Super Capacitor Bank Output

[Risky Rob]

I will be trying to start a deep well water pump and will be building a super capacitor bank to provide a short power boost.

The bank will be about 350 Farads @ 30V, which works out to 5000 watt/seconds. This means I have the potential for a 500 Watt power boost for 10 seconds?
Watt/seconds = 1/2 (capacitance x voltage)

 

[MattiFin]

Correct formula is E=0.5*C*V^2

Energy Stored on a Capacitor

Joule=watt*seconds

you would get 315000 Ws or 1000w for 315 seconds. BUT this would be physically rather large bank and you probably dont have that much capacitance. maybe you forgot to account for capacitance series connection formula?

https://www.allaboutcircuits.com/textbook/direct-current/chpt-13/series-and-parallel-capacitors/

15x350F 2.5v supercapacitors in series gives you 23F capacitance.

 

[Risky Rob]

Wow, thanks. I'm going to do some real world testing. The terminal holes on the pcb board are only sized for an 8 gauge wire, so that might limit the surge capacity. If I can get 500 extra watts to start a motor, I should be OK.

 

[Risky Rob]

Correct formula is E=0.5*C*V^2

Energy Stored on a Capacitor

Joule=watt*seconds

you would get 315000 Ws or 1000w for 315 seconds. BUT this would be physically rather large bank and you probably dont have that much capacitance. maybe you forgot to account for capacitance series connection formula?

https://www.allaboutcircuits.com/textbook/direct-current/chpt-13/series-and-parallel-capacitors/

15x350F 2.5v supercapacitors in series gives you 23F capacitance.

The caps are 6 in series, then 2 banks in series. 12 350F @ 2.5V.

 

[Risky Rob]

So , the capacitance would be ~30 X 30V or 900÷2 or 450 watt seconds?

 

[Bud Martin]

OP: The bank will be about 350 Farads @ 30V
What is the capacitance of each individual capacitor?

 

[MattiFin]

The caps are 6 in series, then 2 banks in series. 12 350F @ 2.5V.

If you have 12pcs of 350F capacitors in series that is 350/12=29F

Discharged from max to zero volts gives you 0.5*29*30*30 =13050 joules or wattseconds.
In real life you cant charge them to maximum voltage as you need some safety factor and you wont be able to discharge them to zero volts either.
(12x capacitors in series would be barely enough for nominally 24v system. I’d rather use 14x in series if this is 24v system)

And You get usable energy out of them only between something like 28 volts to 20 volts before voltage is too low for typical circuitry. (28^2-20^2)*29*0.5=5500Ws

Another way to look at this is from amps and coulombs: 1F capacitor voltage drops 1 volt per second when dischaged with 1 amp current.

29F capacitor bank discharged with 29amp current drops 1 volt per second. Or discharged at 290A voltage drops 10 volts per second.

Or 10 seconds at 14.5 amps, dropping viltage from 28v to 23v

 

[Risky Rob]

OP: The bank will be about 350 Farads @ 30V
What is the capacitance of each individual capacitor?

The nominal value is 500F for each capacitor, but people have tested these as being lower, about 350F. So 12 in series is only about 30F. I am using 30V as the level of charge (2.5V ea)

 

[cs1234]

I know the general consensus around here is super caps are generally a bad idea, just add more lifepo4 if you need more amps, but super caps are fudgin rad and I wish @Will Prowse would do a video integrating some of the packs like these ones..

Super caps for sale, cheap but far away.

 

[Risky Rob]

If you have 12pcs of 350F capacitors in series that is 350/12=29F

Discharged from max to zero volts gives you 0.5*29*30*30 =13050 joules or wattseconds.
In real life you cant charge them to maximum voltage as you need some safety factor and you wont be able to discharge them to zero volts either.
(12x capacitors in series would be barely enough for nominally 24v system. I’d rather use 14x in series if this is 24v system)

And You get usable energy out of them only between something like 28 volts to 20 volts before voltage is too low for typical circuitry. (28^2-20^2)*29*0.5=5500Ws

Another way to look at this is from amps and coulombs: 1F capacitor voltage drops 1 volt per second when dischaged with 1 amp current.

