Hybrid vehicle: series or parallel drive. Which is better? 1

[PatentPro]

PatentPro writes:

I believe that the era of the hybrid vehicle has arrived. Internal combustion engines always operate inefficiently under high loads near the engine's rated power: eg climbing hills, accelerating. But an electric motor, properly sized, can perform efficiently under these circumstances.

The Prius is a parallel drive system. The gasoline motor and electric motor can operate in parallel. The Volt is a series drive system: the internal combustion engine only sees the generator.

Who's right: Toyota or General Motors?

 

[dgallup]

Your premise is wrong. ICE operate at or near peak efficiency at high loads. Turbo diesels in particular have optimum efficiency at full load. Gas engines not quite peak because of fuel enrichment at WOT.

Electric motors efficiency is not as load dependent as ICE but their energy source is quickly depleted at high loads.

Neither the Prius nor the Volt is intended to be operated at high loads, they both suck compared to ICE powertrain in terms of performance.

 

[BrianPetersen]

Who's right, "it depends".

If you accept that a fairly substantial amount of the energy being supplied to the vehicle is via the combustion engine rather than via an electrical plug, then you have to realize that if the combustion engine is going to be running anyhow, a gear-to-gear transmission is more efficient than an alternator, rectifier, battery, inverter, motor, and gear-to-gear drive. So for rolling down the highway, driven by the combustion engine, the parallel solution has the potential to be more efficient.

If you spend your time stuck in city traffic, then a solution that minimizes use of the combustion engine MAY be more efficient.

If you spend your time stuck in city traffic in the middle of a cold spell or a snowstorm, where's the interior heat going to come from? Most of them run the engine to achieve that. In a pure electric vehicle ... good luck with that.

 

[JSteve2]

The parallel provides greater operational flexibility. Also, the series requires a fully capable electric motor rather than a potentially smaller one that is not required to deliver peak motive power output. However, the parallel is more complex, and introduces the possibility of running the ICE at non-peak operating points (not required, but once you have the control option, how can you help yourself?)

If the application is readily amenable to all-electric with little help from the ICE, then the series solution can make sense. If the ICE contributes a large portion of the total motive energy, then the parallel is probably better (i.e. what BrianPetersen said).

 

[ivymike]

parallel ones always won the student competitions.

 

[GregLocock]

Jsteve2- the max efficiency of the Prius engine is 38%. Its optimum delivery curve is above 30% for the entire power range. During a real test it isn't always on that curve, but it is sufficiently close to it for much of the time.

In comparison the Volt will run at around 5 different setpoints. I find it hard to believe that they'll be able to improve the efficiency at those 5 setpoints significantly more than if they just ran it along a curve.

I think Volt is an interesting experiment, but quite why they need to build an experiment as a production car is more to do with politics and marketing than engineering. However, for people who obsess about running costs, as opposed to total cost of ownership, and who drive less than 30 miles a day most days, it seems like a good move.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[JSteve2]

GregLocock - since we're talking about specific vehicles, I think the Volt is built that way for one of two reasons (I have no inside knowledge on the matter). First reason, they may be trying to avoid the complex 3-headed monster transmission that Prius uses. The second reason is that they may be building the Volt as a precursor for a completely electric car, which cannot rely on an ICE for direct motive power, of course, so they have to take the ICE out of the motive loop. It may also just be politics or marketing, but I'm trying to give them the benefit of the doubt on that.

I agree with you that, almost all the time, the parallel gives you superior results from a total efficiency standpoint. In fact, it could always be at least equal depending on what your design goals were.

 

[DanEE]

Quote: The Prius is a parallel drive system. The gasoline motor and electric motor can operate in parallel.
----------------------------------
The discussion might be a bit more on a sound base if your opening statement was a full statement of the Pruis system capability.

The Prius is a capable of a balance of both Parallel and Series operation, the mix of the two controlled by the power balance between MG1 and MG2.. and the use of a very simple, single planetary gearset that integrates the torque split.

If you take to time to find read the concept description in Toyota's patent of the HSD control algorithms you will get a good feel for how dynamic the control is and the use of parallel/series torque split to keep the overall system operating in at peak energy efficiency over a wide range of road conditions. The description is 92 pages long, the longest I've ever seen of any patent I've read.

Sorry don't have a link for it, but online patent data bases are quite easy to search.

DOE, Oak Ridge National Labs has also produced probably the most in depth, independent analysis of the details of the Prius HSD drive system here..

http://www.ornl.gov/~webworks/cppr/y2001/rpt/121813.pdf

A complete read might serve as a good basis for further debate on the subject.

