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

Why such a large engine/generator in a Chevy Volt?

  1. Apr 2, 2017 #1
    I'm curious about series hybrids (Plug-In, or range extended EVs), and this has me wondering. I see that a 2016 Chevy Volt has a 53 KW generator and a 101 HP (75 KW) engine. It can run in full EV mode for ~ 50 miles, with reasonable acceleration and speed. Based on some approximations I've made it seems like something more like a 20 KW generator, and matching engine should be sufficient. Why such a large gap?

    I base this on battery capacity range of several EVs and some general numbers I've heard, that it only takes about 25 HP to maintain a car at highway speeds. When I take the EV range, and assume 60 mph at that range (a mile-a-minute) and the battery kWh, I come up with a ballpark of < 20 KW average for those vehicles. And estimate ~ 25-30 HP engine to account for 10% generator losses? In the ballpark, right?

    I'm further assuming that the 20 KW average would be more than enough for city driving, you get regen braking, and while stopped the engine is charging the batteries, so average power is far lower than the peak for the stop/go acceleration.

    Secondarily - I also know that the Chevy Volt runs the engine in parallel with the motors in some modes. This makes sense, as they avoid the losses in rotary>electrical>rotary conversion, and in the case where the energy is being stored in the battery (like when stopped), the round-trip losses of the battery charge-discharge. But the drive system seems so complex to me. Why not use an engine optimized for a narrow speed/power range (perhaps a high efficiency HCCI?), and couple it through a relatively low power handling CVT? Apply no more power/torque than the ~ 25 HP number from above? That would be far less demanding of a CVT than the kind in ICE cars, which must handle the full power/torque of acceleration. If the engine were only used to assist with ~ 25 HP from say 20 mph to 80 mph, over a narrow RPM range of the engine, a 5:1 ratio range, lightweight CVT could do the matching. CVT's in use today have ratio ranges wider than that. Power not required to propel the car would be directed to the battery (I'm assuming they can control the gen load/output through the field current?).

    Wouldn't that be simpler? What am I missing?
  2. jcsd
  3. Apr 2, 2017 #2


    User Avatar

    Staff: Mentor

    Well, for one thing, if it takes an average of 20kW to drive the car, you can't drive and charge at the same time unless you have more than 20kW. Second, engineers don't design for averages, they design for worst case, then add a safety factor. Third...it looks like the Volt isn't a true series hybrid:
  4. Apr 2, 2017 #3
    Yes, but you don't really need to drive at average high power and charge at the same time. You only need to maintain the charge of the batteries - if you are driving and the engine provides enough to propel, the batteries aren't being drawn down. You can charge them to a higher level when you reach a plug, or when the driving gets more conservative.

    Yes, maybe the 53 KW versus my estimated 20 KW is an ~ > 2x safety factor. Seems high, unless my estimate is way off, and I don't think it is. But if someone was on a 75 mph speed limit highway, with A/C going, I guess it could add up.

    I did comment about the Chevy Volt not being a 'pure' series hybrid, and it seems there would be simpler ways like I mentioned (low power CVT) to get power to the wheels. I wonder why these steps are not used?
  5. Apr 2, 2017 #4

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    I own a parallel hybrid PHEV, I was not interested in the Volt because I wanted a parallel hybrid - so there's a case where advertising blew up in their faces.

    I think where you go astray is assuming because you need maybe 25 hp to hold speed at highway speeds, you need only a 25 hp engine. Putting aside hybrids for the moment, one can use the same logic on ICE vehicles - why do cars start out at around 100 hp? (e.g. the Chevrolet Spark) The answer is that we require cars to do more than a steady speed on a highway: we may need to pass, we may need to go up a hill, we may need to pass when going up a hill, etc. Needing 60 hp, even when "grandma driving" is not all that unusual, although 50 is more common. And there have been times when 102 hp (max all-electric) didn't quite cut it and I was glad to have all 204 hp.
  6. Apr 2, 2017 #5
    Yes, and that's what the batteries and motor do for you. That's actually my point - the +100 HP engines we have are only called to put that out for a few seconds at a time when we go for max acceleration. A hybrid that can run in full EV mode doesn't need that from the engine, it only needs the much lower average power. The batteries/motor will deliver short burst of extra power then, and the average running ICE recharges them over time, between the big burst requirements.

