quote Originally posted by joesfiero: Another example is a spring or torsion rod, which you mentioned yourself. If you have a torsion rod suspension on a car and the car is sitting still, there is X amount of energy stored in the torsion rods. The suspension is not moving (I.E. dynamic torque) and there is a certain amount of force applied to the control arms of the car, keeping it suspended. Now, with the same exact car, if you shorten the torsion bars you will notice the car will sit higher. This is because less energy is being stored in the torsion bar itself and more force is applied to the control arm.

The reason the car sits higher is because you've increased the spring rate of the torsion bar by making it shorter. The same amount of force is still being applied to the torsion bar as before, because the source of the force on the bar is the weight of the vehicle, which hasn't changed. So, by having changed the spring rate, the same weight will twist the torsion rod less and leave you sitting higher. If you were to examine the amount of torque being applied to the torsion bar, you'd find it was the same in both cases.

(Edited to add: Wooo-whoo... page two is mine!)

[This message has been edited by Bloozberry (edited 12-11-2009).]

quote Originally posted by Bloozberry: Steve, no one is disputing the fact that the axle will twist. What's being misunderstood here is that torque is being absorbed and LOST through the twisting motion. If this were true, then theoretically you could attach a super long axle to an engine, and keep it from turning with your bare hands. This isn't true because the twisting of the axle is only stored energy, not lost. As soon as the super long axle was done twisting (and storing the energy), I can assure you that you wouldn't be able to keep it from turning with your hands, or anything less than an opposite amount of torque the engine is producing.

Well, I understood that to be Joe's point, that a certain amount or the rotational force applied (and measured) by the torque wrench would be stored in the (twisted) extension and not applied to the bolt head.

You engine analogy would actually be a different consideration because the engine would continue to add additional rotational force while Joe's torque wrench example would apply only a fixed amount.

I salute both of you, this is the most considered and polite expression of different opinions that I have seen on this board in a long time.

------------------
Steve AT 88GTP DOT com
88 GT\3800 SC\4T65E-HD

quote Originally posted by pmbrunelle: The straight green arrow pointing down shows the force of your hand. The curved arrow shows the twisting. The extension bends because you are pressing downwards. This wouldn't happen with a cross-type lug wrench.

Sorry, but any mechanic knows that when tightening or breaking loose a bolt with a ratchet and extension, you need to hold the head of the ratchet steady, so as to not apply a bending force on the extension. When you hold the head steady, the only movement will be a twisting (torque) force on the extension.

quote Originally posted by Marvin McInnis: Assuming for a moment that you were correct, where does all that energy that is "absorbed" in a long torque tube go? Does the torque tube heat up indefinitely? No. As Bloozberry correctly stated, once steady state is reached the torque out will equal the torque in ... regardless of the length of a torque tube.

The torque shaft, or tube does not in fact heat up indefinitely. The shaft absorbs a certain amount of energy as it is being loaded and it retains this energy until it is unloaded and the shaft twists back. The more torque applied to the shaft, the more it will twist and the more energy it absorbs. The longer the shaft, the more energy it will absorb, resulting in less torque out,

quote Originally posted by Steve25: Joe missed on this one. If you remove coils from a spring it will apply LESS force. It has less steel in it to apply force and therefore cannot do as much work.

Nope, cutting coils will INCREASE the spring rate every time.

quote Originally posted by Bloozberry: Steve, no one is disputing the fact that the axle will twist. What's being misunderstood here is that torque is being absorbed and LOST through the twisting motion. If this were true, then theoretically you could attach a super long axle to an engine, and keep it from turning with your bare hands. This isn't true because the twisting of the axle is only stored energy, not lost. As soon as the super long axle was done twisting (and storing the energy), I can assure you that you wouldn't be able to keep it from turning with your hands, or anything less than an opposite amount of torque the engine is producing.

Blooze, you are confirming what I am trying to say here. As the axle twists, less torque is transmitted to the driven end. As I said before, the energy is not lost, just stored until the axle is unloaded.

