Featured Rotary returns as range extender

Discussion in 'Prius, Hybrid, EV and Alt-Fuel News' started by bwilson4web, Mar 4, 2018.

  1. bwilson4web

    bwilson4web i3 and Prime

    Joined:
    Nov 25, 2005
    21,171
    11,298
    0
    Location:
    Huntsville AL with 2014 BMW i3-REx
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    A range extender engine only has to deal with steady-state, drag power. In the BMW i3-REx it is the generator, not the engine that limits the sustained, highway cruise speed to 70 mph. Sad to say, the data suggests the engine was not optimized for fuel efficiency.

    Bob Wilson
     
    RCO likes this.
  2. Trollbait

    Trollbait It's a D&D thing

    Joined:
    Feb 7, 2006
    11,659
    4,601
    0
    Location:
    eastern Pennsylvania
    Vehicle:
    Other Non-Hybrid
    From your reports, it sounds like BMW gave the genset low priority. Don't blame them; can't have perfect with a deadline. The carbon fiber was their focus, and the gen1 Volt ICE was similarly lacking.
     
    RCO likes this.
  3. hill

    hill High Fiber Member

    Joined:
    Jun 23, 2005
    13,867
    4,488
    54
    Location:
    South OC So Cal & Nashville, TN
    Vehicle:
    2004 Prius
    Model:
    IV
    similarly we all noticed the growth in engine size with the Prius. It almost seems counter-intuitive, but the larger ICE can run more efficiently under many conditions.
    .
     
    RCO likes this.
  4. austingreen

    austingreen Senior Member

    Joined:
    Nov 3, 2009
    12,426
    3,227
    0
    Location:
    Austin, TX, USA
    Vehicle:
    2010 Prius
    Model:
    II
    Let's see on the issues of cost and weight I thought I already covered this. If the engine is designed to last for years and not burn large quantities of oil, costs will be lower for a more efficient 3-cyclinder which may even weigh less. On size, the ford and nissan engines can fit inside carry on luggage. Like the a1 though, a rotary will be easier to fit in the rear of an old design, a 3 cyclinder will need to have the rear designed for it as it is taller. You could fit a tiny boxer 4 cylinder engine in an old design, but that adds costs and weight.

    Mazda SkyActiv-R Rotary Engine Confirmed, Coming In At Least Two Displacements - autoevolution
    It appears the 2019 mazda phev is targetted for north america ;-) but mazda should have learned from bmw and chevy on engine cycles. Bob, a rotary just by its nature is about small size and weight not high efficiency. In order to meet US emissions and not burn large quantities of oil they must run rich with lower compression than a Atkinson engine.

    Its not counter intuitive at all, but some people seem to have some really strange ideas, and they repeat them enough that others believe them. This is helped by the obscure testing done by governmental organizations that favor smaller engines. In the case of the i3 and indeed the 1.5L in the prius c, the engines need to rev higher than a larger displacement or forced induction engine the additional friction hurts efficiency. I would say a 3 banger with around 1.2L displacement would be great for a 20 kwh phev ;-) with engine simply driving a generator. For a prius phev with a small pack and direct driving of the wheels from the engine, I think a 1.5L like the volt and clarity are about right sized. Now the best for toyota there might be to take their new 2L di/pi engine, and remove a cylinder to go to 1.5L but keep it at 41% peak thermal efficiency.

    I don't think the choice of engine was bad at all for the i3, as bmw was going for low cost and had no idea how much the engines would be used. That is a good reason for a design in 2012. They are kind of stuck with it now until a redesign (and i3 is likely to be killed instead of redesigned for a less expensive phev). They have sold over 100,000 and learned a lot about carbon fiber and what people want in a phev.
     