    It's just like a big capacitor on the output of a power supply. My regulator might be limited at 1 Amp, but with a large cap, I can supply bursts of 10 Amps, or whatever I design for. I don't need a 10 Amp regulator to handle those bursts, as long as my average draw is not above 1 Amp.

    edit - Actually the Chevy Volt is a kind of parallel/series hybrid of a hybrid. The engine runs a generator to charge the batteries (series mode), but will also run in a parallel mode through a weird (to me) planetary transmission set-up, in conjunction with the motors.
  7. Apr 2, 2017 #6


    User Avatar
    Gold Member

    The 53[?] kW generator is also one of two electric motors.
    I'm guessing if it were just a generator, you would be correct.
    But since it's also a motor, bigger is better.

    I've never read up on the system before. Incredibly complicated, IMHO.
    I read four different articles, and I'm still not sure how it works.
    Mostly because everyone gives different numbers, and no where did I read if all three motors combined to yield the ≈282 hp. (Motor Trends combined numbers)
    Though Green Car Reports seems to imply all three motors are used in the last hybrid mode, but not exactly how.

    2016 Chevrolet Volt Powertrain: How It Works In Electric, Hybrid Modes [Green Car Reports]
    Describes the 4 driving modes.​

    2016 Chevrolet Volt Dissected: Powertrain, Design, Chassis, and More [Car and Driver]
    Combined electrical power stands pat at 149 horsepower

    The two electric motors are slightly closer in power (one’s still stronger, 117 hp to 64) [= 181 hp electric]​

    CHEVROLET VOLT - 2016 [Chevrolet]
    Electric drive: Power (kW / hp): 111 / 149
  8. Apr 3, 2017 #7
    Thanks, I'll look through those links, I think I've seen a few of them.

    Agree, it seems complex and the sources I've seen so far are somewhat confusing with seemingly different info. And yes, that generator may be used as a motor too (I'm not sure, will try to check that), so maybe the 53 KW is a motor spec and may or may not be used at that level as a generator?

    When I have time, I'll also try to check out the other plug-in-hybrids that can work over the full driving experience in EV mode. I think some limit their EV mode to below X speed, etc.
  9. Apr 3, 2017 #8


    User Avatar
    Gold Member

    They seem to be working on a problem I noticed a few years ago:

    Cost of driving: electricity vs. gasoline [PF]
    Younger Om
    Jul 6, 2008 10:02 am PDT
    There were several apples to apples comparisons at the show, but I'll chose[sic] the co-chairman's 1975 [homemade e-]Porsche 914.
    190 mpg at 20 mph
    200 mpg at 60 mph(don't ask me, you run the numbers)

    Younger Om's question; "Why did he get better mileage when going faster? hmmmm.... Would having two e-motors, one suited for cruising, and one suited for acceleration, mated in some weirdly engineered fashion, be more efficient?"

    Volt's answer; "Yes! And don't ask for the algorithm regarding the mating to the ICE, as it's really complicated. And also, don't ask why we didn't give an option to add all the hp together in the end, as, well, we're not old motorheads."
  10. Apr 3, 2017 #9


    User Avatar

    Staff: Mentor

    Ever drive a mountain highway? What does a Volt do if you need a steady 100 kW for 20 min?

    I've got a couple in mind; maybe we should calculate what is required to drive out of Yosemite, for example?
    Last edited: Apr 3, 2017
  11. Apr 3, 2017 #10
    I think the Chevy Volt does have a "mountain mode", where instead of letting the battery run down, and then go into a more direct ICE -> wheels, and maybe a little charging, it maintains (or attempts to) a full charge on the battery, for reserve.

    That would help, but I imagine the 50 mile EV range of the Volt would be far less doing a mountain climb - maybe 15 miles? A boost of 25 HP from the ICE would increase that, but I really can't say how much. I'd guess, since 100 HP cars can make it, a constant 25 HP ICE might add ~ 25% range? So maybe 18-20 miles uphill? That's a SWAG for the most part of course.