 

[DanEE]

Hybrid System mode comparision by Toyota.. Agree, disagree on what basis?

http://home.4x4wire.com/deddleman/photos/Parallel_Series_Modes.jpg

above (fair use) extracted from this document:

http://www.toyota.co.jp/en/tech/environment/hsd/pdf/a_guide_to_hsd.pdf

 

[JSteve2]

DanEE,
The Prius is not a true series/parallel, in the sense that although the hardware is present, the controls NEVER run in true series mode (i.e. ICE charging, only the electric motor driving the wheels). When the ICE runs, among other things it is always driving the wheels if the wheels are moving. I've studied this thing closely, and I own one (and I love it).

As far as the benefits matrix, I largely agree except there is no reason why "series/parallel" is superior to "parallel" for continuous high output. As far as the Prius specifically (rather than a generic series/parallel), it is no different than the others for idling stop (as the ICE shuts down for this mode), it's only slightly superior for "energy recovery" (the only difference being the engine can charge and drive at the same time, giving one operating condition where the engine has more operating points available to it).

I assumed by idling stop they mean the coast to the stop, if what they mean is the stopped period itself, the series/parallel should score identically to the series (maybe even better if the presumably smaller ICE is more efficient - and they presume the ICE is smaller on series because they ding the series for continuous high output).

I'm not even sure what "high-efficiency operation control" means, except that maybe they mean things like through-the-wheel energy recovery can be a bit complex in a parallel configuration.

 

[ivymike]

this might be the first time... but I think I'm inclined to agree with JSteve on this one.

 

[Thecardoc]

JSteve, at the risk of being picky, reverse in the Prius is completely powered by MG2. If the ICE is used for generation with MG1 and the vehicle is in reverse you do indeed have series operation. We classify the Toyota, and Ford systems as full Series/Parallel hybrids.

 

[JSteve2]

@ Thecardoc:
That's fair, I have no quarrel with that. In most operating conditions, it's not significantly different than parallel, but as I said the hardware is present.

 

[duck0601]

In the interest of inclusiveness, The GM Two-Mode system is full Series/Parallel as well.

 

[DanEE]

One attribute that has been left out of the discussion to this point is that the Prius planetary gearset (sum and difference engines, as they were called and used as such in early analog "computers" in the 1800s) is used to provide CVT transmission function. Very simple mechanics, but complex computer code to control in a seamless manner.

A pure parallel must still have some type of conventional transmission adding weight and cost. The Honda which is pure parallel uses a CVT belt transmission.

Regarding parallel versus series, it is well established that even the best technology MG/Inverter/MGs can not match the power transfer efficiency of mechanical gearsets.

 

[GregLocock]

"it is well established that even the best technology MG/Inverter/MGs can not match the power transfer efficiency of mechanical gearsets. "

That's just a money problem. Best MG is 98.4% (and we could do better), best inverter is basically ((system voltage -1.2V)/system voltage), less a tiny bit, 99% is not unachievable. Admittedly even just a 5 kW system to those specs would cost > $20000 as it would be handbuilt.

Mechanical gears basically lose about 2% per pair, more if sliding takes place.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BrianPetersen]

... and a weight / inertia problem. All motors and generators that I know of that have such high efficiency ratings, have lots of copper, iron, and in some cases rare-earth metals in them relative to their power output. The efficiency may be good but the power/weight ratio is not, in these cases. I'm thinking of the AC traction motors used in locomotives here. In those, you want the locomotive to be heavy, to be able to pull the train. In a car ... not so much.

 

[GregLocock]

Agree. Tho I think worrying about the motor weight and ignoring battery weight is a bit odd!


I should have said 1% per gear pair I think in the previous note.






Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BrianPetersen]

The locomotive AC traction motors that I mentioned weigh something like 3 tons for around 1000 hp. Lots and lots and LOTS of iron and copper in those. Divide by 10 to get a car-engine-sized scale but with the same amount of iron and copper, and it's still 600 lbs, which would be a rather large percentage of the weight of the car or the battery!

 

[wutzu]

Quote: parallel ones always won the student competitions.

At Formula Hybrid 2009, Texas A&M won with what was essentially a gas FSAE car that also had a minimal HV electric powertrain to meet rules. It dominated the competition because it was well-designed and well-tested. More broadly, of the cars that raced at FH09, almost all were parallel (probably 20 parallel, 5 or 6 series), and probably a similar ratio at earlier competitions.

Having parallel drivetrains helps with reliability, which is huge, but as student electric propulsion knowledge catches up and the competition fuel allocation gets cut further (-30% for 2010), I don't think parallel drivetrains will stay competitive.