h a l f b a k e r yStrap *this* to the back of your cat.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
A car's crankshaft is a large lump of
metal that ussually doesn't balance well.
Some cars rely on ballance shafts to
counteract the irregularities and vibration
of the crankshaft.
Instead of this, why not make half of the
crankshaft rotate in the opposite
direction? This would be fairly
easy with
the type of beveled gears found in a
differential.
In a straight four, the back two cylinders
would rotate clockwise, and the front
two would rotate counter-clockwise. You
could also put these gears between each
cylinder, so that every crank would rotate
the opposite way of the ones on either
side. This would be a lot more complex
and expensive, though.
With either scheme, you could use much
more solid motor mounts because there
would be little or no movement of the
engine due to torque and very little
vibration. With a well-ballanced
crankshaft or a boxer engine, there
would be essentially no vibration.
Scroll down to 1966. Great site for GP buffs.
http://www.atlasf1..../preview/jones.html that's ol' reliable Jim Clark piloting ol' unreliable in 2nd for time being - as angel noted - he won. [thumbwax, Oct 18 2005]
diesel motorcycle engine with counter-rotating cranks
http://thekneeslide...otorcycle/#comments Same name, completely different mechanism. Very clever idea. [discontinuuity, Sep 20 2006]
Neander
http://www.neander-motors.com/index.php One piston, two cranks. Two pistons, two cranks. Three... [elhigh, Sep 21 2007]
[link]
|
|
I thought torque steer was caused by having different length half-shafts on front-wheel-drive cars - how does this idea prevent this? |
|
|
This idea would prevent torque steer on rear drive cars, or other cars with longitudinally mounted engines. This wouldn't help out quite as much in a front-drive car. |
|
|
Actually pretty much all 4 cylinder engines have balance shafts. As well as most other configurations--the only inherently stable engines currently made are inline-6 and V12. All others require 1 or more balance shafts to stop vibration. (flat or "boxer" engines aside--they're all stable) |
|
|
At low speeds you can get by without them, and if the engine is really well tuned and you don't mind the car shaking you can remove them for more power, but for any high-RPM usage they're pretty much required. |
|
|
Lets say you have a 4cyl engine. This idea I would think would have 2cyl per half of the split crankshaft. So one shaft is spinning in the correct direction and it drives the tire in the correct direction. The other shaft needs to then have its rotation changed to properly drive the other tire. This sound like a crankshaft differential. Where I cant see it having a real adverse affect on anything, I cant really find why one would do this because it accomplishes nothing. |
|
|
Please differentiate between crankshaft and drive shaft, for they are two completely different things. |
|
|
No, [Antegrity], I wasn't talking about the driveshaft at all. If you wanted to, you could also make it so that exactly half of the engine/drivetrain was spinning one way, and the other half spinning the other way. That would eliminate torque steer in most rear-drive cars. |
|
|
My idea was to have the twisting forces of the engine counteract each other so that you could have stiffer motor mounts, making for a more responsive engine. If you've ever seen an engine being revved with the hood up, you will notice the engine twisting, which has negative effects on performance and the integrity of the chassis. |
|
|
This is not addressing the condition usually described as torque steer. You are trying to address the assymetric loading of torque into the vehicle chassis, and vibration by the clunkiest approach I've ever seen. And who says cranks don't balance? They balance great, or else they wouldn't be in your car. |
|
|
In a front drive car, this concept would do nicely at driving one side of the car backwards. |
|
|
A better solution to combat torque load is to run the driveshaft - at crank speed - through a torque tube to a transaxle in the stern. This works well, and is fully baked by GM in the 60's in a production model. |
|
|
How on earth would you keep sensitive things like engine firing timing correct with the intrinsic gear lash present in any geartrain? |
|
|
The next problem you're going to run into is horrendous noise due to powerplant micro-bending. |
|
|
Engine mounts are designed and located to attempt to minimize its transmission into the passenger compartment. |
|
|
This design would simply evolve into two engines built into one block, feeding a coupled torque converter, perhaps some sort of weird planetary torque converter or something. |
|
|
why not clamp the cam so that the car rotates around it
like a washing machine spin dryer - good for drying your
clothes on a wet day. Distributing the load properly would
ensure vibration free running. |
|
|
BRM used an H-16 configuration in the '60s; effectively two horizontally-opposed eight-cylinder units with their crankshafts geared together. It was not universally regarded as a good idea, although Jim Clark won the 1966 US Grand Prix in an H-16 powered Lotus 43. It was stupidly heavy with poor torque; Jackie Stewart allegedly said that it would have been more use as an anchor than as an engine. |
|
|
does misalignement of the bores not happen when two
engine anoraks can't get their stories straight? |
|
|
Engine noise would increase due to the highly jerky output of your typical crankshaft, and inputing that into the other engine in a largely difficult-to-predict manner. Even if you get the firing timing right to align to the cylinder position, it still may vibrate like a cordless razor, because you would have to accommodate for the largest torsionals in the crank design somehow. I doubt that the engine would find any smooth operation point where everything came together unless the two powerplants were nearly completely isolated by a *very* compliant torque converter, which would unfortunately also be inefficient. |
|
|
One of the jobs of the clutch and / or torque converter is to attempt to absorb and isolate that vibration from the rest of the drivetrain, as it would produce a very rough ride without it. But even clutches and torque converters can't isolate everything. There are still torsional vibrations that pass through. |
|
|
NVH is a weird and spooky science sometimes, and vehicle program timing usually doesn't allow for a lot of in-depth rocket-science NVH analysis, sadly. By the time that the simulation models are generated in any story-telling manner, it's too late to make any significant changes to the car tooling. So there's still an element of guess-work involved. I wish more money were spent there. |
|
|
In general it's not the engine block torquing around, it's the joints between the block and the transmission, and transfer case if you have one that are the biggest 'powertrain bending' locations. |
|
|
This is still one engine with one block, only the crankshaft is split in half. You wouldn't have to adjust the spark or cam timing, the front two cylinders would move up and down normally, just the crankshaft for those two would rotate the other way. |
|
|
I'm not sure what the best way to link the back and front halves of the crankshaft would be. I was thinking beveled gears like in a differential.