    How long do these up-hill climbs last? I have very little experience with mountain driving. I was debating if this thread should go into the "Electrical" sub-forum or the "Mechanical" sub-forum. Now maybe it belongs in the "Earth" sub-forum! :wink:
  12. Apr 8, 2017 #11


    User Avatar
    Gold Member

    In my below case, 75 minutes.
    Coastal trips vary, depending on who I intend on visiting. (family, friend, foe)


    Fun mathematical project. :smile:

    ps. I've only lived in upper Illinois once, about 40 years ago, for about 2 months. I never realized your state was so, um, flat.
    I have trees in my backyard that are taller than your state's highest prominence: 95 feet?. [ref]
    pps. I wonder if roller skates were invented in Illinois?
    ppps. Good Lord, you live in a flat state.
  13. Apr 8, 2017 #12
    Yes, in general, any slight incline we go encounter is just rolling farmland, and then you go back down the other side a minute later. Nothing challenging at all. But a bit of googling indicates that some other states are flatter, namely Florida. And then you get into definitions of measuring 'flat' - it gets involved. I don't think your 95 feet number captures it - I found:

    So state-wide, 1235 − 279 = 956 feet. But anyhow...

    Thanks, good example. But it still seems pretty tame for my hypothetical 25 HP hybrid. Dividing it up into a section for each change in required HP, you go for:
    ~ 16 miles @ 10.9 HP, then
    ~ 13 miles @ 17.5 HP ....

    A 25 HP engine could be supplying this to the wheels, and topping off the batteries as well, if you didn't leave with a full charge. Then you hit the challenge:
    ~ 14 miles @ 35.5 HP, then
    ~ 7 miles @ 26.0 HP ....

    But getting 10.5 HP from the motors to assist the engine for 14 miles shouldn't be a problem. The Chevy Volt should be able to drive the motors at ~ 25 HP for ~ 50 miles. Then just a small amount of electric motor boost for the last 7 miles.

    And in "mountain mode' (attempt to keep the batteries at full charge), there is enough excess HP in the first segments to provide charge for the last 2 segments. An average excess of ~ 11HP for the first 29 miles would offset the draw required at the end. You wouldn't even deplete the battery on that run, you could end up charged higher than you started. But any sudden acceleration like passing would eat some up kWh from the batteries, and probably not recovered by any regen. But still seems like margin under a tough scenario.

    Do those estimates seem ballpark correct?
  14. Apr 9, 2017 #13


    User Avatar
    Gold Member

    They seem reasonable.
    But I'm still wondering what the purpose of this thread is, and what is your ultimate question.
    Is it; "Why is the ICE motor 4 times bigger than it needs to be?"?

    If that is your question, then I would look at lifespans for motors that run at maximum rated capacity: Outboards. (Also airplanes, but I don't know anything about airplanes. I only know about cars and boats.)

    They have lifespans measured in hours. Interpolating the "general internet consensus" on their lifespans, along with how fast my outboard used to push my boat, I come up with a lifespan of 38,000 miles. Not many people would buy cars if they had to replace or overhaul their engines every two years. You are correct, that most cars on the road are barely idling compared to their potential, but that's why some of them last 300,000 miles before they need their first overhaul.

    Boats can get away with it, as they are generally used infrequently enough to last for decades.
    My outboard motor is as 57 years old, and were I to get around to fixing the spark problem, I'm sure it would last many more years.

    The Life Expectancy of the Marine Engine [boatsafe]
    The average marine gasoline engine runs for 1,500 hours before needing a major overhaul.
    What's the life expectancy of my car? [nbcnews]
    You should be driving down the road with a big smile on your face, because you certainly got your money’s worth from that Camry. Consumer Reports (www.consumerreports.org/) says the average life expectancy of a new vehicle these days is around 8 years or 150,000 miles. Of course, some well-built vehicles can go 15 years and 300,000, if properly maintained.
  15. Apr 9, 2017 #14
    OK. I understand and agree with what you are saying about engine life at full power. But look at it this way, and then I'll tie this back to "Secondarily" segment of my OP:

    Yes, a +100 HP car engine won't last long at constant full power (overheat, over-stress, wear), but can run for extended periods and provide long life at that ~25 HP range needed for highway speeds. Now consider the 100 HP engine in the Volt, if it were limited to that ~ 25 HP and a narrow RPM range, and we call on the motor/battery for added boost for acceleration. It would never see the stresses of a 100 HP engine, or need to rev to 6,000 RPM. So the connecting rods, bearings and many other components could be lighter, component could be designed for a more constant 1600-2400 RPM or something. But it would still need to be designed as a 25 HP engine that can run at a continuous 25 HP - so I would not expect it to necessarily be 1/4 the size/weight. It wouldn't be like an 6 HP engine that you occasionally push to 25 HP, it would be a full 25 HP engine.

    And that extends to a transmission for the engine-to-wheels. A CVT that would never 'see' more than 25 HP, that wouldn't 'see' the high torque of acceleration (that would come from the motor/battery). Seems to me a CVT like that would be a pretty small, lightweight, and relatively cheap component. So the engine could run at a fairly narrow power/RPM band. And it could have a higher compression ratio (and therefore higher efficiency) than an engine that needs to accelerate a car. Compression ratios are limited by needing to avoid knocking under acceleration, but then the compression ratio is lower than desired at cruising. Higher efficiency means a smaller engine (relatively). These sorts of compromises would be minimized in a narrow power band engine. You could probably also do away with other complexities like variable valve timing - one condition would probably be OK for a narrow RPM/power range. That saves size/weight/cost as well. And maybe even down to details like the piston rings - I imagine those rings need to be a tight enough fit to contain that max horsepower burst w/o excessive blow-by. So cut the max power by a factor of four, and can those rings be a bit looser, with less friction, heat, wear, etc? Details add up?

    So if any of that makes sense, why don't they do it that way? Obviously, I could be way wrong about some assumptions. Or maybe they just don't want to commit that different of an engine design? But that whole motor/generator/dual-planetary-gear system is very unique. So I don't know.

    As a related aside, here's a true/pure serial hybrid design I've been following, that I think has a lot of merit, for a niche (but significant $) market - a constant speed/power gas/diesel micro-turbine keeping batteries charged in garbage/delivery trucks. Wright was a co-founder of Tesla:

  16. Apr 10, 2017 #15
    What seems overlooked here is how the automotive world, and all consumer products, are "specified" - is reality there are essentially 2 specs - 1) The design specification and 2) The marketing spec ( what they sell to you)

    The 100HP gas engine - would have been chosen to ensure customer satisfaction under all conceivable operating conditions, and a key one of these is to offer acceptable driving when the battery is dead, or the electronics drive system suffers a major failure (inverters in my experience do not kinda-break, they are nearly 100% or they are 0%) - think limp home mode. Automakers will do anything in their power to prevent a completely dead vehicle, once a car "dies" on you you will doubt it forever, and with a "new" technology they can not commercially afford to have any stories of "Chevy Volt leaves customers stranded"

    Trying to look at the marketing spec and reverse engineer the decisions in the design spec from a X + Y = Z perspective is almost pointless - you do not know what they wanted, expected or what they know.
  17. Apr 10, 2017 #16
    While I agree that marketing/perception are an important part of products and often rule over engineering decisions, I just don't see the fit here.

    Full Battery Electric Vehicles are being sold and people accept that a dead battery means a dead vehicle, and they buy them. We buy non-hybrids, and a dead engine/trans means a dead car, etc. We don't have redundant systems in our cars (except for brake master cylinders? maybe a few other key areas). Even with a 25 ~ 30 HP engine, I think they could design a 'limp home mode' if the battery/motor were dead. I did some very rough calculations a while back, and I think the acceleration from a 25 ~ 30 HP engine in a mid-sized car would be roughly equivalent to the acceleration of a fully loaded semi-truck. Not acceptable for normal use, but I guess you could get home. Or maybe the 6:1 CVT range of ratios is far less than a semi? But I don't think it matters, I really don't think the larger engine is to compensate for a dead electrical motive system.
  18. Apr 10, 2017 #17


    User Avatar

    Staff: Mentor

    As much as a half hour: Up to 5,000 feet gain in 15 miles for 30 min. I ran some numbers on a couple of scenarios though and they weren't as bad as I expected. Trouble is, they aren't constant climbs, but have tight turns and both elevation gains and losses -- so they are much worse than the smooth distance/elevation.
    A dead engine/transmission is a warranty-triggering failure. You're describing it as an acceptable built-in feature.