My whole point was simply this, that in an unequal length axle configuration, the longer axle will twist more than the shorter one which results in less torque applied to the wheel with the longer axle for the amount of time that the car is accelerating, causing torque steer (this is only one of the causes of torque steer though) Once the car reaches cruising speed the torque from the motor is greatly reduced and the axle unloads its stored energy. After this happens, both sides of the car have an equal or close to equal amount of torque.

I only gave the examples of using a torque wrench and extension so you could see plain as day the result of using a longer extension. I really dont want to go get out my engineering information and start posting mathematical equations of torque through a driveshaft. Thats not my game at all, I just wanted to post the correct information for the OP because I absolutely 100% know that I know what I am talking about.

P.S. Bloozeberry, I gave you a + rating because you have been so respectful in this discussion. Steve is right, there is too much bickering on this forum and it gets in the way of information. We may have a difference of opinion but at least we never resorted to name calling or anything like that. For that I respect you and hope to be on the same side of the fence as you one day, maybe about another topic though

-Joe

[This message has been edited by joesfiero (edited 12-11-2009).]

quote Originally posted by theogre: Torque steer in Fiero? Yes But not the half shaft, ie fwd, version. Torque and rubber cradle mount can cause this because of the dog bone. Dog bone make the right side of the cradle more than the left side when it accelerate or decelerate. Stick tranny does it more then auto tranny but auto can do it to, especially if the cradle is soft. Twist the cradle and cradle going nut in steering alignment. A arm and tie rod is connected to cradle, top of strut to frame.... That's why how a cradle mount method is importance in 84-87 Fiero... Polly is less but Metal stops that cold.

This is slightly off topic compared to the theoretical discussion, but is solid mounting the cradle the quickest way to stop torque steer on a worn out 84-87 suspension? As I indicated earlier my 84 has bad torque steer and I'm looking for the most bang for my buck on what to replace.

quote Originally posted by joesfiero: My whole point was simply this, that in an unequal length axle configuration, the longer axle will twist more than the shorter one which results in less torque applied to the wheel with the longer axle for the amount of time that the car is accelerating, causing torque steer (this is only one of the causes of torque steer though) Once the car reaches cruising speed the torque from the motor is greatly reduced and the axle unloads its stored energy. After this happens, both sides of the car have an equal or close to equal amount of torque.

I think there was a misunderstanding - we were thinking you thought the unequal length made a difference in the steady-state, not just transitions.

Aside from that though, the wind-up time is pretty small, and torque steer does happen during steady-state conditions.

Thanks Joe. I think we've pretty much exhausted our abilities to explain our differing points of view trying to get the other to change his mind without repeating ourselves too much. At this point, I don't think either of us is going to change the other's mind. For my part, I'm going to leave it at that and let anyone who reads this thread decide what philosophy they want to abide by. A "+" for you too.

Sorry, but I'm not willing to let incorrect information like this go unchallenged.

quote Originally posted by joesfiero: The torque shaft, or tube does not in fact heat up indefinitely.

Yes. So where else does the "absorbed" energy go?

quote The shaft absorbs a certain amount of energy as it is being loaded and it retains this energy until it is unloaded and the shaft twists back. The more torque applied to the shaft, the more it will twist and the more energy it absorbs.

Yes ... but for a real-world shaft the amount of twist and the time it takes to do so will be miniscule. Tire hysteresis is probably just as significant a factor in torsional flexibility as axle twist, but that's just informed speculation on my part.

quote The longer the shaft, the more energy it will absorb...

Yes.

quote ... resulting in less torque out,

Absolutely not! That erroneous conclusion is the flaw in your reasoning. Under steady state conditions, torque out = torque in.

quote As the axle twists, less torque is transmitted to the driven end.

Absolutely not! From Engineering Mechanics for Structures (MIT): "[It is a] fact that the torque does not vary as we move along the axis of the shaft." (Emphasis added.)

quote My whole point was simply this, that ... the longer axle will twist more than the shorter one ...

Yes.

quote ... which results in less torque applied to the wheel with the longer axle for the amount of time that the car is accelerating ...

