Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Regenerative Breaking

  1. Feb 29, 2008 #1

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    Are there any IC cars out there that do regenerative breaking by running the engine as a compressor while slowing down?
     
  2. jcsd
  3. Feb 29, 2008 #2
    In actual production? I don't know of any. I worked on a project with Ford that used an accumulator to store brake energy, but I lost track of where they went with it. I don't think it went to production.
     
  4. Feb 29, 2008 #3

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    It just seemed like an easy thing to do, but that nobody was doing it.
     
  5. Feb 29, 2008 #4

    RonL

    User Avatar
    Gold Member

  6. Mar 1, 2008 #5

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    But nobody's trying to bottle up the compressed gas?
     
  7. Mar 1, 2008 #6

    RonL

    User Avatar
    Gold Member

    I have put a lot of thought into ways to recover energy from, cars and trucks, and to recover and store this much would involve the use of large, and heavy tanks. The reduced payload would not be justified by the energy recovered from compressed air.
    Somewhere in my past searches on the net, i did see where GM (i think) used a (50KW ?) generator driven by the diesel engine.
     
  8. Mar 2, 2008 #7

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    I guess I'm missing something, but assuming that the system can put the energy back into the drive, it doesn't seem like it would have to be that large.

    Back of the napkin calculations suggest that a 60,000 pound semi at 45 mph has about 6 megajoules of kinetic energy. Assuming the engine does 24:1 adiabatic compression, that works out to roughly 232 atmospheres ~ 23 million newtons per square meter. At that temperature and pressure a system would have to accomodate on the order of .235 cubic meters-that's about a 55 gallon flux. Since this compressed air is very hot, a regenerator (heat exchanger) should give some more improvements.

    For a car that's the 10th of the size of the semi, that's a 5 gallon air tank, and a motorcycle that's 1 100th the size of the semi, the tank is a half-gallon.
     
  9. Mar 2, 2008 #8

    RonL

    User Avatar
    Gold Member

    Looks like i need to work on metrics a little, I'm still old school:redface:
    I would look at the compression as 24:1 X 14.69 and think i would end up with a max pressure of a little less than 350 psi, and volume of air in cubic feet, determined by number of pistons, and rpm at start of braking and declining as speed slows down.

    The consideration of thermal exchange is great. One other idea is to make frame rails, and cross beams out of tube material and store compressed air in them, i once built a utility trailer using a pipe frame that my truck compressor kept under pressure, worked very well.:smile:
     
  10. Mar 2, 2008 #9

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    You're discounting the pressure from compression heating.
    If the compression is adiabatic we get:
    [tex]VT^\alpha=costant[/tex]
    so
    [tex]V_0T_0^\alpha=V_1T_1^\alpha[/tex]
    [tex]\left(\frac{T_1}{T_0}\right)^\alpha=\frac{V_0}{V_1}[/tex]
    [tex]\frac{T_1}{T_0}=\left(\frac{V_0}{V_1}\right)^{\frac{1}{\alpha}}[/tex]
    [tex]T_1 \approx T_0 \times (24^{\frac{1}{1.4}}) \approx T_0 \times 9.67[/tex]
    (This is absolute temperature, that is, Kelvin, or Rankine if you're wierd.)

    Plugging that into the ideal gas law gives:
    [tex]P_1 \approx P_0 \times 24 \times 9.67 \approx P_0 \times 230[/tex]

    ~ 3400 PSI after compression.

    Reiviewing this calculation suggests an in-cylinder temperature of ~2200 degrees C/4000 F before ignition, which seems much too high. Wikipedia suggests pre-ignition temperatures on the order of 700-900 degrees C, so it's more like:
    [tex]P_1 \approx P_0 \times 24 \times 3.5 \approx P_0 \times 84[/tex]
    so pressures on the order of 1000 PSI assuming 14.5 PSI at the beginning of the compression.

    In terms of storage, neither of these higher pressures should be a particular challenge, especially if there is an efficient thermal exchange system.
     