    Americans in particular have a very low tolerance for compromise. If you are Tesla and are ok selling 50,000 cars a year to people who accept compromises, ok, but GM isn't Tesla: they (ultimately want to sell 500,000 cars a year to people who consider them to be interchangeable with their gas cars.

    I'm a fan of Elon Musk, but I think his investors are going to lose their shirts. A car company is not a tech company. Tesla needs to sell cars by the hundreds of thousands and turn a profit(!?) in order to be worth the valuation they are getting now.
  19. Apr 10, 2017 #18
    Only if the car would actually not be able to perform in those scenarios. A few posts back, I took a stab at the scenario OmCheeto presented, and if my numbers are reasonable, it didn't seem that challenging for a 25-30 HP engine. And if you haven't depleted the battery (and we hadn't in that scenario), you still have the electric motors to kick in for those tight curves and higher than average climb segments (which will be offset a little by the lower than average segments).

    Musk is clearly a super-bright guy, and able to execute. I'm amazed that they have several models of cars, and shipping in some volume (not just DeLorean level shipments). Breaking into that industry beyond a specialty car (which can get some exemptions, and/or tons of re-use of existing cars - somewhat like their Roadster was based on a Lotus Elise chassis) just strikes me as mind-boggling difficult. I'm hesitant to downplay what he might be able to achieve with Tesla, but I agree, the numbers just don't seem to make sense. I just can't see it holding that valuation, unless it all comes from other products like the batteries themselves, solar panels, or maybe something new.
  20. May 8, 2017 #19
    Following up, I recently viewed this video which I found interesting. Whereas the Chevy Volt has the dual planetary transmissions, the 2017 Honda Accord Hybrid has... NO transmission! The ICE is engaged only when you would normally be in top gear, through a clutch. In other modes, the ICE is connected to the generator only, and the electric motor is driving the vehicle.

    Interesting that two such seemingly extremes are used in similar vehicles. Edmunds rates the Honda as "Best-in-class fuel economy", but it is not a plug-in, so you won't get those cheap electric-only miles.

    I also found it interesting that the Honda has a 1.3 kWh battery, but the larger traction motor can also draw at least 30 horses (23 kilowatts) from the hybrid battery during acceleration. That's a C rating of ~ 18x, but probably limited to ~ 10 seconds of acceleration or so?

    I didn't see much info on EV mode, but I'm guessing you might get ~ 10 minutes if you were doing light driving, average 10 HP? I imagine some modders will come up with a trickle charger for that 1.3 kWh battery - why not leave home with a full charge and get at least a few miles with cheap electricity?
  21. May 8, 2017 #20
    Did they design this engine from scratch solely for the Chevy Volt application?

    My guess is that this engine is a version of their most effective short-term option for this application that they had at the time. It is based off of the GM SGE engine platform. If you're going to design an engine from scratch, you want it to be useful in as many applications as possible. Not just one single application that doesn't even sell very well.

    The engine requirements probably were focused around
    - Cost (economy of scale plays a big role here)
    - Sufficiently mitigated NVH to please the target market
    - Ability to produce a requisite torque curve to work well with the generator.
  22. May 8, 2017 #21
    I had thought about the idea that they wouldn't want to commit to a unique engine design (the Chevy Volt is still pretty low volume), but they did create a unique transmission. Though fewer variables with a transmission compared to an engine. But the engine would not need to be totally unique, I would imagine a re-tweaked version of an existing block, etc.

    But I think all the things you mention combined probably pretty fully explain it. As far as NVH, I suppose an engine running at a constant speed throughout would sound 'funny' to people - more so than a CVT, as at least that is responding to throttle pedal input. An ICE running mostly ON/OFF mode might be pretty weird to a lot of people. Like if you are crawling in traffic, and suddenly the engine starts and revs up to 2,000 RPM to top-off the battery, that could be unsettling.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted