Why not Prius MPG??

[George S. Bower]

Before I get into the technical discussion I would like to say that this is NOT an anti Volt, pro Prius thread. It is simply a technical discussion delving into posed question. RPhooper posted a good thread today about how to improve the Volt for Gen2 from an MPG and cost point of view,and, in order to come up with some good suggestion we must first understand why the Volt does not get Prius MPG in CS Mode.

An excellent place to start is with a post from WOT as follows. I just went back to this post and did what he suggested. ie open 2 windows and pull up the interactive planetary animation and nomograph. Relabel the sun, planets and ring in the proper Volt nomenclature and start playing with it.

WOT Quote:

To more easily visualize the effect you can actually use this Prius nomograph animation
http://eahart.com/prius/psd/
For it to make sense you will need to appropriately "rename" the "stick" shifters and RPM readouts:
Blue (MG1) becomes MG2
Brown (ICE) becomes OUTPUT (to final drive)
Yellow (MG2) becomes MG1 and/or ICE driven speed of the ring/internal/annulus gear (and dependant on clutching factors)
(you can even use a bit of masking tape on your monitor to label as such to keep it straight)
*also keep in mind the ratio of the Prius is 4.66:1 but the 4ET50 is geared a bit higher at 4.22

Starting with everything flat & level you can slowly raise the blue shifter slowly to 6500 rpm by using the brown OUTPUT lever (previously ICE) to simulate normal EV mode in CD.on the Volt this would equate to an output speed that delivers ~70mph after the final drive and wheel circumference is through with it.

To go any faster in order to mitigate the inefficiency of overspeeding MG2 (Blue) we can first uncouple the ring/internal/annulus gear from it's stationary position (locked to case by C1) then add a variable reactionary influence from either MG1 (when C2 only is applied in CD mode) or ICE (when both C2 and C3 are applied in CS) or BOTH.

To observe it's dramatic effect (again at 4.66:1 not 4.22) slowly drag the yellow lever upward to until it's RPM is 1508. While doing so, MG2 RPM is quickly decreased to 2653 rpm yet the output speed does not change! (again 2653 is not quite accurate, the Volt would be 7.6% faster due to the higher gear ratio but close enough) After that you would merely move the speeds of MG1/ICE (both assuming CS) and MG2 in unison up to 100mph (something this nomograph animation doesn't permit)

It's not perfect but you get the picture.It's NOT a 2-mode (OR a Prius) but a similar use of variably split ratios of planetary gearset

Of course a fringe benefit of this 3 way coupled "output split" mode is any regenerative opportunity in this mode could be potentially created and maximized by using either 1 or both MG units depending on the inverter controlled pathway back to the HV battery.

After playing games with engine speed, vehicle speed and loading I have concluded that the existing planetary gear setup DOES allow the ICE to operate with enough freedom to put it on it's best sfc point for a given load. It does this at the expense of driving one or both MG's to a slightly off design condition, however, the motor maps I've seen have a fairly wide efficiency band as long as one does not drive the motors to an overspeed condition. So, IF what I just said is true, then why can't the Volt achieve Prius efficiency. We essentially have eliminated a poor ICE match as the culprit. Sooooo, then why??

I don't know and I must say I am even more puzzled about this than before.

Ideally we would have the guy at GM that runs the off design computer model of the Volt explain it. Failing that I would first ask WOT, or anybody else on this forum to present a logical reason.

I suspect that it is not just one thing but a multitude of things. First on the list in my mind is still that the Volt is operating in pure series mode for much of the time and therefore is suffering from conversion loss penalties (although this should be mitigated with the output split mode). Second we know that the higher wt of the Volt has to have an effect as Tom's sensitivity coefficients have shown (even though the Volt has a superior regen system). Thirdly: Is the ICE cycle efficiency a culprit (as opposed to ICE match point on it's map), I guess it does NOT have direct injection but I would assume it is using VVT so it should be able to match the Prius ICE efficiency.

Why?????

[George S. Bower]

I just discovered an interesting condition w/ the nomograph. At lower vehicle speeds and high loads, if you try to increase ICE rpm and load to take as much power as possible from the ICE you start driving the main traction motor backwards and reach the rpm limit. I wonder if this is a limiting factor??

[rphooper]

George,
thanks for mentioning my thread, this question has been bugging me too, but I wanted to keep my post in a positive light.
I cant begin to follow your comparison of the Prius and Volts gears, it was all a bit over my head. My view is that the Volt in CS mode is less
efficient then the Prius for two reasons, both related to it being an EV first and a hybrid second.
First off weight, it has a 16kw battery to lug around in order to drive 40 miles without gas, thats a lot of weight most of which serves no purpose once its in CS mode.
Second the GM engineers stated from the beginning that they wanted the CS mode experience to be the same as that for EV mode. In EV mode the Volt has a lot of low end torque and and is a powerful ride, blowing away the Prius. I believe in order to keep the Volt's EV and CS driving experience seamless the Engineers sacrificed fuel economy for power, quietness and a smooth transition between modes.

I believe that GM could build a pure hybrid version of the Volt, where the CS mode fuel economy is the most important factor, that would get much better CS mode fuel economy, but would have less power, a rough transition between modes, and the engine coming on at odd times.

I also believe we will see third party tuning services for the existing Volt, that can significantly increase its gasoline mileage, at the cost of performance and ride.

[George S. Bower]

[QUOTE=rphooper;44290]George,
My view is that the Volt in CS mode is less
efficient then the Prius for two reasons, both related to it being an EV first and a hybrid second.

rp,
There is something in your statement that rings true. We know that the Prius is limited in pure EV mode because it drives MG1 to overspeed. Similary the Volt's transmission, being optimized for EV, has limitations in power split mode. Perhaps something similar to the condition I posted in #2.

[G35X]

Since the ICE of the Volt is operating with the throttle fully open most of the time (http://gm-volt.com/wp-content/upload...vior-graph.jpg), you cannot improve its efficiency further unless you turn it into an Atkinson-cycle variety. I agree with rphooper that if you want to make the Volt behave like the Prius, first of all, reduce its weight. I think a version with 8kWh battery with 20-mile EV range is viable from the marketing standpoint… lower cost, better CS mode mileage.
On the other hand, if you want the Prius to behave like the Volt, increase battery capacity, reduce the size of ICE and increase the size of the main motor. Mechanically there is no conceptual difference between these two cars… have three propulsion sources and combine their output for the best efficiency and performance. The only difference is philosophical. (Although, the Volt system is more complicated with two clutches.) The reason why GM did not let the main motor drive the output shaft directly is, I think, the cost and efficiency of the higher frequency/voltage power supply (and the motor itself).

[hermperez]

To elaborate a bit on previous replies.. the engine may not be as efficient as the one in the Prius.. once they release the compression ratio then we will know a bit more.. and the weight of course. A more efficient engine may have more NHV issues and since the engine is tied to the wheels then we cant have that.

[Cab Driver]

To elaborate a bit on previous replies.. the engine may not be as efficient as the one in the Prius.. once they release the compression ratio then we will know a bit more.. and the weight of course. A more efficient engine may have more NHV issues and since the engine is tied to the wheels then we cant have that.

They have released the compression ratio. It is 10.5:1 just like all other recent normally aspirated family 0 1.4 liter engines. See: http://media.gm.com/media/us/en/vehi...olet.tab1.html

With regard to weight, Car and Driver measured it at 3755 pounds. See: http://www.caranddriver.com/reviews/...test-road_test

I think the bottom line is as several of you have suggested; the Volt is several hundred pounds heavier and the engine / transmission has not been optimized as aggressively for fuel economy as the Prius.

We know that GM could have gotten significantly better fuel economy by using direct inject instead of port injection. They didn't because the significantly higher cost of DI would not ever be recovered in gas savings for the 80% or so of drivers who will use very little gas anyway. (It also would have added development time to incorporate DI.)

[hermperez]

Thanks Cab Driver, that settles it then.. its an official GM website. So its just a low rpm version of the Cruze engine and there goes all hope for high mileage in the CS mode.. the Volt will not be a do-all-beat-all hybrid.

[Koz]

This is the kind of thread I really enjoy. Wish I had spent more time understanding engine design but that won't stop my opionating.