Each crankshaft would have a 45 degree beveled gear fixed on the end, with several smaller gears turned 90 degrees from the crank and placed in a cage radially. If this cage in the center was mounted on some kind of torsion springs, it could rotate back and forth to absorb some of the shocks from the difference in power pulses. |
|
|
[ rasberry re-tart ] Is the problem you are trying to solve:
1. The torsion caused between the engine mounted at the front of the chassis, and the diff mounted at the rear, due to engine torque, or,
2. The torque due to the angular acceleration of the engines crankshaft (the motion of the engine you notice whilst looking under the bonnet and revving your engine) ? |
|
|
#1 is solved as per the comment from [Pa've], with the torque transfer tube. |
|
|
Running 2 cranks joined longitudinally (as per your suggestion, 2+2) would create all the above mentioned problems in other members comments(backlash, excessive stresses, noise, elasticity, poor timing, weight, etc). And although it would remove the engine-caused longitudinal torque reaction, it would introduce a yaw or pitch torque reaction. |
|
|
Also, you would still need compliant engine mounts, as the problem you describe is not really a source of engine vibration. |
|
|
BUT, if you used a clear crankcase and mounted the engne in a motorbike frame, it would look cool and people would wonder what the hell was going on. For this reason, I give you a + |
|
|
Yes, I wasn't sure if this would solve any problems, but it would look pretty cool. |
|
|
Look folks, take it from a guy who designs engine and transmission parts for a living. Gears always have lash in them, and engines always run roughly, and the combination of the two always create the need for something springy or slippery to smooth out the torque flow. Statically enough lash is present so as there's a few degrees of slack; dynamically the teeth literally bang together back and forth as they spin if there's no damping. |
|
|
A crankshaft, even with a properly-sized flywheel, with the accelerator pedal set at a constant rate, accelerates and decelerates at a rate that's several orders faster than the crank rotation, so that if you graph the angular velocity of the crank vs. time for even 10 ms the spikes will be very pronounced, especially in a four-banger; +- 15% of speed isn't unheard of at idle. Combining these spikes with the gear lash needed for efficient operation will simply throw off the engine's timing from one bank to the other, and be enormously noisy to boot. |
|
|
torque steer is totally irrelevant and not linked to this idea proposed this idea is about lowering vibrations for better comfort
also torque steer is nothing to do with halfshaft lengths this comes into play on black ice where it makes the vehicle spin uncontrollably
thanks
Nick |
|
|
I've often wondered why engines for front-wheel drive cars are not mounted sideways (likewise for mid/rear engined sportscars). I'm sure someone has tried it but it's not commonplace. Anyone know the reason? |
|
|
What do you mean by 'sideways', [marklar]? Transverse crank orientation is the norm these days for front-drive installations. There are exceptions, though. |
|
|
I am the Norm these days. Lovely string of annos, though. |
|
|
You're essentially describing a square-four engine. I'm not particularly familiar with them; just happened to stumble across a bike magazine detailing the restoration of a square-four Suzuki. |
|
|
Check the Neander motorcycle link. They've taken this exact idea and built a working engine with it. Bulky, but a torque monster. |
|
|
Thanks for the link [elhigh], very interesting. Twin cranks with twin conrods was not where I thought this idea was going! |
|
|
a better recipe for baking would be counter-rotating driveshafts, this would ensure the vehicle stays level and goes straight during hard launches |
|
|
/stays level and goes straight/ |
|
|
Rotates about the engine, spinning both tyres wildly but in opposite directions. Or am I misunderstanding something? |
|
|
/You're essentially describing a square-four engine./ |
|
|
I was of course wrong, nothing of the sort is being described. A square-four, with two parallel cranks driven in opposite directions by half of the pistons each, makes sense. Trying to do it to a straight-four, with a reversing bevel gear set, is horrible - as the cylinders fire, the torque will pass up and down the shaft, so the torque across the gear set will keep reversing and the gears will snap back and forth through their clearances. I can't think of a faster way to break a gear set, except by using an angle grinder. |
|
|
So we make a problem where formerly we had a compromise. Sound reasoning. |
|
|
I like it, send it to Jenny girl, she does it like that, I think !!, :-) s. |
|
| |