One speed/one load genset for electric vehicles 3

[icelander]

The general direction of combustion engine development is towards increased complexity to retain efficiency over a wider range of speed and load, with Daimlers DiesOtto as possibly the pinnacle. Going the other way, designing a combustion engine for one speed and one load would shed loads of compromise and complexity. Flow simulation now makes "easy" optimizing such an engine (presumably 2 stroke, valveless, minimum peripherals, cheap to produce) to the maximum efficiency/emissions point. A similarly simple and optimized single load/single speed (on-off) electrical generator would absorb the mechanical force, batteries serving as load buffers in vehicular applications. Since no major players seem to follow such a strategy the question is simple.
Why not?

 

[GregLocock]

Two stroke? So if you want it to be efficient you need direct injection - is that really a low cost alternative?

However, I do like the idea of a single point tuned system, if Ed Danzer is still here he could fill you in on a free piston generator.





Cheers

Greg Locock

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

 

[SomptingGuy]

If your vehicle is a locomotive, you are describing standard practice.

- Steve

 

[icelander]

Thank you for your response. These forums are without equal in quality of information, I believe.
Standard locomotive practice, precisely. However, the diesel/electric locomotives I remember used output control through the conventional diesel engines, rather than this proposed simple, optimized "single point tuned" on-off engines/generators and output control on the electrical side. Some ships used for other than constant speed transport (research vessels, cruise ships...) have multiple gensets and turn them on or off as needed, keeping those running at optimum load points. A large electric motor is direct-coupled to the propeller obviating the need for a gearbox.
In "Chevy Volt" type cars, or urban buses, the logigal route has always seemed to me to have a minimum sized chemical energy engine (IC) supplying just enough electrical power for a reasonable maximum travel speed for a reasonable time, minimum sized batteries/capacitors (with plug-in charging) buffering and storing to keep acceleration at acceptable levels.
This could keep weight and bulk, if not cost, of the whole system down to the current levels of straight IC powertrains, but with efficiency increased beyond even straight ICs theoretical possibilities. IC powertrains are neccessarily bloated in size and weight by consumer requirements (private cars) or necessity (urban buses) to supply seamlessly variable power from zero to bursts of multiple that required for travelling. In gensets for hybrids today, the engines are generic ones designed for wide range prime moving and keep the associated complexities and compromises. Costly, heavy, bulky, and unnecessary compromises for one load/one speed generating.
Toyota just showed a 4 seater family car (all carbon fibre, 400 kg, not in any showroom soon) with a 500cc engine, series hybrid coupled, but no sign of "single point optimization". Actually, in years of searching I´ve never found discussion of the concept on an engineering level, let alone researched. Surely, I have missed something...?
Free pistons and linear generators are ideal in concept - but why do they not seem to materialize? Or have they? If so, their popularity could stand some improvement. There must be a reason/reasons for this, besides the power of convention?

 

[Bribyk]

The trend right now seems to be building lighter cars, and rightly so. I came across a fuel efficiency comparison between 15 cars from 1992 and their current models. All but two of the '92 models had better fuel efficiency than the 2007's and most '92s were 500-750 lbs lighter. They had few or no power seats/windows, heated cupholders/seats/mirrors, DVD/Nav, larger alternator/battery, NVH insulation, etc... or inflated horsepower marketing numbers.

 

[BrianPetersen]

One thing to keep in mind is that with single-point tuning but inherently variable loads on the vehicle, you are going to have some losses associated with charging and discharging the energy storage system (regardless of what that happens to be). I suspect that an optimum would be found involving "not quite single point tuning", in which you accept some losses of running the engine off-design in return for (under appropriate load conditions) not having to charge and discharge the energy storage system.

The general pattern has been that "series" hybrids can (if properly designed) be better in pure city driving, but less efficient than gear-to-gear mechanical drive in pure motorway driving.

