Charging EV with varying input power

[tbader]

I wanted to post this as a thought experiment. I wanted to preface this by saying I do not plan to build/experiment with anything being discussed as high currents/ high voltages are being dealt with and I am no expert.I am curious on designs for a system that will take the energy from a renewable system and transport it to a car battery. Let’s say you have a system (regenerative breaking, solar, etc), whatever the case may be, and that does not act as a steady power supply. Obviously, it would not be recommended to charge the battery directly as there will large fluctuations in the battery as well as with the current. With that being said, a supercap bank would be a good “middle man” in the system to regulate the input to the battery. I am interested in complete systems that can do just that on large scales. For the battery, let’s just assume an 85kWh battery, with a voltage of 403V. The input power from the the system that will charge the battery outputs voltages anywhere from 20-100 and the current output varies from 5-60 amps.

Here is an example of the configuration of the battery pack:

Pack - 16 modules wired in series. 403V fully charged.
Module- One of the 16 sets of cells fully charged output of 25.2V from sixsub- groups wired in series.
Sub-group - One of the six sets of cells within a module. These are all connected in parallel to form a high capacity 4.2V battery (fully charged). These consist of about 74 individual cells. There are 96 total sub-groups.I would appreciate ideas and thoughts on how to deal with a system that varies drastically with the output and regulate it to charge such a large battery?

 

[donpacino]

Look up buck boost converters https://en.wikipedia.org/wiki/Buck–boost_converter

In some architectures the high voltage lines on electric vehicles use those converters (or similar converters) to combine different sources onto one bus.

 

[billy_joule]

I am curious on designs for a system that will take the energy from a renewable system and transport it to a car battery. Let’s say you have a system (regenerative breaking, solar, etc),

I think most, (maybe all?) off the shelf AC motor controllers for EV's can do regenerative braking nowadays, some have schematics floating round online. There's a few open source design out there too, here's one.

 

[CWatters]

What is the max charging current the battery will allow? That determines how much of the available power can be put into the battery. Any extra will have to be either wasted, or stored by other means.

 

[f95toli]

Maybe it would be worth looking up the specifications for the battery pack and regen for a Formula E car? Whatever system they use should be state-of-the-art.
The system in use this season was developed by Williams, and I believe quite a lot of the information is public.

I do know the drivers/teams do have to continuously change the amount of braking throughout the race depending on how much energy has already been used. Battery temperature is a very important factor: regen is more complicated on hot days,

 

[tbader]

What is the max charging current the battery will allow? That determines how much of the available power can be put into the battery. Any extra will have to be either wasted, or stored by other means.

The receptacle on the car can take 396V@225A which is almost 90KW of power, At least that is from the supercharger. Obviously if you use a standard Tesla home charger, they can output 80amps to the battery at 240volts. My question/thinking was a super cap bank that can supply a trickle charge to the battery and regulates the fluctuations.

 

[tbader]

Look up buck boost converters https://en.wikipedia.org/wiki/Buck–boost_converter

In some architectures the high voltage lines on electric vehicles use those converters (or similar converters) to combine different sources onto one bus.

I have done some research on them but a base buck converter won't be helpful from my understanding. Is there a type of converter like a full bridge boost converter that would be more applicable? I am not sure if you can combine the converters in series, as that could be another option?

 

[tbader]

Maybe it would be worth looking up the specifications for the battery pack and regen for a Formula E car? Whatever system they use should be state-of-the-art.
The system in use this season was developed by Williams, and I believe quite a lot of the information is public.

I do know the drivers/teams do have to continuously change the amount of braking throughout the race depending on how much energy has already been used. Battery temperature is a very important factor: regen is more complicated on hot days,

I am going to have to look into some of their systems, i appreciate the tip. Do you know if they just take the energy generated and just feed it back into the battery? My thinking was if you had a supercap bank that as the battery uses energy, the supercap bank supplies a constant trickle charge to the battery.

 

[mheslep]

The link assumed lead-acid batteries. That's already out-of-date.

Lead acid is much heavier than alternatives, but it remains for the moment the lowest up front cost rechargeable battery technology per kWh stored, about $150/KWh. Over a battery lifetime lithium ion may now have a cost edge over lead acid.

In any case the number of moles of metal, lead or lithium, required for backing up a half TW load for seven days remains the same as calculated by Murphy in his Do The Math series, and the global lithium resource is several orders of magnitude short of building such a cubic mile of battery pack.

 

[donpacino]

I have done some research on them but a base buck converter won't be helpful from my understanding. Is there a type of converter like a full bridge boost converter that would be more applicable? I am not sure if you can combine the converters in series, as that could be another option?