  11. Mar 3, 2008 #10

    RonL

    User Avatar
    Gold Member

    Well NateTG, I'll concede to your calculations, but will stick to my statement.
    Did some research, and realized I don't know jacks***, so i have to use what i have, and that is what engineers have tagged with a spec plate, two tanks R-12, rated at 450 PSI @ 650 degrees F., they each weigh about 200 pounds, and are 12" Dia.X 72" long.
    How many of these would it take to receive the energy we are talking about?
    Will the high temperatures flash your lubricating oils ? and how quick can temperature be removed, before the tanks get too weak to handle the pressure. It seems in my simple way of thinking, that in almost every process of handling energy, mechanical or thermal, a large energy movement requires a large weight of material to handle it.
    I think i fit into George Clooney's definition of his two friends "dummer than a bag of hammers", but at age 65 i have done a lot of things and still have all my fingers and toes, and i really do read the engineers spec. sheets.
    As for compressed air, i have a link to a site that has something i think is really interesting ( a drawing ) i would change it a little, if you are interested I'll post it.

    Ron
     
  12. Mar 3, 2008 #11

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    Socrates knew more than everyone else, because he knew that he didn't know anything. I've got a book, you've got experience. We both know that experience is more valuable.

    It depends a lot on the working pressure range and the efficiency of the compression. (Clearly, more energy is stored if the tanks go from 0 to 650 PSI relative, than if they go from 640 to 650.) In practice it seem to be more of a question of what the stored energy is worth and how much is needed for the desired application.

    (Bearing in mind that I have no useful experience with any of this stuff, I would be inclined to put some kind of temperature and pressure escape valves into the system, and test it.

    There's also the options of exotic materials, lots of engineering, and limited reliability. The technology in Formula 1 race cars, for example, is absurdly small for the amount of power involved, but also costs millions.

    I'm always looking for new and different things. (Not that I'm all that likely to ever accomplish anything with them.)
     
    Last edited: Mar 4, 2008
  13. Mar 3, 2008 #12

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    I think the problem with compressing air using the engine for regenerative braking would become a problem when you tried to harness that energy on acceleration. You would have to inject the compressed air back into the cylinders, which would cause an extreme lean condition when paired with combustion and could be quite catastrophic (unless the super-compressed air was only used without any combustion in the cylinder).

    On the plus side, the idea of using a compressor and tank to store energy rather than batteries and a generator is a sound one and is being worked on by other people right now:

    Hybrid UPS Delivery Truck in Development
     
  14. Mar 3, 2008 #13

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    I'm not sure how you come up with this number, but a piston compressor with a compression ratio of 24:1 and an intake absolute pressure of 1 atm (14.7 psi) will only reach a maximum output pressure of 24 atm (352 psi). There's no way it would put out 10 times that without a mulit-stage compression :confused:

    With that being said, simply using [tex]PE = P * V[/tex] calculates the total volume needed to absorb 5.5 MJ to be in the range of a 600 gallon tank.
     
    Last edited: Mar 3, 2008
  15. Mar 3, 2008 #14

    RonL

    User Avatar
    Gold Member

    I'm glad to see an article like this, it takes me back to around 1972 or 73, a popular mechanics story of a man that bought two Ford fairlanes, with 6 cylinder engines, and standard transmissions. He ran fuel mileage tests on both cars to establish economy, then removed the transmission on one,and installed a pump for hydraulic fluid, and one or two accumulators, the engine was controlled by pressure sensors, and would only be used when pressure was needing an increase, in order to start, or maintain speed. Any slowing, or downhill condition would recharge the system.

    I think the results were around double the fuel milage.

    I have seen a lot of ideas that have gone nowhere, only to resurface years later, as if they have just been thought of.:uhh: The cycle of life just keeps on going.:smile:

    But if i remember right, i was buying gas for around .40 cents per gallon.
     
    Last edited: Mar 3, 2008
  16. Mar 4, 2008 #15

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    My original calculation assumed adiabatic compression at 24 to 1 and came up with that factor of 10 due to heating (the calculation is posted above). I don't have any authoritative references, but wikipedia suggests that diesel engines typically get compression heating to 700-900 degrees C or 1000-1200 degrees kelvin, which is a factor of 3-4 in absolute temperature. This led to the revised estimate of 1000 psi.

