More on one speed/one load genset for electric vehicles
More on one speed/one load genset for electric vehicles
(OP)
This belatedly picks up an interesting thread from earlier in the year, titled "One speed/one load genset for electric vehicles". Please refer to thread 71-207534.
Great stuff, hard facts. Thanks. Brian Peterson rightfully points out the benefit of chamber size in the quest for efficiency. Lopping 5 cyl.s off a Sulzer and installing the remaining one in my daughters Honda would benefit the environment through 50% efficiency fueled by recycled road tar. Impressive lbs/ft at 180 revs. Some weight increase, though.
Let's slip down the size/efficiency graphs somewhat from full bore oil tanker engines to the generic 2 liter 4-banger, and swap it out of the minivan for 1/6 of a 12 litre I6 truck engine. Same swept volume, hp similar or down a tad, same torque, but revs go down 60%. Impossible for NVH reasons with a normal drivetrain. But when not asked to cover 900-2000 revs to bridge gear gaps and load fluctuations? Staying perennially put at BSFC? Married to a dedicated single-point generator? Harmonics guys could counterweight the sows ear into a silk purse, what?
Seriously though, Brians excellent info actually rather underpins the "single-point" point, I think.
My initial purpose was to find the negative facts to strike the single-point idea off the "interesting enigma" list. Things are getting interesting, as your contributions, not least Ed Danzers, to a large extent lead the opposite way.
Let's see. Best point BSFC of VW TDI 197 g/kWh. + 15% deviation = kissing 230 g/kWh + engine not always happy + what about the drivetrain + who's driving and, the part missing from most BSFC maps – what kind of time does that engine actually spend converting that "knocking on 47% thermal efficiency" into useful shove and odometer spin? How much time does the engine spend "off the map", in the idle or near idle zone where efficiency knocks on 0? A lot of time, in fact. So valuable chemicals die a futile death, spinning and sliding and wearing all the pricey gubbins in a complex engine adding 10% or more to the vehicle weight.
An engine that is started before departure and keeps burning away until after the trips end, will inevitably give us a mean efficiency for the trip of... a few percent? More? Let us measure the actual energy expended during the trip to push air and roll tyres (strain gauge on drive axle?). Let us then measure the BTUs missing from the tank. The discrepancy is chilling. Where did the 47% efficiency go?
Try this test yourself. Drive your modern car at a steady speed, and observe the consumption display. Now drop a gear. Then drop another. Maybe one more. You have now roughly doubled the amount of fuel (mis)used to travel the same distance at the same speed. We're not talking 15% here, but 50%.
This also drives home that steady travel power needs are in the region of 10% of many consumer cars' installed maximum output, used for occasional spikes of acceleration. Western consumer perceived car acceleration "needs" of 2008 are now bordering on the silly, this coming from a life-long motorsports man. A Citroën 2CV gives you more fun than a Toyota White Goods 4.0, any day or night. In spite of having a really inefficient engine for these computed times. Fisker is weighing down his Karma with a 250 hp 2.0 turbo solely for generative purposes in a full-EV. Effectively installing two whole propulsion systems with associated weight and cost.
We are, in fact, on to the whole system energy equation here. The baseline is the classic straight IC driveline, straight average driver, average conditions. Can we arrange solid concepts and common components to do better? If so, then how?
Electric vehicles have a few established benefits. Like being well understood, having been around for centuries. Like being eminently scalable (LectraHaul in your garage, sir?). Like being wonderfully efficient with a long enough extension cord (I'll take mine without batteries, please). Only taking energy to give energy, and when they do, at up to 97% efficiency if some proponents are to be believed.
Note that EVs score twice here. Efficiency when drawing juice, and not drawing any when not propelling. Adding regenerative abilities to the motor drops efficiency but these losses are soon won back in normal use, especially in cities.
There´s more. Packaging, bulk, power-to weight, noise attenuation, parts count, user friendliness ............. (add remarks on dotted line, knowledgeable electric guys), even longevity. Electricity makes a great car, truck, bus, mega-dumper, or ship and always has. A modern E vehicle can beat a modern IC vehicle easily.
If only you can leave out the batteries.
Let's plug then. Trolleybuses, trains, trams etc whir away all over, in the Alps too, encrusted in smoked ice. The downsides are expensive, conspicuous and maintenance heavy infrastructure, and where that ends, your mobility ends.
