Last post on Aug 25, 2008 at 12:05 AM
You are in the Hybrid Vehicles - Archived Discussions
This discussion is ARCHIVED. To reactivate the discussion, post a request in the Lost? Ask the Hybrids Host for directions! discussion.
What is this discussion about?
Engine, Fuel System, Transmission, Hybrid Cars, Coupe, Convertible, Hatchback, Truck, Sedan, SUV
#45 of 206 Hybrid Diesels by lucas
Feb 22, 2005 (1:52 pm)
You mentioned direct drive wheel motors,here's my take.
The only popular uses I know for direct drives are
disk drives; to permit compactness and avoid the speed variation problems associated with rubber belts.
And clothes driers to replace the two step-downs to the drum and the single phase motor; the one equipped with that fault prone centrifugal start-switch. But in this application whether the three phase direct drive pancake motor with inverter will provide the improvement of a more quiet and compact machine is open to question.
However the problem with this technology in automobiles is two-fold.
Firstly they add significant weight to the wheels and this unsprung mass, as it is called, out on the wheels will bring a large amount of vibration back into the vehicle unless you are driving on very smooth concrete which mostly of course you don't.
Secondly the base RPM is low (600rpm on the wheel)so the horsepower will be low also. You only get to use the magnet and copper materials 10 times per second whereas at 12,000rpm you can use the same materials 200 times a second. It can mean the difference between 10Hp and 100Hp in this context. For simplicity I have been proposing a 10:1 planetary reducer built into the end bell as they did on the original EV1 back in 1990. Take a look at my post #31.
It can be argued that an inverter grade induction motor is not significantly less efficient than a permanent magnet synchronous machine as you mentioned. But avoiding the expense of exotic neodymium magnets should also assist reducing its cost to Walmart proportions.
I want a $500 motor not the $4000 say, that you will find on the PRIUS. When I was involved with electric vehicles, I used to have a saying of not "sacrificing yourself on the altar of high efficiency". You can always find a place elsewhere for the same money which will provide a better improvement overall.
Having an inboard motor makes the sharing of coolant with the ICE more secure and further reduces the frame size you can get away with on this motor. My point is that liquid cooling will allow the specific power density of the motor to be increased.
Regarding diesels - these are slow machines - even so called high speed diesels top out at 4000rpm. They have to. The first part of the stroke is constant pressure; the fuel continues burning while the piston moves down, you have to allow time for that process to occur; sure the elevated compression ratio they use brings the Air Standard Efficiency from 36 to 44%. If the diesel were required to work at full rated power, I would be with you on diesels unfortunately the nature of motoring is to need +100Hp while accelerating for 8 seconds or so followed by a considerably longer duration when the demand is 10Hp or less. The higher rpm of the conventional ICE is much more desirable in this application IMO. This probably explains why you won't be seeing F1 cars with diesel engines anytime soon.
Turbos, well of course, in fact I think all vehicles should be fitted with turbos. Too bad the 3 cyl on Honda's INSIGHT didn't get one.
After I read Hugh McInnis' book on Turbochargers I was a believer.
Unfortunately these devices need some TLC, manufacturers have not installed the sensors and the microcontroller technology needed to prevent John Q. Public from abusing them, they've got a bad rap. In the hands of sports car aficionados and SAAB they continue to survive. But you don't see enough of these cars on the road to give me the confidence that service support would not be an issue.
I'd be a lot more enthusiastic if Corollas and Civics started championing this technology.
I would however have no trouble being a first time buyer of a SHEV with its induction motor and inverter drive because this type of technology is now meat and potatoes in the factory automation business. Motor generator sets are no big deal either even though the automobile version would be light years from the usual Honda standby set.
Feb 23, 2005 (11:48 am)
"Yamaha 'Waverunner'. This three-seater personal watercraft employs a 1.0liter 4cyl engine which produces 170Hp 10,000rpm. Just TWO cylinders of this marine engine would be equal in power to the atkinsonised version of the Echo 1.5L that powers the Prius."
The Waverunner engine is spinning at 10,000 rpm! If rev-limited to 5000 like the Prius, it would only generate ~85hp... about the same as the Prius engine.
No miracle engineering there. They simply made the waverunner spin faster (and less fuel efficient).
#47 of 206 Waverunner engine
Feb 24, 2005 (12:18 am)
I was just pointing out that higher speed engines are already considered doable for the consumer market and not merely some near-term project that would need big R&D dollars. I wasn't intending to suggest that it was a replacement candidate for the Prius engine just that a two cylinder version of that engine could provide the same power.