    RCO likes this.
  5. bwilson4web

    bwilson4web i3 and Prime

    Joined:
    Nov 25, 2005
    21,171
    11,298
    0
    Location:
    Huntsville AL with 2014 BMW i3-REx
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    So I went to www.fueleconomy.gov to map out plugin hybrid efficiency:
    [​IMG]
    • 2018 Prime vs Ioniq - one has higher efficiency in EV and they equal on gas
    • 2016, 2017, 2018 i3-REx - fuel efficiency is relatively poor
    Some of us may have a difference of opinion but without empirical data from products in hand or citing credible sources, try to avoid discussing opinion pieces. Silence is not always concurrence for example Wankel: Wankel engine - Wikipedia

    The only problem I have with the announced Wankel range extender is insufficient power.

    Bob Wilson
     
    #25 bwilson4web, Mar 8, 2018
    Last edited: Mar 8, 2018
    telmo744 and RCO like this.
  6. austingreen

    austingreen Senior Member

    Joined:
    Nov 3, 2009
    12,426
    3,227
    0
    Location:
    Austin, TX, USA
    Vehicle:
    2010 Prius
    Model:
    II
    I'm not quite sure what that is supposed to show. Depending on your point, I will have to violently agree or disagree.

    You are correct that we will have no idea the volume, weight, or efficiency of the mazda range extender until they build it and people can test it. There can be good or bad implementations of range extenders.

    The bmw i3 engine/generator+ other items adds 270 lbs mazda says theirs will add about 200 lbs. I have no idea what either engine weighs but the bmw motorcycle using a similar engine weighs 550 lbs. The weight includes not just the engine, but motor/generator, radiator, gas tank, fluids, manifolds, emissions control, exhaust, wiring, controllers, sound and heat insulation.

    For reference a similarly sized honda motorcycle engine weighs 125 lbs about the same weight as a 3 cylinder suzuki engine that sells in cars that cost less than a bmw motorcycle ;-) Nissan has a 1.5L 3 cylinder that weighs 88 lbs, but the ford ecobost 1L with iron block, exhaust manifold and turbo weighs 215 lbs about the same as the 1.8L in the prius prime. The toyota 1.3L 4 cylinder weighs 250 lbs though, so much heavier engines could get picked.

    On size, the ford 1L is 11.25" long and 8" wide and appears to be about 15" tall. The 1.5L Nissan is 20" tall and looks to be about 15" long and 8" wide. Either of these could shave weight if put in an Atkinson implementation.

    You can do a good or bad implantation for any of these.
    Compare Side-by-Side


    Compare Side-by-Side



    Looking at rwd i3s (has stickier tires than the i3) the tesla rwd 3 gets 10 more mpge in the city and 24 more mpge on the highway despite being much heavier with stickier tires. The i3 does well in the city though, and it was designed as a city car. The model 3 is designed to be more aerodynamic and this greatly improves highway mileage. The extra weight of the rex and extra reserve in the battery drops range by 10 miles and bev efficiency by 3%, some of this could be recouped by a front motors regen capacity. I really like the idea of a simple rex for lowering engineering and manufacturing costs.

    Nissan's e-power note has what is probably a engine/generator that is optimized for a compromise between low cost, low weight, small size, and efficiency. It's 1.2L 3 cylinder engine drives a generator, but only has a tiny 1.5 KWH battery. It mostly churns at 2500 rpm, and uses about 25% more gasoline than a prius.

    The engines in the prime, volt, and clarity had 40%, 39%, and 39% efficiency compared to electricity from the plug, the bmw i3 32%. I would think with a more efficient engine with a greater range of power sweet spot that rex number would move to at least 35%. The i3 is over 20% lighter than a clarity phev, it should be closer in efficiency with a better engine choice. We will see if a mazda rotary does better or worse, if it makes it into a car.