29F capacitor bank discharged with 29amp current drops 1 volt per second. Or discharged at 290A voltage drops 10 volts per second.

Or 10 seconds at 14.5 amps, dropping viltage from 28v to 23v

Thank you, I will be using these with LiFePO4 batteries and I plan to have about 30V on each. The inverter is 230V output, and 90% efficient, so the super capacitor boost will be something like 10 seconds at 1.3 Amps. The surge time for the pump motor is probably shorter than 10 seconds.

 

[Risky Rob]

I know the general consensus around here is super caps are generally a bad idea, just add more lifepo4 if you need more amps, but super caps are fudgin rad and I wish @Will Prowse would do a video integrating some of the packs like these ones..

Super caps for sale, cheap but far away.

I've watched some super capacitor tests and they often don't actually have the nominal capacitance. People with huge battery banks don 't need them, but if I can get an extra 1.3A @ 230V for 10 seconds, I'll take it. The cost was about $120.

 

[cs1234]

I've watched some super capacitor tests and they often don't actually have the nominal capacitance. People with huge battery banks don 't need them, but if I can get an extra 1.3A @ 230V for 10 seconds, I'll take it. The cost was about $120.

Be sure to let us know how it goes, super cap news and experiments around here are very rare.

 

[Risky Rob]

Be sure to let us know how it goes, super cap news and experiments around here are very rare.

I'm going to initially test the super capacitor bank by itself through a 24V split phase LF inverter. One thing I've read is to be careful the capacitor bank is the same voltage as the batteries. You could damage a battery otherwise.

 

[cs1234]

I'm going to initially test the super capacitor bank by itself through a 24V split phase LF inverter. One thing I've read is to be careful the capacitor bank is the same voltage as the batteries. You could damage a battery otherwise.

You need to make sure the bank is the same (or very close to) voltage when you connect it, if it isn't, it will VERY RAPIDLY try to make itself the same voltage.

The actual capacitor bank can support as high of voltage as you want though, say that 48v capacitor module in the link I posted, it just won't be using any of the upper limit of voltage if connected to a 12v or 24v system and will probably be wasting capacitor capacity (say it really fast over and over). Once you install it inline with the rest of your system it doesn't have the ability to get voltage higher than what flows through it.

 

[Hedges]

I'm going to initially test the super capacitor bank by itself through a 24V split phase LF inverter. One thing I've read is to be careful the capacitor bank is the same voltage as the batteries. You could damage a battery otherwise.

You need to make sure the bank is the same (or very close to) voltage when you connect it, if it isn't, it will VERY RAPIDLY try to make itself the same voltage.

The actual capacitor bank can support as high of voltage as you want though, say that 48v capacitor module in the link I posted, it just won't be using any of the upper limit of voltage if connected to a 12v or 24v system and will probably be wasting capacitor capacity (say it really fast over and over). Once you install it inline with the rest of your system it doesn't have the ability to get voltage higher than what flows through it.

Of course precharge capacitors through resistor before paralleling with battery.

Total energy in capacitor is 1/2 C V^2, but the energy available to inverter is difference between that energy when V = initial battery voltage minus that energy when V = low voltage disconnect of inverter.

Capacitor has a series impedance (resistance), and battery does too. That determines current sharing when inverter draws voltage lower, and voltage vs. current curves determine how much current is available.

Lithium battery can probably supply all the current inverter needs to start any motor. BMS maybe not. A system that bypassed BMS FETs and had BMS wired to remote on/off of inverter, if it provides suitable delay, could work.

What is surge of pump? If 40A 240V, about 10kW, from 24V battery that's about 400A. A 2C surge for one second from 200Ah battery? That would seem something it could handle. Maybe just get a BMS with specs to cover that.

I've watched some super capacitor tests and they often don't actually have the nominal capacitance. People with huge battery banks don 't need them, but if I can get an extra 1.3A @ 230V for 10 seconds, I'll take it. The cost was about $120.