I agree, it may not be as fuel efficient as the Prius engine while running 10,000rpm but it could be as much as 100lbs lighter so there would be an automatic increase in mpg for that reason alone.
Dare I say that some of the problem lies with your obsession for efficiency. Imagine if the Prius engine was an Atkinsonised V8 3.0L it could run half the speed and so be even more efficient (using your logic) than the current 1.5L. It goes without saying (but I'm going to anyway!) that the increase in vehicle mass with the V8 would cause the mpg to take a hit. You weren't expecting to lug all that extra weight around for free were you ?
#48 of 206 Re: [stevewa] by robertsmx [toyolla2]
Feb 24, 2005 (8:40 am)
When discussing efficiency, you have to realize losses not just horsepower. An engine half the size of the Yamaha engine you quote could produce half the power (85 HP) at 10K rpm assuming similar efficiencies, and ultimately match 1.5-liter Atkinson Cycle motor as used in Prius in peak power output. But to relate power and fuel efficiency, we really need to look at BSFC.
Efficiency of a hybrid drive train is going to be determined primarily by two factors. One, how much of the energy is recycled. And the efficiency of the non-electric motor involved. Parallel versus Series isnít going to be an issue. It is all about efficiency.
You may have seen a Honda example of a prototype hybrid vehicle earlier from me that involves an ICE mated to electric motor in Series as well as in Parallel mode. The active mode depends on demand. For normal cruising, the vehicle ran in series mode, with ICE charging battery pack running electric motor hence the vehicle. During start up and acceleration, ICE and electric motor run in parallel mode for greater power delivery due to increased load requirements.
One of the best examples of a series hybrid is a diesel locomotive.
I was just pointing out that higher speed engines are already considered doable for the consumer market
High revving engines have been around in motorcycles for a long time. The problem isnít with the output itself, but how you manage to put it down.
It would be possible to replace a 3.0-liter, 150 HP motor with a 1.0-liter, 150 HP motor in a 3000 lb sedan. These are peak numbers of course, but assuming similar efficiencies, the 1.0-liter motor will need to spin at three times the rate of the 3.0-liter motor. Gearing will have to take care of this aspect, and spinning at 10K rpm versus 3000 rpm for same power output can be perceived or felt very differently in the same car. Not in terms of power, but in terms of NVH. And, high rpm isnít the best way to increase fuel economy.
As for weight, yes, if the 3.0-liter engine weighed 310 lb. as does Hondaís V6, and the 1.0-liter engine weighed 126 lb. as does Hondaís I-3 (in Insight), than you do gain 184 lb. in weight and the vehicle can be lighter. But, thatís probably going to, may be, compensate for added losses of higher revving design (in the engine, and the transmission itself).
#49 of 206 BSFC versus Vehicle Mass TRADEOFF
Feb 25, 2005 (7:14 pm)
robertsmx, In your first paragraph you say that when
... discussing efficiency, you have to realize losses not just horsepower.
...But to relate power and fuel efficiency, we really need to look at BSFC.
If we were discussing a stationary power plant I would agree entirely with those statements and the Specific Fuel Consumption (in lbs/Hp-hr)is king. But carrying excess weight on a mobile platform will require more horsepower and you will have more losses no matter how good your SFC is.
As an extreme example even though we know jet engines are dreadfully inefficient we don't see WestJet or any other no frills airlines running around in Super Constellations no matter how more efficient their engines might be. So I am going to stick to my guns on this one that while SFC is important - in automobiles (and airplanes) Mass is too.
Your last sentence of your post seems to infer a softening of your position
...vehicle can be lighter...thatís probably going to, may be, compensate for added losses of (a) higher revving design.
I've always contended that such high revving will only occupy a small fraction of time over the full driving cycle anyways. The steeper gradients or the stoplight Grand Prix, that kind of stuff. I would anticipate that most of the time the ICE would be below 4000rpm 50mph so engine wear would not be an issue. As a reality check to what a typical car needs most of the time let me refer you to this piece that I just dug out, an old 1983 advertising copy from Ford. It says " Aerodynamically designed, LTD needs only 6.7Hp to push it through the air at 50mph".
Oh and by the way, ...if saving 184lbs does compensate for the less efficient high revving...
If we could be a little more optimistic here, supposing it does more than compensate... in which case, maybe, you just gained an increase in the torque/mass ratio of the vehicle.