Pure series operation requires that the electric drive be capable of full output power for acceleration. The halfway-house used by Prius etc requires the electric drive to be only big enough for moderate acceleration and constant low-speed driving, resulting in much smaller motors, power inverters, and battery amp loading. Right now, there are cost, weight, size, efficiency, and battery-life tradeoffs with designing the electric drive to provide the full output power for maximum acceleration. Prius, for example, gets away with a relatively compact battery pack because it doesn't get called upon for providing all of the vehicle's acceleration.

 

[BrianPetersen]

One other small addition. Charging and discharging a battery is (typically) around 85% efficient provided that you don't call for currents approaching the maximum capacity of the battery. If you take a look at the BSFC chart of a modern turbo diesel engine (which, by the way, I think would be the most appropriate for that type of application), and you find the maximum efficiency point, and you identify the area around that point which gives you a 15% leeway, you'll find that the range between maximum and minimum power under that constraint is quite wide.

 

[icelander]

Thanks for the charge/discharge loss figures, they go some way towards answering the question above.

Brian Petersons point about leeway around BSFC is most valid too. However, the core of this single-point idea is taking the best BSFC of the best modern turbo diesel and besting it by some margin, and, simultaneously, dropping considerable complexity/expense (no wastegate/vanes, no valvetrain...) but getting higher power-to-weight than even BMWs currently brilliant best. Packing the best known optimum fluid dynamics and combustion processes in simple, conventional mechanics (piston, cylinder, crank). The result being a cheapo assembly with record efficiency. No?

Incidentally, the very best modern TDi engine has quite a way to go to get the sort of BSFC a "simple", low parts count Sulzer 2-stroke marine engine churns out round the clock. The typical TDi is a rather expensive thing too, and is elvolving further down that road.

The Sulzer can perhaps be classified as a single-point engine - anybody with knowledge there to share?

(Perhaps the reason for this thread should be given. I run a design bureau creating niche vehicles and am involved in forming a proposal for the recreation of public transport in our country´s only city. I firmly believe the "bus" must be renewed in its entirety, including customer interface, image, technology and environmental efficiency (perceived and real). A practical route is to carefully mix good existing technologies rather than "invent". Reassemble other peoples good work. Always looking for the better mousetrap - that works.
Electric buses are not new, and neither are chemical energy range extenders to save battery weight. A hydrogen fuel cell city bus programme just closed here and no follow-up is in sight. For good reason.)

 

[CCycle]

Icelander,

Agreed, good thinking right down the line.

But set a high goal. Say we were successful in making an IC engine with 50% thermodynamic efficiency. Now what do we have? Well the cars now getting 20 mpg will get low 30s mpg. We have done nothing to solve the real problem. We have billions of people in developing nations that want to get on board with cars and energy rich life style. The world just can not supply the fuel nor handle the pollution even with engines getting 50% efficiency.

What we need is a pluggable hybrid with enough battery capacity to go 100 miles. This would shift 85% of transportation energy off anthropogenic carbon fuel and on to the electric grid where we have several renewable options such as solar, hydro, biomass and nuclear. Having this, the stand by engine becomes quite simple. It would be small and run at its optimal efficiency. These engines would preferably deliver their power as electricity, not shaft power.

The problem we have with the hybrids today is the same as they had with the electric cars 100 years ago. We do not have an adequate battery. But we are getting close. The Lithium battery looks great. The engineering challenge is to make it low cost and so it doesn't burn.

I just returned from a sitting in the drive up lanes of a local bank. Of the six or eight cars waiting, I was the only one that turned off my engine. As I sat there with my eyes burning from exhaust fumes I thought, this is absurd. An idling engine is 0% efficient.

 

[BrianPetersen]

Tracked down some more info. Re the comparison between an auto engine and a Sulzer marine engine, do not forget to account for the difference in scale of the engines. It's easier to get the best operating point of a large-displacement low RPM engine to have a better BSFC than for a smaller displacement higher RPM (and therefore lighter) engine. As of a couple of years ago, the VW TDI 110hp engine at its best operating point have the lowest BSFC of any auto engine. Truck engines can have a little better BSFC - but they're bigger, and a lot of that is BECAUSE they are bigger. The level of technology is about the same.