    [​IMG]
     
    RCO likes this.
  7. bwilson4web

    bwilson4web i3 and Prime

    Joined:
    Nov 25, 2005
    21,171
    11,298
    0
    Location:
    Huntsville AL with 2014 BMW i3-REx
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    Ok,
    [​IMG]
    • Although the Prime and Ioniq show the same fuel consumption, the Prime uses ~10% less electricity. This suggests the motor-to-wheel via the stepped, dual clutch gear or the motor/electronics are less efficient. Speculation, if the Ioniq plugin uses an induction motor, the Prime permanent magnet motors would be more efficient.
    • The three BMW i3-REx models have similar electrical power to the Accord and Volt but the fuel consumption is ~ 20% higher. The training manual shows no cooled-EGR nor are there any references to an Atkinson cycle but the rated power is half that of the motorcycle version of the same engine. This suggests the engine efficiency may be a problem. If the generator torque load is high enough to keep the engine from faster rpm, then the pumping losses would be reduced and it may not have a partial throttle operation problem. This is an area I hope to investigate. But notice the larger battery (and weight) versions suffer a significant loss of electrical efficiency. Weight matters.
    The Prime and i3-REx are excellent City cars. The Prime is ~20% more electrically efficient over the i3-REx so it is my first choice in +50F weather and 3 stop trips. The longer EV range i3-REx, 3x the Prime or 10 stops, and manual engine control makes it a better choice in cold weather.

    The Prime is an excellent Highway car, ~70% more efficient than the i3-REx. Just the i3-REx highway mileage, 39-40 MPG, is affordable. Not great but better than many and averaged in with the EV, much better.

    So I look forward to seeing the metrics from the Wankel range extender even though it will be hard to reach Prius thermodynamic regions. It looks like the Achates might be the next major step up in efficiency but there may be scaling issues.

    Bob Wilson
     
    RCO likes this.
  8. Rmay635703

    Rmay635703 Active Member

    Joined:
    Oct 16, 2016
    715
    470
    0
    Location:
    Somewhere in Wisconsin
    Vehicle:
    2013 Chevy Volt
    Model:
    N/A
    A smaller ice can be also made to be much more efficient than a larger one but then top speed and acceleration are affected.

    My 360cc car would get mid 60’s high way and up to 100mpg city if I was hypermiling the whole way.

    That same car can also get mid 30’s with dual carbs and electronic ignition
     
    RCO likes this.
  9. austingreen

    austingreen Senior Member

    Joined:
    Nov 3, 2009
    12,426
    3,227
    0
    Location:
    Austin, TX, USA
    Vehicle:
    2010 Prius
    Model:
    II
    I'm not sure if this is still true, but lots of information on the engine and operation. https://attachments.priuschat.com/attachment-files/2016/07/108943_12_W20-Engine.pdf I would suggest there is no reason for partial throttle on the i3-rex. The generator should keep the engine at full load, and being small displacement there will be lower pumping losses versus those 1.5L and 1.8L engines. We do likely have lower thermal efficiency as well as higher frictional losses. Lower power is the the result of choosing a generator to max out at 4300 rpm, which is probably already past efficiency peak of the engine, and higher speeds would cause worse nvh. The generator runs at 94% efficiency (23.5 KW) when engine is at 4300 rpm making 25 kw. Looking at the electrical efficiency the average efficiency of this engine in use has to be bellow 33%.

    I provided the tesla model 3 as an example of a 17% heavier rwd car with stickier tires but better city electrical efficiency. BMW definitely can improve here even if it gets heavier and keeps awd The bmw engine is a good choice for a scooter, but not for efficient highway driving. Add a cylinder, increase the compression to go further into atkinson mode, and not much weight would be gained, but it should get to 37% efficiency of the toyota 1L while requireing less cooling and less vibration and harshness. No cooled egr or other costs needed. When the car was designed, the rex was a good choice, and it makes even a better combination with the new bigger battery, but doing a redesign you would have more displacement.


    I look forward to see if a opposed piston engine can work in a car. We should see a turbo charged one in a pick up soon, but my guess is it will be at least decade before anyone tries it as a range extender. The skyactiv-x should be in a car within the next 2 years, and engineers and journalists are driving the prototype now. That also requires forced induction, but should work fine as a 1.2L 3 cylinder in a range extender. That would definitely be heavier than a wankel but mazda is talking about 44% efficiency in the test mules, and 50% with skyactiv-3.
     