Then if you are not saying engine efficiency is paramount - no matter how much it weighs - where do you draw the line. I speculate Toyota went with the most advanced 1.5L engine block that was available at the time, they were taking enough risks.
We know they certainly don't view high RPM as being a problem; they now have MG1 turning 10Krpm. In fact if they decided now on a liquid cooled motorcycle engine there is wiggle room on the PSD to have its ratio changed to suit.
You make a point of the rpm being inversely proportional to the swept volume in the cylinders i.e. a 1.0L needing to run three times faster than a 3.0L. 2100rpm v 700rpm.
The waverunner engine produces 16.5Hp per 1000rpm.
Remember the LTD needing 6.7Hp 50mph, so while cruising in that car 1000rpm this engine would still have 10Hp in reserve via the throttle.
A two cylinder version of the waverunner in this car (8.2Hp/1000rpm) would be quite highly loaded with only 1.5Hp in reserve. Increasing engine speed to 2000 rpm would restore a 10Hp reserve.
If we go back to the real world, 2000rpm is just abit more than twice the minimum speed of the 3.0L VVTi 1MZ-FE engine, mine idles around 750 I can tell you.
It has a capability of 36Hp/1000rpm assuming constant torque, giving 27Hp at idle.
My conclusion is that most of the time this 0.5L engine need run only just over twice as fast as a conventional V6. Not 6 times faster as you would have proposed might be expected. Not forgetting that it is the sole ability of the PSD via MG1 to change its ratio so rapidly that makes this system usable.
They have to make the Prius lighter and less expensive, I believe this is one of the ways open to accomplish that.
Other than that let us agree to disagree.
robertsmx, your postings certainly raise some points, too many to address in one go.
By the way NVH = ? please
The recycling of power is important but I see no system having advantage over any other where electronic regeneration back to the main pack is concerned.
One more thing.
The efficiency of the prime mover is not the be-all of the system as you infer. The transmission is also key - and it pivots on which of the SHEV or Prius PSD methods can do this function in the most efficient way.
The SHEV is basically a system biased towards using larger power electronics.
The PSD is a system biased towards mechanics.
I did a post on separation of losses, some weeks back, because SHEVs get dismissed with what I call "lazy explanations".
But if you want a supreme example of what can happen when a group of mechanical engineers try their hand at electrical engineering.....
Next time it will be "Shark week for Honda"
I will be discussing the IMA with figures gleaned from Honda's own website, in answer to previous posts.
#50 of 206 Re: [stevewa] by robertsmx [toyolla2] [robertsmx]
Feb 27, 2005 (12:19 pm)
I have enjoyed very much your well-written messages on automobile engineering. I do learn something from you. You must have lot of automobile engineering experience.
Mar 05, 2005 (10:30 pm)
Since this board has more to do with physics and electrical/engineering I hope no one who owns a hybrid will get upset with statements written here in good faith. It may be a truism that the effect of advertising is not merely to attract new buyers but to affirm to new owners that they made the right choice.
To those I say, knowledge and experience are something you gain just after you needed them.
To ruffle feathers is not my intent any more than writing here is part of a popularity contest.
The Honda IMA is a motor affixed to the engine crankshaft. A 3-phase controller and a 144v NiMnH battery complete the system.
Figures in bold are from Honda's website for the Honda Civic Hybrid.
It's been written, ad nauseum, that the IMA's sole purpose is to reclaim kinetic energy normally lost by mechanical braking, the IMA becomes a generator and supplies current to the 144v battery, it therefore becomes a load on the shaft and slows the vehicle down. The slight brake pedal travel is detected as gentle braking so the system works on the IMA first , later more aggressive use of the brake pedal will engage the mechanical system.
On acceleration the IMA becomes a motor and draws current from the 144v battery. This happens as long as the ICE revs are kept low. Aggressive acceleration keeps the revs high and the IMA may not be able "to motor". Honda's figures tell this story, they don't hide this fact, but no-one to my knowledge has ever interpreted this data in a forum anywhere. To most people this stuff is as unintelligible as a Mutual Fund prospectus anyway.
The IMA is rated at 10kw and Honda further says their motor produces 36 lbs-ft 1000RPM but this is only represents 5.1 KW when you do the math. Assuming the torque remains constant, by extrapolation we can probably expect the full 10kw 2000rpm.
So torque 36 lbs-ft 1000rpm
increases to 36+87 lbs-ft 2000 rpm
with the engine now pulling at full torque.