Plugging wherever you stop then? Speed charging and widespread charging opportunities would possibly have a much greater impact on E-vehicle acceptance than the elusive Überbattery, as it can be implemented sooner. Zap up an extra 60 miles while your burger fries. Zing a full charge in the K-Mart parking lot. Nice, but not here. Work is being done on this – never forget the ubiquitous 230V nightheater outlets in Sweden – but is still stuck in the future.
For us stuck in the present, onboard chemical-to-electric energy conversion is here. There is a great economic and technological advantage in de facto existence. Could this chemical-energy-to-mileage conversion be made considerably more efficient, not only in the future, but now? Economically?
Brian points out the approx. 15% battery charge/discharge losses and the efficiency of the Prius serial approach reducing the size, though not complexity, of the electrics. Valid, but when tested by Auto Motor und Sport magazine in factual everyday use against a couple of popular family cars of far superior performance and utility (VW and Peugeot diesels), the Prius consumes 20% more fuel. Sobering facts. Where are the diesel hybrids?
The evidence presented in these posts and elsewhere and mathematics rather support the theory that a dedicated present-day-tech single-point engine can further the parallel hybrid to the point of parity with the most efficient straight ICs and then some. Now. In terms of tech and $&€. In the future, it'll get better, with free-pistons, continuous combustion, external combustion, linear electrics et al, but that, gentlemen, is the future. A boggo piston&crank can do it now.
How?
By shedding all the accumulated complexities of compromise. By single-mindedly optimizing towards simplicity. Minimizing materials, mechanicals, bulk and waste. Leading possibly to a high pressure, high specific output (valveless) DI 2 stroke relying on modern expertise in fluid dynamics and harmonics. Turbocompounding, recuperation and partial adiabatic strategies following.
What it boils down to is this: Offsetting partially decreased transmission efficiency through increased chemical energy conversion efficiency.
Now, as before, the optimal transmission is no transmission. Attaching an IC engine crankshaft straight to a wheel 100% transmission efficiency. This efficiency is scuppered by the ICE being inefficient except within narrow parameters. So for a century engines have been compromised to fit transmissions and vice versa. A juggling act.
In this post, it is suggested that transmission inefficiency inherent in electrical production, partial storage and use, can be more than offset through as near as possibly absolute chemical-to-electrical conversion efficiency in a low-cost, optimized on/off single-point piston engine.
Losses for single-point optimized electric driveline:
ICE: 50%. Transmission path: 5-15%. Total system efficiency: 40-45%. Not including regeneration.
Losses for conventional ICE+gears driveline:
ICE: 55-100%. Transmission path: 3-8%. Total system efficiency: 0-40%. No regeneration.
Furthermore strengthening our approach as opposed to pure ICE+gears or pure electric with a humongous battery:
1) The battery of our single-point vehicle can take in external energy (be plugged in), saving fuel
2) Kinetic energy can be recovered and reused
3) Waste heat can be used for cabin heating
4) Comparable system weight to straight conventional ICE+gears
5) System cost between current hybrid and conventional drivelines
Yes?
No?
Please challenge!
Great stuff, hard facts. Thanks. Brian Peterson rightfully points out the benefit of chamber size in the quest for efficiency. Lopping 5 cyl.s off a Sulzer and installing the remaining one in my daughters Honda would benefit the environment through 50% efficiency fueled by recycled road tar. Impressive lbs/ft at 180 revs. Some weight increase, though.
Let's slip down the size/efficiency graphs somewhat from full bore oil tanker engines to the generic 2 liter 4-banger, and swap it out of the minivan for 1/6 of a 12 litre I6 truck engine. Same swept volume, hp similar or down a tad, same torque, but revs go down 60%. Impossible for NVH reasons with a normal drivetrain. But when not asked to cover 900-2000 revs to bridge gear gaps and load fluctuations? Staying perennially put at BSFC? Married to a dedicated single-point generator? Harmonics guys could counterweight the sows ear into a silk purse, what?
Seriously though, Brians excellent info actually rather underpins the "single-point" point, I think.
My initial purpose was to find the negative facts to strike the single-point idea off the "interesting enigma" list. Things are getting interesting, as your contributions, not least Ed Danzers, to a large extent lead the opposite way.
Let's see. Best point BSFC of VW TDI 197 g/kWh. + 15% deviation = kissing 230 g/kWh + engine not always happy + what about the drivetrain + who's driving and, the part missing from most BSFC maps – what kind of time does that engine actually spend converting that "knocking on 47% thermal efficiency" into useful shove and odometer spin? How much time does the engine spend "off the map", in the idle or near idle zone where efficiency knocks on 0? A lot of time, in fact. So valuable chemicals die a futile death, spinning and sliding and wearing all the pricey gubbins in a complex engine adding 10% or more to the vehicle weight.