    At least in theory, compression ignition engines are much less sensitive to running lean than spark ignition engines since there's no potential for pre-ignition. (I was under the impression that existing diesel tech allows for running a leaner mix to scale back power.) Since I'm off in theory-land, post-combustion air ignition might also be possible. Originally I was thinking that an air start/run could be used to reduce idling, but I don't know how much fuel consumption goes into that.
     
  17. Mar 4, 2008 #16

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    Hmm, I see where you are coming from there. I find it doubtful that the air will really gain that much heat for a variety of mitigating factors, but it is possible. I'll just add that the air will cool down as it is pumped into the storage tank (large, cold thermal mass), and as air is pumped through the engine during the regenerative deceleration it will cool the cylinders in the engine, reducing the heating effect. Since the air will end up cooling in the tank depending on how long it is stored, it's probable the system will only have air at 300-500psi to work with.

    This is true, so it seems that this specific hybrid technology would only be compatible for direct-injection gasoline or diesel engines. Still, the major concern would be exhaust gas temperatures as leaning the mixture out too much could damage the pistons or valves very easily.

    It seems to me that a better way to go would be to have a compressor/turbine in-line with the transmission that could handle regeneration and acceleration.
     
  18. Mar 5, 2008 #17

    NateTG

    User Avatar
    Science Advisor
    Homework Helper

    (Still in pie-in-the-sky land.)
    If the engine doesn't have fuel-injection, then using the cylinders to compress air gets a lot trickier anyway. Turning off fuel injection should be easier than running a separate air intake.

    Assuming that this is a compression ignition engine, and it's running a dedicated compressor, are there particular downsides to going to 2-stroke diesel?

    I was originally thinking this sort of thing would be a relatively simple modification, but reflection on the storage issue especially suggest that it's not that easy.
     
  19. Mar 5, 2008 #18

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    Well, a 2-stroke design would in theory have a higher specific power, but 2-stroke diesels tend to be more complex compared to their gasoline couterparts. A 2-stroke diesel would have to have exhaust valves with a properly sized scavenging blower to be able to compress air and then release it to a tank at the pressureized state. Then for utilizing that pressurized air, either the intake would have to be pressurized (which could present its own set of problems) or the air would have to be injected directly into the cylinder mid-compression stroke.

    If it was easy, someone would have already done it ;) In some ways, a pressurized air hybrid system is more difficult to implement than an electrical one (when the engine's cylinders are used for regenerative compression and acceleration). I do think however that a properly utilized compressor/turbine located at or around the engine's crank shaft could help remove a lot of technical hurdles, while providing functionally similar operation to that of current design electrical hybrid systems.
     
  20. Mar 5, 2008 #19

    RonL

    User Avatar
    Gold Member

    I'm having a hard time getting my brain to take this in, it seems you need to work against a resistance in order to slow the heavy truck, car, or moving object.
    I think of my shop compressor, and why i need a 5HP motor, and when does it apply it's power. When the process starts, the first draw of amps goes high, kicking the motor and compressor into action, but because the tank is empty, or low in pressure, the amps fall to a low state as little power is needed to push air into the tank, as the pressure builds, the amp draw goes higher and when the tank is at max pressure the motor is working the hardest.

    I guess a primary tank that is always at a peak pressure, and dumps the overload into empty tanks, but still something doesn't seem quite right:confused:
     
  21. Mar 5, 2008 #20

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    That is essentially correct. You may have also pointed out a problem- what is the regenerative braking simply doesn't slow the truck fast enough? It seems to me that a regenrative system would be most useful if it alone was able to absorb the entire (or a significant part of) the kinetic energy stored in the vehicle in a "short" amount of time; say as fast as 10-15 seconds. It could be the flow rate required to achieve that would not be available through the engine's cylinders alone...

    The initial high current draw is because the internal moving parts of the compressor are accelerated to operating speed by the motor very quickly. Once the piston and assorted parts have accelerated, the current draw that it falls to is essentially how much is required to overcome friction, and compress the air. Due to the design of the compressor, it takes more energy to compress the air as the pressure in the storage tank rises.

    I don't think the primary tank in an air pressure hybrid system would be at peak pressure all the time any more than a battery in an electrical hybrid system is fully charged all the time. It's likely that even if the system is very efficient, most if not all of the stored energy will end up being used to accelerate the vehicle again after a stop.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Regenerative Breaking
Loading...