    RCO likes this.
  10. Rmay635703

    Rmay635703 Active Member

    Joined:
    Oct 16, 2016
    715
    470
    0
    Location:
    Somewhere in Wisconsin
    Vehicle:
    2013 Chevy Volt
    Model:
    N/A
    Too bad they don’t list the BSFC chart for the engine then engine + generator

    My 360cc car has peak BSFC at 5200 RPM, the flame front speed VRS piston speed determines the ideal RPM for motor efficiency.

    Anyway the following is interesting and I could indeed see them not running at peak efficiency and partial throttle given...


    During the development of the BMW I01 the objective was to eliminate interference noises by the combustion engine and, at the same time, generate a sound pattern typical of the brand. This combined with a comfortable extension of the range. For this reason the output power and thus the speed of the range extender are influenced by various parameters. On the one hand, these are defined by the acoustics and range strategy, and, on the other hand, the ambient temperature also has a significant influence on the output power.
    In order to bring the combustion engine and the catalytic converter to operating temperature, the combustion engine is operated at a speed of 2,200 rpm - 2,400 rpm during the warm-up phase (5). This is dependent on the speed driven and, at the same time, represents the idle speed of the combustion engine. The period for this warm-up phase is roughly 360 seconds depending on the ambient temperature.
    The speed varies accordingly between 2,200 rpm and 4,300 rpm and also the output power of the combustion engine depending on the speed driven and the state of charge of the high-voltage battery. The resulting acoustics of the combustion engine are regulated conveniently by controlling the speed subject to the speed driven. In order to extend the range the engine speed is also influenced by the state of charge of the high-voltage battery.
    The following graphic high (The graphic does not match a typical BSFC Chart)
     
  11. hill

    hill High Fiber Member

    Joined:
    Jun 23, 2005
    13,867
    4,488
    54
    Location:
    South OC So Cal & Nashville, TN
    Vehicle:
    2004 Prius
    Model:
    IV
    Chart?
    .
     
  12. bwilson4web

    bwilson4web i3 and Prime

    Joined:
    Nov 25, 2005
    21,171
    11,298
    0
    Location:
    Huntsville AL with 2014 BMW i3-REx
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
    austingreen and hill like this.
  13. Trollbait

    Trollbait It's a D&D thing

    Joined:
    Feb 7, 2006
    11,659
    4,601
    0
    Location:
    eastern Pennsylvania
    Vehicle:
    Other Non-Hybrid
    The tl;dr is that BMW choose to run the range extender at speeds that match the driver's expectation of what the engine noise should be vs. running it for efficiency.
     
  14. austingreen

    austingreen Senior Member

    Joined:
    Nov 3, 2009
    12,426
    3,227
    0
    Location:
    Austin, TX, USA
    Vehicle:
    2010 Prius
    Model:
    II
    I am not 100% sure is your car is a subaru 360, but if so it complies with my proposed engine sizing. That was a 356cc twin cylinder 2-cycle air cooled engine in a car that with driver was only a little over 1000 lbs. Unfortunately emissions make 2 cycle engines difficult, although they can still be done. I would assume a 20 kwh phev complying with today's safety standards would be at least twice as heavy.

    That engine pumped out 15 kw at your peak rpm. My guess is you would want double that at the most efficient in a new mazda phev, or 35 kw in a car the i3's weight and aerodynamics. I don't think you could do air cooled today, but perhaps oil cooled (lots of the porsche engineering is available and off patent as well as motorcycle). Still with today's emission this correstponds to 2x (for 4 strokes instead of 2) and 2x for power. That makes 1424 cc. We can downsize from there because of fuel injection, better valving, higher efficiency, and upsize a little to give it atkinson valving, and that comes to 1L-1.2L depending on implementation. A rotary would likely be about half that displacement like a 2 cycle, but .... 0.33 sounds like it will have to rev very high and get out of max efficiency speed, or use a large battery buffer.

    The only place I see them running partial throttle is at the 6 minutes of warm up, and I'm not sure it is even partial throttle there. It appears once warmed up the engine runs at full load between 10 kw and 25 kw (23.5 kw after the generator at the highest power).