An engine that is started before departure and keeps burning away until after the trips end, will inevitably give us a mean efficiency for the trip of... a few percent? More? Let us measure the actual energy expended during the trip to push air and roll tyres (strain gauge on drive axle?). Let us then measure the BTUs missing from the tank. The discrepancy is chilling. Where did the 47% efficiency go?
Try this test yourself. Drive your modern car at a steady speed, and observe the consumption display. Now drop a gear. Then drop another. Maybe one more. You have now roughly doubled the amount of fuel (mis)used to travel the same distance at the same speed. We're not talking 15% here, but 50%.
This also drives home that steady travel power needs are in the region of 10% of many consumer cars' installed maximum output, used for occasional spikes of acceleration. Western consumer perceived car acceleration "needs" of 2008 are now bordering on the silly, this coming from a life-long motorsports man. A Citroën 2CV gives you more fun than a Toyota White Goods 4.0, any day or night. In spite of having a really inefficient engine for these computed times. Fisker is weighing down his Karma with a 250 hp 2.0 turbo solely for generative purposes in a full-EV. Effectively installing two whole propulsion systems with associated weight and cost.
We are, in fact, on to the whole system energy equation here. The baseline is the classic straight IC driveline, straight average driver, average conditions. Can we arrange solid concepts and common components to do better? If so, then how?
Electric vehicles have a few established benefits. Like being well understood, having been around for centuries. Like being eminently scalable (LectraHaul in your garage, sir?). Like being wonderfully efficient with a long enough extension cord (I'll take mine without batteries, please). Only taking energy to give energy, and when they do, at up to 97% efficiency if some proponents are to be believed.
Note that EVs score twice here. Efficiency when drawing juice, and not drawing any when not propelling. Adding regenerative abilities to the motor drops efficiency but these losses are soon won back in normal use, especially in cities.
There´s more. Packaging, bulk, power-to weight, noise attenuation, parts count, user friendliness ............. (add remarks on dotted line, knowledgeable electric guys), even longevity. Electricity makes a great car, truck, bus, mega-dumper, or ship and always has. A modern E vehicle can beat a modern IC vehicle easily.
If only you can leave out the batteries.
Let's plug then. Trolleybuses, trains, trams etc whir away all over, in the Alps too, encrusted in smoked ice. The downsides are expensive, conspicuous and maintenance heavy infrastructure, and where that ends, your mobility ends.
Plugging wherever you stop then? Speed charging and widespread charging opportunities would possibly have a much greater impact on E-vehicle acceptance than the elusive Überbattery, as it can be implemented sooner. Zap up an extra 60 miles while your burger fries. Zing a full charge in the K-Mart parking lot. Nice, but not here. Work is being done on this – never forget the ubiquitous 230V nightheater outlets in Sweden – but is still stuck in the future.
For us stuck in the present, onboard chemical-to-electric energy conversion is here. There is a great economic and technological advantage in de facto existence. Could this chemical-energy-to-mileage conversion be made considerably more efficient, not only in the future, but now? Economically?
Brian points out the approx. 15% battery charge/discharge losses and the efficiency of the Prius serial approach reducing the size, though not complexity, of the electrics. Valid, but when tested by Auto Motor und Sport magazine in factual everyday use against a couple of popular family cars of far superior performance and utility (VW and Peugeot diesels), the Prius consumes 20% more fuel. Sobering facts. Where are the diesel hybrids?
The evidence presented in these posts and elsewhere and mathematics rather support the theory that a dedicated present-day-tech single-point engine can further the parallel hybrid to the point of parity with the most efficient straight ICs and then some. Now. In terms of tech and $&€. In the future, it'll get better, with free-pistons, continuous combustion, external combustion, linear electrics et al, but that, gentlemen, is the future. A boggo piston&crank can do it now.
How?
By shedding all the accumulated complexities of compromise. By single-mindedly optimizing towards simplicity. Minimizing materials, mechanicals, bulk and waste. Leading possibly to a high pressure, high specific output (valveless) DI 2 stroke relying on modern expertise in fluid dynamics and harmonics. Turbocompounding, recuperation and partial adiabatic strategies following.
What it boils down to is this: Offsetting partially decreased transmission efficiency through increased chemical energy conversion efficiency.
Now, as before, the optimal transmission is no transmission. Attaching an IC engine crankshaft straight to a wheel 100% transmission efficiency. This efficiency is scuppered by the ICE being inefficient except within narrow parameters. So for a century engines have been compromised to fit transmissions and vice versa. A juggling act.