Its possible that I misunderstood your post; but as someone who has worked on EVs, I completely disagree with your point that a buck boost converter will not work

 

[tbader]

Its possible that I misunderstood your post; but as someone who has worked on EVs, I completely disagree with your point that a buck boost converter will not work

I highly agree that they will work but I was saying a basic boost converter will not work for larger power. From my understanding, other forms of "boost converters" such as single switch forward, dual switch forward, half bridge and full bridge converters may be better for larger power but I am not sure on the benefits/differences for each? Or i can connect a bunch of boost converters in series to amplify the voltage even more, but I believe that is not safe to do.

 

[CWatters]

I think the OP needs to identify exactly what the energy source is. It's straightforward to do a basic voltage step up/down as required but that alone may not be sufficient if the source is regenerative braking. The devil is in that detail.

I once saw some solar powered model aircraft flying. They were so marginal on power that they used variable pitch propellers and varied the pitch to keep the solar cells operating at their max power point. If the motor drew too much current the solar cell voltage would fall and the total power would drop. Likewise if the motor drew too little. If you used your hand to make a shadow on the cells the prop pitch changed automatically to optimise the cell output. Very cool.

Something even more complicated is required with regenerative braking. You can't treat it simply as a variable power source. The generator has to act like a variable load. Others have already mentioned this.

 

[tbader]

I think the OP needs to identify exactly what the energy source is. It's straightforward to do a basic voltage step up/down as required but that alone may not be sufficient if the source is regenerative braking. The devil is in that detail.

I once saw some solar powered model aircraft flying. They were so marginal on power that they used variable pitch propellers and varied the pitch to keep the solar cells operating at their max power point. If the motor drew too much current the solar cell voltage would fall and the total power would drop. Likewise if the motor drew too little. If you used your hand to make a shadow on the cells the prop pitch changed automatically to optimise the cell output. Very cool.

Something even more complicated is required with regenerative braking. You can't treat it simply as a variable power source. The generator has to act like a variable load. Others have already mentioned this.

I wasn't referring to anything in particular, just wanted to know how to deal with variable power source that can generate large amounts of power for minutes and then a small amount of power at other times. There is no constant output from this energy source. Weather that's solar or regenerative or something else that behaves similar and doesn't output a constant supply. If the generated power is so marginal, how do you regulate such a supply to charge a battery? My thought was a supercap bank that acts as a trickle charger and can output a constant supply to the battery. Would this be inadequate?

 

[donpacino]

basic boost converter will not work for larger power

where did you hear this?

 

[tbader]

where did you hear this?

Should not have used the terms "not work" , was trying to say other forms of "boost converters" such as single switch forward, dual switch forward, half bridge and full bridge converters look to be better for higher power applications. I could be wrong, trying to learn more about the advantages/disadvantages of each.

 

[donpacino]

Should not have used the terms "not work" , was trying to say other forms of "boost converters" such as single switch forward, dual switch forward, half bridge and full bridge converters look to be better for higher power applications. I could be wrong, trying to learn more about the advantages/disadvantages of each.

Some other converters might have higher efficiency, but they will be more complicated. There will be more components, more weight, etc. Sometimes simpler is better. There are basic Buck, Boost, and Buck-boost designs on many EV platforms.

If the generated power is so marginal, how do you regulate such a supply to charge a battery?

Refering to regenerative breaking specificly. In most cases regen breaking actually won't recharge the battery. Instead it will be stored in capacitors (like you stated). But then used as energy for other loads. In order to truly recharge the battery you'll need to bring in more energy you are using. In the case of a series hybrid for instance, the layout is such that the engine (acting as a generator) charges the batteries, which then drives the electric motor.

But the battery, generator and electric motor are often on the same DC Bus. What actually happens is you have a rectifier to convert the AC to DC, possibly some sort of level shifter, then the main HV bus that connects the various High power components. you have an inverter to drive the electric motor, and you have the HV batteries. This HV bus usually has a lot of capacitance on it (this is your cap bank).

So when the electric motor "calls for power" (IGBTs open), the power often comes directly from the generator (which is stored in capacitance). When there is excess power generated, the battery charges. If you are generating too much power and can't charge the batteries anymore, you need to decrease power generation.

If you want or need to trickle charge the battery (for whatever reason), you would need a special setup. It is also likely that this would have to be a special mode of operation, as the battery would not be able to supply power. But the same principal would apply. take your power, convert it to DC, level shift it if needed, store it in a cap bank, use it when you need it. however note many of these hvbatteries do not need to be trickle charged.

please note this is an oversimplification.

 

[tbader]

Refering to regenerative breaking specificly. In most cases regen breaking actually won't recharge the battery. Instead it will be stored in capacitors (like you stated). But then used as energy for other loads. In order to truly recharge the battery you'll need to bring in more energy you are using. In the case of a series hybrid for instance, the layout is such that the engine (acting as a generator) charges the batteries, which then drives the electric motor.