Reasons for sub-Prius mileage:
1. ICE efficiency
2. Weight
3. NVH and "normal" driving experience concerns (?)

I think we should also establish what we believe the goals going forward should be. Since the gas mileage on a EREV40 should be about 90% highway, I think that CS mpg should be maximized for this driving: 15-25kw steady loads for long periods. For the ICE, I believe the priority should be:
1. Lower cost (won't help increase CS mpg but getting EREV into the mainstream is more important)
2. Higher efficiency (target minimum average 40% BSFC at steady 15-25kw car load)
3. Lower weight (575lbs GVW more than the Prius can't be ignored)
4. Smaller footprint (not directly a CS mpg saver but makes packaging easier and allows for smaller vehicle implementation)

Possible improvements for the ultra high efficiency branch of EREV (note, I think they should be developing a higher power, higher performance variant for the Converjes, Equinoxes, small pickups, etc.):
1. ICE
a. Kind (Atkinson, DI, HCCI, Deisel-compression reciprocating, Wankel, Atkinson-Rotary, Turbo,etc)
I. Atkinson- most promising from a current production/tech standpoint, currently implemented successfully in Hybrids, currently capable of 37+% BSFC, reasonable weight/size, reasonable cost, down scalable. Further BSFC improvements difficult.
II. Turbo- Could allow 2-cylinder ICE capable of 25KW without turbo assistance. High cost, added weight/size offset some of smaller core engine advantages
III. DI- good efficiency, current tech in production, down scalable, reasonable weight/size. Further BSFC improvements difficult, cost is high.
IV. Wankel- currently in production, excellent size/weight, potential for low cost in volume. BSFC is a big question for steady 15-25KW loads, although it is better than many presume, can it be made good enough? Reliability, newer tech and ocasional steady loads could render this a non-issue.
V. Deisel- well known, high production, very efficient, very reliable. Heavy, big, expensive.
VI. HCCI- very good efficiency, down scalable(?). not in production yet, high weight, high cost
VII. Atkinson Rotary- Excellent efficency, excellent size/weight, potential for low cost. Not in production, reliability unknown

2. Control can be augmented to include a "extended drive" (ED) mode. Setting this would adjust the efficient maps with the knowledge that you will be driving for extended periods at steady 15-25kw loads, instantaneous efficiency concerns would be minimized as well as NVH concerns.

3. Weight can be reduced in many ways beyond just the battery and engine. An EREV platform should be designed from the ground up. Low weight seats, lids, roof, skin, glass. Any weight reduction must be mindful of cost. GEN1 at the right price has huge market appeal, even with the current CS mpg.

I'm very puzzled as to why Atkinson-like tech was not implemented in Gen1. GM uses this tech in their two-mode, so they have experience with it. Does the advantage diminish when operated at WOT most of the time? Is there some other technical limitation or disadvantage?

RPhooper or another poster on his thread stated that a GM engineer mentioned that they are not working on ICE improvements for the next generation. I am very disappointed, if this is accurate. Hopefully there was miscommunication or GM is just being coy again.

[srschrier]

I tend to believe development teams at GM are already pursuing some of the ideas in this thread. Weight reduction, a smaller lighter battery pack. Maybe by Gen-2 or 3, assuming market demand sells enough Volts, we'll begin to see smaller lighter batteries with storage densities comparable to Gen-1's "Model T" pack.

Slightly off topic, a simplified driver interface would be welcome, anything that safely reduces the vehicle's cost. The digital instrumentation in Gen-1 Volt is impressive, but what for? Could a basic analogue instrument panel be offered as an option? An instrument panel without so many buttons, without the multi-layered touch sensitive screens, without iPod connections, etc., an ultra-simple instrument panel that provides the driver with enough information to drive safely. Gen-1's digital instrumentation may appeal to younger generations but the car's price is so high many cannot afford it.

Does the word (EREV) continue to be an accurate description of the Volt? It looks like the platform has evolved into a plug-in hybrid, a "super plug-in hybrid" (SPHEV). Even so, the Volt's arrived with more advanced engineering and will be in showrooms well ahead of Toyota's plug-in.