Absolutely not!

On the other hand, in addition to half-shaft angle there are several other variables that can and do affect torque steer even under steady-state conditions:

• Caster
  
• Scrub Radius (including camber effects)
  
• Left/right weight distribution (both static and dynamic)
  
• Steering angle
  
• Steering geometry (steering axis inclination, Ackerman, tie rod angle, etc.)
  
• Axle flex joint characteristics (plain U-joints vs. CV joints, etc.)
  
• Road characteristics (pavement type, wet/dry, lateral slope, etc.)
  

Even with identical length half-shafts, any left/right imbalance in these variables will result in torque steer effects, although caster, scrub radius, weight distribution, and road characteristics will probably dominate under straight-line conditions.

N.B. If torsional deflection due to different length axle shafts were indeed the primary problem, it would be easy to solve by simply making the longer axle slightly stiffer. Since the torsional stiffness of a circular shaft is proportional to the 4^th power of its radius, only a slight increase in the diameter of the longer axle would be needed to match its torsional stiffness to that of the shorter axle. This is exactly what some car manufacturers do, but torque steer effects still remain.

[This message has been edited by Marvin McInnis (edited 12-14-2009).]

Im going to throw another wrench into the works....

If you hold the two wheels still, and continuously build torque to the differential this is what will happen:

The long axle starts to twist some, the short axle twists considerably less. This is a given.

What does the differential do, it spins 50% of the difference between both axles deflection, and delivers exactly 50-50% torque.

Since the differential splits the torque 50-50, then the wheels will have exactly the same torque on them.

In fact, as the torque is increasing, and the long shaft is twisting, for that brief moment, the differential itself is also delivering the same reduced torque at the short axles wheel as the wheel on the end of the long axle by spinning the spider gears slightly to eat the deflection. So I fail to see how a long axle can deliver different torque as a short axle.

However, there is a difference in torque output based on the angle of the axle. Bloozberry touched on this earlier. As an axle angle changes from perfectly straight, the force it exerts changes from purely torsional, into a force that is perpendicular to the direction of the axle movement plane. Picture an axle that is at a 90 degree angle, It will deliver Zero torque, and become strictly a lever. Anywhere in between is a percentage of both forces. This non-torsional force can produce some unwanted forces at the steering knuckle, as well as reduce the delivered torque to the wheel

The shorter shaft is at a sharper angle, and since the "lever force" (im calling it this from my example) is greater due to the angle, and it is from a shorter lever, a much larger amount of force is delivered to the steering knuckle then would come from the other side with the long axle. However when the axles are straight. The torque delivered is exactly 50-50.

[This message has been edited by Fierobsessed (edited 12-14-2009).]

quote Originally posted by Fierobsessed: Im going to throw another wrench into the works.... If you hold the two wheels still, and continuously build torque to the differential this is what will happen: The long axle starts to twist some, the short axle twists considerably less. This is a given. What does the differential do, it spins 50% of the difference between both axles deflection, and delivers exactly 50-50% torque. Since the differential splits the torque 50-50, then the wheels will have exactly the same torque on them. In fact, as the torque is increasing, and the long shaft is twisting, for that brief moment, the differential itself is also delivering the same reduced torque at the short axles wheel as the wheel on the end of the long axle by spinning the spider gears slightly to eat the deflection. So I fail to see how a long axle can deliver different torque as a short axle. However, there is a difference in torque output based on the angle of the axle. Bloozberry touched on this earlier. As an axle angle changes from perfectly straight, the force it exerts changes from purely torsional, into a force that is perpendicular to the direction of the axle movement plane. Picture an axle that is at a 90 degree angle, It will deliver Zero torque, and become strictly a lever. Anywhere in between is a percentage of both forces. This non-torsional force can produce some unwanted forces at the steering knuckle, as well as reduce the delivered torque to the wheel The shorter shaft is at a sharper angle, and since the "lever force" (im calling it this from my example) is greater due to the angle, and it is from a shorter lever, a much larger amount of force is delivered to the steering knuckle then would come from the other side with the long axle. However when the axles are straight. The torque delivered is exactly 50-50.

