? I would imagine that if you made a 'pack' of these in parallel to your main pack -- they would effectively 'charge' each other and always be at the same voltage. Then when you gave it lots of throttle they would both drop in voltage (the Lithium due to sagging, the capacitors discharging the amount of the voltage sag). Then when the discharge was over the lithium cells would recharge the capacitors up to the voltage levels of the lithium cells. Is the correct?
If the above is correct -- how quickly would the capacitors pull energy from the lithium cells after quick discharge? Would the lithium cells continue to discharge at a high level into the capacitors to the point that I would have effectively just 'time-shifted' the the load and not really evened it out at all? You know... it doesn't pull as hard from the pack when you hit the throttle... but it continues to pull hard after you let off. In the end, gaining you nothing...?

And the third question, which may be moot based on the answers in the second. If capacitors drop in voltage proportional to their charge level, then we would only really have the capacity of the delta from the 'charged pack' level and it's 'sagged' voltage level. So lets say I'm normally at 157v and I sag to 130 ... I only really get the 27volts delta in capactitive capacity... right? That's only 17% of the capacitors capacity. ... meaning one would really need 5-6x (or so) of their desired ah 'omph' in capacitor farads. Is this correct?
You can find 70F ones that aren't physically much larger than the 9 Farad one's above for only $11 (so, still affordable) -- that'd be a ~3ah capacitor pack (of which you could use up to .51ah during a massive voltage sag... right?


Anyone have any answers, thoughts, corrections, or comments?
 
See less See more
#2 ·
I think you've hit most of the points pretty well. You're not timeshifting the hit on the battery, but not reducing the overall load, really. Someone had a huge supercap bank that demonstrated a decrease in peak draws, but it wasn't well documented. I don't think anyone's tried a reasonable experiment. Also, good point about the working voltage. Only the 27 volts (less actually) would ever be used, so you can't even tap into most of the power in the caps.

When you consider that you need over 100 of those 70F caps just to equal one AA, it doesn't seem like it could be worth it. I think modern lithiums are good enough to not care that much.
 
#3 ·
when they wrote
---------------------------------
a big capacitor can even out the voltage by absorbing the peaks and filling in the valleys.
------------------------------------

they are talking about spikes and valley inside the PWM ripple , a controller generaly works at 10Khz , so you have the IGBT ( or the mosfet ) changing from on to off every 0.0001 second ... inside that time a capacitor can supply the power to keep everything smooth but this is the time range of cap usefulness ... thousandth of seconds .. no more.

Supercap are a different story , the can hold much more power , but almost nothing compared to battery , as rule of thumb consider this equivalence .. 72F = 1Ah .. this mean that a Supercap of 72F can store the same energy of a battery of 1Ah of same voltage.

So if your current is for example 100A , you are going to empty your 9F supercap in less than 4.5 second IF you have a supercap dedicated controller.

If you just put battery and supercap in parallel you can only use very small percentage of this because in supercap energy is stored from 0 to full voltage , in parallel with battery you can use only from full voltage to sag voltage so about 10% of total energy.

In short words , if you have 100A of current your 9F supercap will help the battery for the first 0,5 second , after this time your battery will be alone again.

Sometime 0.5 second can be helpful because is the very fist moment when the car start from still and your motor works in a very low efficiency range and current are very high

so , check your current , check if the same money can give you more battery and make your choice
 
#4 ·
BTW.

i forgot to mention supercaps NEED a BMS , if a supercap exceed the 2.5V will short so you will have a short circuit across the main battery wire because once one supercap fail all the other will fail in chain.


So if you are still thinking about supercap .... first search also for a way to protect them form overvoltage a cheap solution is to use a simple zener+resistor but at the price of derating the supercap to 2.0V because zener are not so precise and you cannot take the risk ( remember one supercap broke = ALL supercap broken)

The zener + resistor will add about 1$ for each supercap and you also can use the supercap only up to 2.0-2.1 volt.


on the other way there are supercap much more powerful than the one you linked , for example this cost 12.50$ and store 350F.
http://www.tecategroup.com/store/index.php?main_page=product_info&cPath=18_20_32&products_id=1221


To reach your 150V nominal battery pack you will need at least 100 supercap ( how much is your full charged voltage .. and add at least 15V for safety)

so for (12.50+1$ ) * 100 = 1350$ you will be able to discharge 100A for about 17.5 seconds or 200A for about 9 second or 400A for 5 second ( at high current there is more sag so you can use more supercap)

OF course if you use a dedicated booster for the supercap you can use much more energy from the supercap but price are very high
 
#5 ·
Thank you for the information.

A different idea...

What if you hooked the caps up to each individual lithium cell? Like put a 5V cap (so there would be no risk of shorting upon failure - and fuse it just in case anyway) across the terminals for each cell in your main traction pack.

Would that simplify things at all and still stiffen the pack, or simply make a bomb?
 
#6 ·
Thank you for the information.
A different idea...
What if you hooked the caps up to each individual lithium cell? Like put a 5V cap (so there would be no risk of shorting upon failure - and fuse it just in case anyway) across the terminals for each cell in your main traction pack.
Would that simplify things at all and still stiffen the pack, or simply make a bomb?
this is what is called Hybrid battery , Bollere' car is using this solution becuse they owns both a supercap factory and a Lithium battery factory , so they combine the 2 elements in one single pakage ( and also integrate an electronic to take care both of lithium cell and supercaps cell)


using supercaps wired directly on each cell is a lot safer , but you are wasting more power because you use less voltage for each supercap ( you need 2 supercap to take care of a single Lifepo4 cell) , it works a little better for Li-po .. but Li-po don't need supercap.

It works really well for Lead battery because each cell have a max voltage of 2.4 and also act as a shunt
( it works so well because supercaps were invented to work with lead)



If you want to stay on the safe side and not taking the risk of trashing 1300$ of supercaps, wiring each cell with his own supercap bank is a clever , safer but a little more expensive way
 
#7 ·
You need the capacitors in series to get higher voltage, not in parallel.

Also, the capacitance gets lower if you put capacitors in series (see this link).

Two 9F 9V capacitors in series would give you 4,5F at 18V. If you go up to 150V+ then you have almost no capacitance left...

You'll probably be better of finding capacitors that match your pack voltage and them putting those in parallel.
 
#8 ·
I found these 9 Farad 2.5V capacitorswhich can be purchased for $8.62 per. At 2.5V I'd need a set of 63 or so to match my pack voltage. That's 541 dollars. Not bad, really.

Two problems - capacitors in series act like resisters in parallel - so your 63 bank of 9 Farad capacitors would have a capacity of 9/63 Farads

To get 9 Farads at 150v would need 63 x 63 capacitors - too expensive!

BUT you would need more than 9 Farads - you calculated you would need 8 Farads BUT that was with 150v "droop" - you would need to restrict the "droop " to ~1.5v to get any benefit so you would need 80 Farads

$8.62 x 63 x 63 x 10 = $342,000
 
#9 ·
#10 ·
Except that those can only take 100A, so unless the main batteries are doing most of the work anyway, they'll fry.

Maybe 2 or three in parallel could work, but the high c-rate for small cells is often based on the fact they are small, and can dissipate heat quickly. If you stick a pile of them together they lose that advantage.
 
#11 ·
Texoma did this back in 2009 using Supercaps that don't need to be put in serial nor do they need balancing boards. I just bought 49 200V Supercaps to mimic what he did back then for my car. You can search for his post by using the term "Fluxcapacitor." Cool idea, I hope it works for me.

Good luck,
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top