In this link, the BSFC map of a 1.9 litre 90 hp VW TDI engine can be seen:

http://forums.tdiclub.com/showthread.php?t=200427&page=2&highlight=BSFC+map

Note that the best point is 197 g/kWh. If you allow a 15 percent deviation (to simplify drivetrain design) and follow the 230 g/kWh line, max load can be as high as 3500 rpm and 12 bar BMEP, and lowest load can be as low as 1000 rpm 5 bar BMEP, although in practice this engine will not be "happy" at such low RPM; 1500 rpm 5 bar BMEP is a more practical minimum that won't feel like "lugging". Still, this gives a range of more than 5 between max and min power output that are within (roughly) 15% of the best-efficiency operating point.

If you have a tailored application, it might be possible to get a few points better BSFC by eliminating the off-design conditions, but for one thing this will be a difficult challenge (VW TDI engines are regarded as being close to state-of-the-art in this respect) which suggests that the margin for further improvement is probably low for an engine on that scale. For another, if the objective is having low fuel consumption of the vehicle as a whole, the importance of considering the WHOLE system (vehicle, weight, aerodynamics, usable space inside, rolling resistance, transmission / hybrid drive / energy storage as the case may be) can't be overstated.

It's true that if you separate the power-generation from the vehicle-propulsion, you can design out some of the transient conditions and the costs associated with those, and you can downsize the engine to match it better to the load. For example, perhaps you could eliminate the fancy variable-vane turbocharger that the TDI uses to get good transient response, and substitute a fixed free-running turbocharger, no VNT no wastegate. Possibly the servo-controlled EGR could go back to an old fashioned vacuum operated one. You'll still need the (upcoming) common-rail injection equipment, though.

Regarding the above post, plug-in hybrid normally isn't regarded as viable for a transit bus application, which is what I gather the original poster is working on. You can't be plugging in a vehicle that's expected to be on the road for 16 hours a day. All-electric drive is interesting for Iceland (which has an abundance of geothermal power), but the question is how to do it on a practical scale for a city bus. The forklift solution is to change out the entire battery packs - but a battery pack for a transit bus that's big enough to give a viable range (100 km would give maybe 2 - 4 hours of operation?) would be something colossal in scale. Maybe there's a way to do it? I've yet to see that done.

GM has hybrid-powertrain buses in production - they are in use in Toronto (and many other places). Don't know how well they are doing in service.

Definitely avoid making the big mistake of assuming that one technology is universally applicable to everything, because it isn't. The statements that the envirofreaks seem to be making (that anything and everything that Detroit is doing is bad) is every bit as wrong as the statement that everything Detroit is doing is perfect.

 

[GregLocock]

If you need electric buses to operate 16/24 then use a trolley bus (may be a problem with snow). If you need big batteries changed often use a battery trailer.

I am astonished to read that a small 50% efficient engine would be a waste of time.

Cheers

Greg Locock

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

 

[icelander]

CCycle,

how true. Indias "one lakh car" is a major environmental threat for these reasons precisely. The most effective environmental single sweep is pinching the purses of your fellow drive in laners through European levels of fossil fuel taxation. And its not about to happen. Still, we've been waiting "a while" for The Battery. Is it about to happen? Ummm...? Anything we can do until then is worth trying.

Actually, 50% is exactly the efficiency figure quoted for the Sulzer and similar city-block sized CI simpletons. They run at constant speed and load for days on end.

Another fact: In Sweden most cars are plugged in daily. But alas, only in winter, to keep the petrolswigging pistonpumper warm to start next morning. Proves though that infrastructure is no problem and car owners' will to plug exists. Most Western hemispheral car trips are to work, then shop, then back. Plug it in enough, and we have the tech for a viable EV now. Right now. "All" we need to resolve is the holiday drive. And getting the worlds most dangerous and frightened animal, the Consumer, to accept.