    Yep it was not optimized for efficiency ;-)
    Actually it probably is the efficient range of the engine, it simply is chosen for small size and nvh not efficiency. To bob's point the compression is 10.6, and it will be pegged in otto mode to provide enough power to the battery. Alternatively adding anouther cylinder and building to 12.5:1 expansion, would allow a more efficient atkinson mode and a cooler running engine.
     
    RCO likes this.
  15. bwilson4web

    bwilson4web i3 and Prime

    Joined:
    Nov 25, 2005
    21,171
    11,298
    0
    Location:
    Huntsville AL with 2014 BMW i3-REx
    Vehicle:
    2017 Prius Prime
    Model:
    Prime Plus
  16. Trollbait

    Trollbait It's a D&D thing

    Joined:
    Feb 7, 2006
    11,659
    4,601
    0
    Location:
    eastern Pennsylvania
    Vehicle:
    Other Non-Hybrid
    Seeing how Mazda's SkyActiv diesel passed Japan emissions without SCR, but couldn't pass them in the US, I'm guessing the US has stricter emission regulations than Japan.
     
  17. hill

    hill High Fiber Member

    Joined:
    Jun 23, 2005
    13,867
    4,488
    54
    Location:
    South OC So Cal & Nashville, TN
    Vehicle:
    2004 Prius
    Model:
    IV
    It could be a thing where, like European countries, only certain types of pollutants are considered - whereas other pollutants aren't even on their radar.
    .
     
    RCO likes this.
  18. austingreen

    austingreen Senior Member

    Joined:
    Nov 3, 2009
    12,426
    3,227
    0
    Location:
    Austin, TX, USA
    Vehicle:
    2010 Prius
    Model:
    II
    270 grams/kwh minimum is about 31% thermal efficiency, about where I pegged it from the other data. Toyota's inexpensive 1L gets 37% efficiency, and in a 1.2L more expensive 3 cylinder you could probably get 40% peak thermal efficiency or around 210 grams/kwh in a fairly large sweet spot of power.
    It's the test. Mazda could do the easier European and Japanese tests, but with the US cycle NOx was either too high, or power allowed too low. Mazda is going to try it again. Mazda also failed emissions on the rotary in 2014. They now think they have the technology to pass. I'll wait to see the epa sticker with a car review to see how it does. Current mazda diesel would pass with SCR, but that adds cost that mazda thinks will kill sales.

    For 2 cycle the problem is warm up. For a traditional piston cylinder it is possible to go from 4 cycle to 2 cycle and back with a camless design. The first production camless car appears to be ready for production in china. Maybe in a couple of years a 4/2 cycle will be available as a range extender. This probably would be a very small and light turbo charged engine that would be a lot more expensive than a 1L 3 banger. Let's say you built a tiny 0.8L electrically supercharged or turbo charged 3 cylinder that could go from miller cycle 4 stroke, to 2 stroke, and used the pressure to help scavange exhaust. That could probably be smaller, more powerful, and more efficient than the bmw rex. I picked the size 0.8L because it is exactly half of a 1.6L V6 F1 engine, which are currently being built for efficiency. Mercedes has a very expensive one that hits 50% peak efficiency. You can either get up the horsepower from higher boost and rpms, or 2 cycle. In mercedes' case they go to 11 (11,000 rpm for the production car version).

    If the goal is only 31% thermal efficiency and lots of charging and discharging of the battery in the new mazda phev, then they can probably do that, but IMHO that is not a great goal.
     
    #38 austingreen, Mar 14, 2018
    Last edited: Mar 14, 2018
    RCO likes this.
  19. Trollbait

    Trollbait It's a D&D thing

    Joined:
    Feb 7, 2006
    11,659
    4,601
    0
    Location:
    eastern Pennsylvania
    Vehicle:
    Other Non-Hybrid
    Like particulates from gasoline engines in the US.
     
    RCO and Rmay635703 like this.
  20. El Dobro

    El Dobro A Member

    Joined:
    Jul 12, 2011
    5,713
    2,266
    1
    Location:
    NJ
    Vehicle:
    2014 Chevy Volt
Loading...