We can expect that the IMA torque component, assuming constant 10kw Power, will decay inversely to speed. Right ? Sorry, Wrong. Torque roll-off is much more severe than that.
See the figures from the Honda website, I noticed from their figures that at 3000rpm the IMA provided only 18lbs-ft of torque?
But by the inverse law for constant horsepower shouldn't 36 lbs-ft 2000rpm
become 24lbs-ft 3000rpm. What's with this?
My calculator shows 18lbs-ft 3000rpm to be only 7.67kw. Where did the 10KW and 24lbs-ft, you would have expected, go to? And where will the torque be approaching 6000rpm? Try 3-4 lbs-ft.
Well the problem is with the inverter being limited to 144v battery voltage. The first thing you have to be aware of is that the frequency of the applied voltage must always match the speed of the motor. This applied voltage produces a rotating magnetic field around the stator of the machine and must be in synchronism with the powerful magnets attached to the rotor. This is fairly easy to achieve with a crankshaft encoder feeding positional data to the electronic controller.
But here are the problems. The stator winding producing this magnetic field has inductance. Inductance opposes changes in current. To make the motor rotate faster, higher frequencies have to be applied. The inductance produces even more opposition to this faster changing current. The net effect is that the controller finds increasing difficulty getting current into the motor. Torque drops.
And there is another effect that is even more insidious. The magnets on the rotor produce a strong static magnetic field. When the rotor starts rotating this field similarly rotates with the rotor. The field passes through the stator winding and generates what's known as a back-emf.
You want back-emf because that value multiplied by the current is a measure of motor power at that instant.
This voltage increases with rotational speed until it reaches the battery voltage at which point it will be impossible to inject current into the motor. Torque will drop to zero.
I predict this starts happening after the motor passes through 3000rpm at which point unless it it is disconnected the motor will behave as a generator and start charging the battery whether it needs to be charged or not.
Of course the designers bank on you not being able to tell when this torque drop is happening. Neither do designers hang out in the advertising offices. So misleading sales information that adds the IMA power to the engine power sounds reasonable but is pure hype. Sales will need a greater MSRP for a hybrid, why not let the customers think they are getting a slightly more powerful system in the horsepower department.
A case can be made that engine production spreads of +/- 5Hp at the top end are quite acceptable. This variation swamps the IMA power anyway so who cares.
That doesn't change the fact that at the low end the extra 36lbs-ft helps extend the useful torque band down to 1000rpm so you may be able to hold 2nd gear from 10 to 50mph and achieve respectable acceleration.
Of course for long road trips the IMA does nothing for mpg compared to what a tall gear in a non hybrid Civic could be doing.
As I stated in a previous post for the HCH I think they should have set the IMA base speed for 1000rpm and picked up 72ft-lbs instead of the 36ft-lbs. I notice in the HAH they are using an 850 rpm 15Hp IMA. The Accord is not significantly heavier than the Civic so the reclaim energy is going to be roughly the same.
The system components will therefore be similar too.
I rest my case.
Despite all this I can't help but say that winding electrical machines for low base speeds to improve the torque bandwidth of the gasoline engine is a misapplication of electrotechnology in automobiles. GM's EV1 had no problem regenerating to a stop with the same motor that would take it to 60 mph in a little over 8 seconds.
Mar 07, 2005 (12:09 pm)
"Of course the designers bank on you not being able to tell when this torque drop is happening. So misleading sales information that adds the IMA power to the engine power sounds reasonable but is pure hype."
They didn't hide anything. The Insight brochures include this graph, which *clearly* shows the actual engine+motor horsepower & the fall-off in motor torque at 2000 rpm:
(Please Note: The Prius' electric motor does the same torque falloff at 2400 rpm.)
#53 of 206 Re: IMA [toyolla2]
Mar 07, 2005 (5:27 pm)
Although I use IMAís capabilities very sparingly, one item that you have not brought to light is that IMA acts as a quasi-balance shaft for the 3-cylinder ICEís in all Insightís.
In terms of FE, IMA starts from a straight or forced Auto-stop is a force multiplier that a non-IMA equipped Honda lacks. There will always be traffic conditions not conducive to high FE that a non-IMA equipped Civic/Accord will not be able to take advantage of. The Honda hybridís on the other hand can achieve spectacular results even under the worst traffic conditions because of IMA. That is if the traffic conditions are handled properly with the IMA equipped Honda.
Wayne R. Gerdes