An extra 250W won't do anything for you.
But if delivered in 0.5 seconds, an extra 5kW is probably what you need.

The nominal value is 500F for each capacitor, but people have tested these as being lower, about 350F. So 12 in series is only about 30F. I am using 30V as the level of charge (2.5V ea)

30F, drained from 28V to 24V would deliver 3120 Joules or Watt-seconds, according to my math.
So 6kW for 1/2 second. That could be useful.

But look up resistance, and if possible impedance at 60 Hz (ripple current drawn from battery.)
Determine how much current will be drawn from battery vs. capacitor.

 

[cs1234]

You guys and your math. Just hook it up and see what happens!

 

[Risky Rob]

You guys and your math. Just hook it up and see what happens!

There are lots of variables. For instance my LF inverter has 2 big caps on the main board, and the pump control box has a starting cap.

 

[Hedges]

Capacitors on AC and DC side do different things.
On AC side, it is phase shift for starting windings.
DC side, stored energy and reduced impedance to deliver a surge of power.

What is rated current of motor? I expect surge to be 5x that. Inverter needs surge rating that high for a second, not just milliseconds.

As cs1234 says, Just Do It.
How does inverter perform with well pump, everything as-is?

 

[cs1234]

I see super caps as a solution looking for a problem. You really have to play to their strengths to make good use of them. They are fantastic at capturing and releasing high amps but suck for capacity and long term storage.

I see them as being useful in very specific planned scenarios, but not really "integrated" into your main system, more of a separate subsystem.

Imagine an air conditioner or fridge with a high starting surge, but relatively low running amps and it runs on a cycle then "turns off" for a bit. Now imagine that ac/fridge being kind of a distance from your main system, or not wanting to design your otherwise appropriately sized / efficient system to accommodate that device. This is where a super cap bank might come in handy. Size the super cap bank to be large enough to supply the starting amps and then have a small lifepo4 or dc/dc charger to handle the rest of the load once it peaks up and recharge the super cap. You could even skip the small lifepo4 if you want, and supply the recharge/running current from the dc/dc charger connected to your main system using relatively cheap cable because you are keeping the amps low.

Assuming we are talking AC devices, not DC devices here, you would give this system it's own inverter, like an AIMS inverter, inexpensive/inefficient but capable of high amp loads. It would work especially well if the AIMS style inverter has one of those sleepy modes where it only really ramps up when huge loads come on.

The other advantage to using a separate inverter from your main inverter is you can play to the super caps strengths, which is higher starting voltage and being able to safely drain down to really low voltage without harming it. Using a separate dc/dc charger controller (a configurable one, like a Victron) and an inverter that has a good high voltage and low voltage range, you can really work that super cap down so long as you use some fat cables from it to the inverter. Super caps are also super light weight and if building your own, you could really cram them in wherever space allows or use all kinds of odd package sizes.

Another use case I could see for them would be similar to the above, but for if you need to run a different voltage system, say like an RV. You could use a 48v for your main system and battery storage, and then use a large enough ultra cap to handle your peak 12 volt loads, and just recharge it with a dc to dc charger system that is high enough to handle the sustained loads.

tldr Super caps are cool, but they aren't batteries and you have to create (find) specific problems to really maximize their awesomeness. If you just shove them inline with your normal batteries and inverter, they don't really get to shine. They work great if you are physical space, temperature or peak current limited.

 

[Risky Rob]

Capacitors on AC and DC side do different things.
On AC side, it is phase shift for starting windings.
DC side, stored energy and reduced impedance to deliver a surge of power.

What is rated current of motor? I expect surge to be 5x that. Inverter needs surge rating that high for a second, not just milliseconds.

As cs1234 says, Just Do It.
How does inverter perform with well pump, everything as-is?

The one time I tried to start the 230V pump, I think the problem was voltage sag. The pump ran for a few seconds, then the inverter shut down. I was very green, and not familiar with the process. Since then, I've learned more about my equipment and how to start large 115V motors. My next step is to test the 24V capacitor bank. I will charge it with my solar charge controller, as high as my inverter will accept and run some AC loads.