How about keeping the CEEVRE (Chemical Energy Electrical Vehicle Range Extender...?) on a shelf for the day-to day, and drop it in for the long haul? Like the luggage box? That would be a crossover hybrid, right?

Still hope somebody can direct me towards practical solutions others have found along these lines laid above. Reports, papers, but best of all, a reputable producer selling such units...

 

[BrianPetersen]

VW TDI best efficiency point works out to be around 41 - 42% thermal efficiency.

Are you designing a city bus, or are you designing a personal transport vehicle? The requirements of the powertrain are not the same!

For a city bus, a trolley (streetcar) is not a bad idea, we have those in Toronto and winter is not a BIG problem for them. But you need the tracks, and the overhead conductors, and there is certainly some maintenance involved. No fancy batteries, no technological breakthroughs.

I agree completely about the plug-in hybrid deal for personal vehicles. It's only a matter of time before they're in production.

Re the Indian Tata car ... the real breakthrough would be convincing North American SUV drivers that something like this is really all they need!

 

[CCycle]

My apologies. I said transportation where I meant automobile. Brian is correct, a city bus is not a candidate for pluggable hybrid technology, hybrid maybe but not pluggable hybrid. I also did not mean to discredit efficiency advances. We must keep the efficiency as high as possible. But it will probably come with issues.

Thirty years ago, after the energy scare of '73 in the states there was a scramble to produce high efficiency electric motors. The high efficiency motor was great on things like freon compressors or blowers where the load was more or less constant. But if the motors were run off their design point they were no more efficient and perhaps less efficient than a standard motor.

Which brings us to Icelander's original post. Future engines will be more efficient and hopefully simpler but will benefit even more from load matching.

 

[EdDanzer]

There are many hurdles to overcome with vehicle drive trains. I believe one reason electric vehicles will never be mainstream will be cost per mile over 100,000+ miles. To build a car that will have mass market appeal it must have a good range between charges and short charge times. For commercial vehicles this may never be possible.

I have published a White Paper

http://www.dehyds.com/html/standby_power_white_paper.html

that you should read to understand one of the difficulties in designing a drive train with an engine and generator to charge batteries and run electric motors large enough to provide starting fully loaded on a hill as well as operating at highway speeds on long steep grades.

One of the most frustrating things about comparing technologies to specify for an application is determining what data is hyped and what is real. Another difficulty is quantifying what you really need to the task the vehicle will be used for. Current drive cycle information is close to useless for designing a vehicle drive train.


Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[SomptingGuy]

Looking at that VW BSFC map, it seems as if you get the best mpg in 4th gear at 90 mph, which I find odd and hard to believe. It does mention drag coefficient and frontal area, so I assume they are taken into consideration.

- Steve

 

[BrianPetersen]

You're reading the chart wrong. The *effiency of the engine*, taken in complete isolation from everything else, is indeed quite likely to be best in that condition, but the *fuel consumption* will not be.

The engine makes the 70 horsepower that it takes to go 90 mph more efficiently than it makes the 20 horses it takes to go 60, but the fact that it's using about 3.4 times more power to go 1.5 times as fast doesn't make up for it.

This is why I've been saying all along that you need to consider the complete system, not just each component in isolation from everything else.

 

[SomptingGuy]

I've not been myself the last few days. For some reason I read it as mpg, even though the numbers would be ludicrous.

- Steve

 

[SomptingGuy]

Hmm, even that's a stupid statement. I need more medicine and an early night!

- Steve

 

[Cronik]

Interesting thread.

One possible solution to the trade off between larger battery packs for longer range and the weight penalty associated with them is the use of inductive power transfer.

While this is not practical for typical automotive use, it could work very well for a bus service due to it well known travel path and frequent stops.

Embedding an inductive transfer loop in the road under the bus stop to give the batteries a quick charge while passengers are entering and exiting the bus.

Food for thought