well, are we now going to calculate the "wind-up" times?
the longer axle will take slightly longer to "load-up", and will spend slightly more time "unwinding"

using rubber erasers to make a mechanical model is really helpful here

Looks like we have quite the debate here! Theres been a lot of great points posted guys, thanks

------------------

Check out my build!
https://www.fiero.nl/forum/Forum3/HTML/000100.html

quote Originally posted by Marvin McInnis: Sorry, but I won't let incorrect information like this go unchallenged.

quote Yes. So where else does the "absorbed" energy go?

quote The shaft absorbs a certain amount of energy as it is being loaded and it retains this energy until it is unloaded and the shaft twists back. The more torque applied to the shaft, the more it will twist and the more energy it absorbs.

Pretty much answered it right there.

quote Yes ... but for a real-world shaft the amount of twist and the time it takes to do so will be miniscule. Tire hysteresis is probably just as significant a factor in torsional flexibility as axle twist, but that's just informed speculation on my part. [QUOTE]The longer the shaft, the more energy it will absorb... Yes.

So right there you are not only admitting that a twisting shaft absorbs energy, you are stating it in your own facts.

quote "[It is a] fact that the torque does not vary as we move along the axis of the shaft." (Emphasis added.)

From your own link you posted "The cross-sections of a circular shaft in torsion rotate as if they were rigid in-plane. That
is, there is no relative displacement of any two, arbitrarily chosen points of a cross section
when the shaft is subjected to a torque about its longitudinal, z, axis. We prove this assertion
relying on rotational symmetry and upon the constancy of the internal torque as we
move down the axis of the shaft.
We first show that radial lines must remain straight by posing that they deform, then show a contradiction
results if we do so."

What it seems to me they are saying is that their studies are based on the shaft NOT twisting, in which it would show contradictory results of the torque as you move down the shaft.

We have all determined that the axle does in fact twist, and the longer the axle the more twist. Now, as the axle twists, it is absorbing torque, it takes torque to twist the shaft. So that being said, how in the hell do you apply 30 ft.lbs of torque on one end of a shaft, say the twisting of the shaft takes 5 ft. lbs, and end up with 30 ft. lbs. still at the other end? Like I said before, the axle twisting absorbs the energy which is not lost, the shaft releases the energy by torquing the shaft the other way as it is unloaded, meaning the drive end will spin up, the shaft will twist, and then when the torque is reduced, the driven end of the shaft will "catch up" via rotational force, or torque from the shaft un-twisting.

I understand this all happens pretty quickly, and there are two things wrong with you beating me up about my statements.

1. I NEVER stated that axle twist was the ONLY reason for torque steer. I clearly stated that it is one of many possible causes of it.

2. We are not talking about an electric motor here with a flat torque curve. A gas motor has a rising torque curve until it reaches max torque while accelerating. As more torque is applied to the shaft, the more it will twist which takes, as I said before, until your torque curve maxes out or you stop accelerating.

If you still dont believe me, please try the torque wrench example I stated earlier. It will explain a lot when you realize the longer extension does not deliver as much torque as the shorter one once there is load on it and the extension twists.

-Joe

quote Originally posted by joesfiero: What it seems to me they are saying is that their studies are based on the shaft NOT twisting, in which it would show contradictory results of the torque as you move down the shaft.

So you are right and MIT, too, is wrong? A lot of tuition-paying MIT students are going to be really disappointed.

quote If you still dont believe me, please try the torque wrench example I stated earlier. It will explain a lot when you realize the longer extension does not deliver as much torque as the shorter one once there is load on it and the extension twists.

I don't know how many other ways I can say it ... you are incorrect! Torque out = torque in, regardless of the length of the extension.

[This message has been edited by Marvin McInnis (edited 12-15-2009).]

quote Originally posted by Marvin McInnis: So you are right and MIT, too, is wrong? A lot of tuition-paying MIT students are going to be really disappointed.