In this post, it is suggested that transmission inefficiency inherent in electrical production, partial storage and use, can be more than offset through as near as possibly absolute chemical-to-electrical conversion efficiency in a low-cost, optimized on/off single-point piston engine.
Losses for single-point optimized electric driveline:
ICE: 50%. Transmission path: 5-15%. Total system efficiency: 40-45%. Not including regeneration.
Losses for conventional ICE+gears driveline:
ICE: 55-100%. Transmission path: 3-8%. Total system efficiency: 0-40%. No regeneration.
Furthermore strengthening our approach as opposed to pure ICE+gears or pure electric with a humongous battery:
1) The battery of our single-point vehicle can take in external energy (be plugged in), saving fuel
2) Kinetic energy can be recovered and reused
3) Waste heat can be used for cabin heating
4) Comparable system weight to straight conventional ICE+gears
5) System cost between current hybrid and conventional drivelines
Yes?
No?
Please challenge!





RE: More on one speed/one load genset for electric vehicles
I'm in total agreement that today's conventional powertrains have gotten grossly oversized in the interest of top speed (that can't legally be used) and acceleration (which is scarcely used by most drivers). There's a fair bit to be found by NOT using ANY fancy high tech hybridizing at all and simply "right-sizing" the powertrain - but it will require people to accept that they don't actually use their 0-to-60 in 6 seconds capability. Plenty of us older folks grew up with old air-cooled VW Beetles and 60-hp first-generation Civics, and they kept up with traffic (mostly) ...
RE: More on one speed/one load genset for electric vehicles
The Volt use the electric motor to match the IC engine output to the vehicle demand power, when it is in IC on mode. The Prius does the same sort of thing but doesn't have fixed setpoints.
Cheers
Greg Locock
SIG:Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips.
RE: More on one speed/one load genset for electric vehicles
The idea here is to take a leaf out several books.
I.e. engine tuner practice: After being "down-tuned" 10% or more from their achievable power peak, for drivability, 500cc motorcycle GP 2-strokes belted out 200 hp on carburettors. Drivability being a relative term in this mad case! Those bikes reportedly needed suicidal geniuses to complete fast laps.
I.e. F1 turbos: When these 1.5 litres reached their 1000-1500 hp peak decades ago, they were said to have been some of the most fuel efficient IC engines ever (myth or truth?). Running on deep-chilled poison though. Until they grenaded a few miles on.
I.e. Hugo Junkers's aircraft diesels: These "clean" 2-strokes had power-to-weight ratios equalling contemporary sparked powerplants, without turbos, at considerably lower specific consumtion. 70 years ago.
Volt and Prius both use classic 4 stroke 4 valve 4 cylinder engines as "sub-prime movers", consuming as much space and mass as in a "straight" car without electrics. It is logical to use the century of development behind these units to throttle them to power demands, at least as long as 30% efficiency (Greg Locock) is still acheivable. "Single-point" tuning such an engine would, as Brian rightfully reasons, possibly be counterproductive because of the electric charge/discharge losses. At best maybe, not worth the investment.
Imagine an IC one quarter the bulk and weight of 2 litre fours, with apprx 90% reduced parts count (probably one valveless cylinder, scaleable by adding cylinders),. "Peakiness" is not an aim but an unfortunate effect of seeking the absolute in ICE performance, simplicity, or efficiency. Development would determine the "sharpness" of the peak, trading off-peak operation losses against electrical train losses. The output of this seemingly simplistic result of fluid dynamics modelling would be absorbed by a system optimized generator. Without this relatively drastic downsizing and simplification (of the parts count, not combustion quality), single-point tunings' drawbacks are probably too large. On the twin merits of low mass/bulk and fixed ca 50% efficiency it still seems logical to me - until proven otherwise - that total system efficiency could benefit enough to consider this path.
Obviously, this whole idea would at best give a net return of investment - no "50% more for 50% less!!! Oil companies will fight it!!!" breakthrough. Do such things exist? The Prius is an example. Promising technology, properly researched, funded and marketed, falling short on delivery. Needing another few investment billions. This single-point concept needs old-fashioned development graft as any other for realisation, but it sure is fun (and cheaper!) exploring with you the honest pros and cons . . .
PS - Why not same single-point idea w. hydraulics & accumulator instead of electrics & battery? How are you doing, Ed Danzer?