It would most likely be perpetual motion if you had more energy than you are using but why wouldn't it charge the battery still? It couldn't act as a range extender. Obviously regen breaking doesn't produce that much but let's say you had a solar system for whatever reason on your car, could that energy in super cap bank still charge the battery and act as a range extender? And what size super cap bank would you need to be large enough to charge the battery in an EV?

And there is no benefit to trickle charging? It wouldn't act as a benefit to consistently charge the battery? Just curious on your opinion

 

[mheslep]

In production EVs, all the energy goes to either battery charge or friction braking, not capacitors. The Tesla S battery can absorb up to 60 KW of braking power

http://www.caranddriver.com/reviews/2013-tesla-model-s-test-review-the-braking-accelerator-pedal-page-2

 

[donpacino]

In production EVs, all the energy goes to either battery charge or friction braking, not capacitors. The Tesla S battery can absorb up to 60 KW of braking power

http://www.caranddriver.com/reviews/2013-tesla-model-s-test-review-the-braking-accelerator-pedal-page-2

You do not generate that much power from regen breaking in comparison to the amount of power used by the motors, and other electrical loads on EVs. There are often capacitor banks on the HV link between these systems. When regen breaking kicks in, very often the energy is small enough that most if not of the energy from regen gets thrown right back into the loads.

 

[mheslep]

There are often capacitor banks on the HV link between these systems...

Can you refer to such an EV using capacitors? I know some specialty cases have done so (e.g. race cars), but if any production EV models have included capacitors for energy storage, they've not disclosed it.

 

[Windadct]

For the inverter to work properly - you pretty much NEED a capacitor bank close coupled to the switching devices (IGBT/MOSFETs), and this can be an intermediate energy storage, as then a Buck/Boost charge / discharge element can manage the energy flowing to / from the battery.

But doing some quick napkin math -- a 100KW converter may have about 6mF (1100V - FIlm caps - actually that would be a lot for a vehicle) -- Assuming the Battery is 650V and during breaking the DC bus can go to 900V before the energy needs to go somewhere else ( the Max DC link Voltage for 1200V IGBTs is typically 900V)

So to change the DC Link voltage from 650 to 900 during regen 6mF * ( 900^2 - 650^2) = 2325j of Energy - Really not that much ( a 2000Kg car moving at 30KPH - has 68Kj? ) - Assuming 2 seconds of breaking time the system is dealing with 34KW - which is remarkably close to the rating of the home chargers. So the battery can probably take this energy at that rate,

But for short periods the battery can absorb a large amount of energy, since most of the restriction on the charging rate is to prevent over heating the battery - but I am sure there are still other limits.

For larger power ( high current) the basic buck and boost topologies are fine. The other topos typically good for low noise, higher switching frequencies, overall "cleaner" electrical power

 

[mheslep]

The inverter capacitance stores and dumps a few joules every switching cycle. The KE of a 3000 kg vehicle at 30 mpg is nearly 300 kilojoules, i.e. the difference between capacitors and super capacitors.

 

[tbader]

For the inverter to work properly - you pretty much NEED a capacitor bank close coupled to the switching devices (IGBT/MOSFETs), and this can be an intermediate energy storage, as then a Buck/Boost charge / discharge element can manage the energy flowing to / from the battery.

What configuration supercap bank would be optimal for absorbing all the energy as an intermediate energy storage?

 

[mheslep]

What configuration supercap bank would be optimal for absorbing all the energy as an intermediate energy storage?

No need for large capacitor based intermediate storage with good battery tech. Tesla batteries can handle energy at 100 KW, from a charger or from the regenerative brakes.

 

[tbader]

No need for large capacitor based intermediate storage with good battery tech. Tesla batteries can handle energy at 100 KW, from a charger or from the regenerative brakes.

Hypothetically, if it was a constant large power input (e.g 80KW) for a longer period of time, there is no inefficiencies and adverse effects on the batteries?

I know the superchargers charge at a high power output

 

[mheslep]

Hypothetically, if it was a constant large power input (e.g 80KW) for a longer period of time, there is no inefficiencies and adverse effects on the batteries?

I know the superchargers charge at a high power output

High charge rates for long periods cuts a little into battery life. Braking power last for seconds, and rarely gets above a dozen kw.

 

[tbader]

High charge rates for long periods cuts a little into battery life. Braking power last for seconds, and rarely gets above a dozen kw.

So the ~90kW Tesla's absorb from supercharging for about 40min does not cut into battery life?

 

[mheslep]

So the ~90kW Tesla's absorb from supercharging for about 40min does not cut into battery life?

Batteries come with so many charge cycles at a given rate of charge/discharge. Increasing the rate of charge reduces the number of cycles. Typical cycle life chart (not for the Model S):

This might be part of the reason Tesla has been exhorting its owners to use supercharging stations only for long distance and away from home travel, and not for everyday use.

...As a frequent user of local Superchargers, we ask that you decrease your local Supercharging and promptly move your Model S once charging is complete….”