RE: More on one speed/one load genset for electric vehicles
The Prius isn't significantly different in fuel consumption from Honda's much simpler IMA because they're into the diminishing-returns regime, where exponentially more complexity is rewarded by only marginal gains in efficiency.
Keep in mind that an undersized but hard-working engine is not necessarily any more efficient than a slightly larger but more mildly-tuned engine, and in some cases it can be *less* efficient. There is a very strong diminishing-returns phenomenon. Case in point, I believe on the next-generation Prius, Toyota is going to use a slightly *larger* engine ...
VW has been playing around with hybrid "twin-drive" powertrains combining a TDI engine with an electric "assist". They have been using the 3 cylinder version of the TDI in this application - basically the same engine used in the next-size-smaller standard vehicle, the Polo.
Keep in mind also that in a steady-state motorway driving condition, the most efficient powertrain is the one that has the least losses ... and among the technologies that we know how to do, that's a manual transmission with gear-to-gear power delivery connected to a "right-sized" engine. No hybrid no nothing, any of that is just extra weight along for the ride.
RE: More on one speed/one load genset for electric vehicles
Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers for professional engineers
RE: More on one speed/one load genset for electric vehicles
Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers for professional engineers
RE: More on one speed/one load genset for electric vehicles
RE: More on one speed/one load genset for electric vehicles
VWs down-sized-diesel+electrics hybrid has been a long time coming and is not yet available. Perhaps they have been struggling with cost-effectiveness, as they have hinted. There has to be a strong case for installing two different powertrains in one family car, as already mentioned. Economy must increase substantially to warrant the extra weight and expense.
However, rather than a gasoline car with added electrics, the concept of this single-point thing is simply an EV with the minimum of added ICE range extender to substitute a maximum of battery weight. If that equation doesn't show improvement over either straight diesel & gears or straight EV, why bother? Therefore, my somewhat fanciful connotations to high-performance piston engines from the past.
A full-weight fuel-burn powertrain on top of the EV gear (Volt-type parallel hybrid) must replace considerably more than its own weight in batteries to start making sense.
And yes, my earlier post broke out of bounds of this stellar forums' standard of engineering reality. My hope in starting this post was to try to make sense of a quandary. It's coming along nicely. Thanks.
RE: More on one speed/one load genset for electric vehicles
Cheers
Greg Locock
SIG:Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of Eng-Tips.
RE: More on one speed/one load genset for electric vehicles
Well said; I would add to that performance and range.
RE: More on one speed/one load genset for electric vehicles
RE: More on one speed/one load genset for electric vehicles
RE: More on one speed/one load genset for electric vehicles
RE: More on one speed/one load genset for electric vehicles
Basically, the Prius has no clutch, it's direct drive from crankshaft to wheels; nothing but gears. It has infinitely variable speed reduction, which allows it to work without a clutch. Unfortunately, torque multiplication is not infinite, which is why it needs battery/electric at low speeds to get moving. That's what the "hybrid" part is for, its only purpose is to enable that amazingly efficient transmission to work, nothing else.
Once it was decided to use hybrid drive, tuning the engine for atkinson cycle running was another benefit. Also, the ability to turn the engine off at idle became part of the whole concept of managing engine speed independent of vehicle speed. This transmission requires a very advanced control system; a "model based" feedback system, a simple PID won't work which is probably why it hasn't been done before, the underlying technology just wasn't there.
Most competing products from Ford, GM, Honda, etc don't use the Prius transmission, which basically reduces "Hybrid" to crude marketing gimmick, doomed to failure. As Jeremy Clarkson says: "say again why I want to buy a car with 2 motors?"
RE: More on one speed/one load genset for electric vehicles
Perhaps things might be getting better at EPA due to the unrealistic numbers the EPA gave hybrid's prior to 2008. According to the EPA they have changed the test procedures for 2008 to give a real world number for mileage: http://www.epa.gov/fueleconomy/ I am still very skeptical about mileage claims.
RE: More on one speed/one load genset for electric vehicles
Much fuss has been made of the capability of the Prius to run electric-only, and the incapability of Honda IMA to do that. Well, I have in fact test-driven a Prius, and I couldn't drive it without the engine starting. Yeah, it is possible to pull away extremely slowly and at low speed electric-only, but that's not the way I drive and it's not the way most people drive. Net result: no real world benefit ...
The upcoming Ford Fusion is supposed to do 70 km/h electric-only, and the next version of the Prius is supposed to have not only a bigger gasoline engine but also a more capable electric powertrain. Remains to be seen what *that* is going to do.
My late model Jetta diesel has been using 5.0 L/100 km recently, with none of this hybridization nonsense.