I never said the studies were wrong, just that you were interpreting them wrong.

quote Originally posted by Marvin McInnis: I don't know how many other ways I can say it ... you are incorrect! Torque out = torque in, regardless of the length of the extension. I am done here.

So I take it you didnt try the torque wrench thing?

-Joe

[This message has been edited by joesfiero (edited 12-15-2009).]

Joe, since you base your arguments on your example of the torque wrench, let me attempt to convince you one more time through several examples why your understanding of torque and torque transfer isn't correct. The first concept you believe relates to torque transfer, and that is, that a torsional force applied to one end of a shaft is not transfered entirely to the other end because a measurable portion of the force is absorbed to twist the shaft. So, you argue, that the force available at the other end of the shaft is equal to the applied force less the force needed to twist the shaft. The second concept you believe is that the longer the shaft, the larger the amount of force absorbed by the shaft and therefore the output torque is propotionately less at the other end.

So, here's where I argue strictly from an experimental point of view that your first concept of torque transfer isn't logical:

Example 1. Let's say you take a one foot long torsion spring like the one at the top of your garage door, only smaller, and hold an end in either hand. According to your reasoning, if you started twisting the spring with your right hand while holding your left hand steady, then you wouldn't have to hold the spring as tightly in your left hand because some of the force exerted by your right hand is absorbed by the spring. Now then, whether you view the fact that your right hand is twisting and your left holding it still, or vice versa, is only a matter of perspective. If you agree with that concept then, how could it possibly be that by twisting the spring the same amount with one hand results in a different amount of force needed to twist the spring with the other hand?

Example 2. Using your own example of the torque wrench, you imply that torque wrenches must be designed for use with a specific length shaft (not the handle, but the extension shaft in the same axis as the bolt) to ensure that the proper amount of torque is actually applied when tightening a bolt. According to your reasoning, if you used anything but the manufacturer's designed extension length, you would introduce error in all your torque applications. Well, for starters, you'll notice that nowhere in the literature that comes with a torque wrench is there any such recommended extension length. But even more compelling would be to try an experiment with two torque wrenches attached to each other through a common extension. If you were to set both wrenches to click at say 50 lbft and hold one steady while you turned the other one, your reasoning would have us believe that the one you're turning would click sooner than the one being held still since some of the torque being applied by the wrench is absorbed by the extension. Once again as in the first example, the wrench that is doing the turning is only a matter of perspective, so you'll find that both wrenches will click at exactly the same time. Hopefully you see that whether it's a motionless torque wrench at the other end, or you replace it with a bolt or anything else, it will experience the full torque applied by the wrench being turned.

The second concept you would have us believe is that the longer the shaft, the larger the amount of force absorbed by the shaft and therefore the output torque is propotionately less at the other end. To understand why this can't be the case, then consider the example of the garage door spring again. If you were to compare two springs side by each, one that is one foot long and the other that is two feet long, it's obvious that you understand you would have to turn the two foot spring with your right hand twice as many turns before arriving at the same torsion on both springs. Consider then that the springs are wound up, one twice as many turns as the other, if you were to replace either hand by a torque wrench on the long spring, your reasoning would have us believe that the right hand wrench would measure 50 lbft, but left wrench would measure something less. Furthermore, your reasoning would also suggest that if you were to repeat the same experiment on the shorter spring, there would only be half that difference between the two ends. I've already argued in the first experiment that torque IN equals torque OUT, so ultimately if you apply 50 lbft on the short shaft or the long shaft, the two different length shafts will output the same torque. If they output the same torque, then the length of the shaft has no bearing on torque.

You could argue that, "aha! it will take twice as many turns on the garage door spring that is twice as long and THAT'S what I'm talking about". The only result here is the delay in achieving the torque on the longer spring until it is wound up. The thing is, an axle has a significantly higher torsional spring rate than a garage door spring. So when we're talking about say 170 lbft of torque on a stock V6 Fiero axle, the short axle might twist in the order of several thousandths of a degree and the long axle maybe four times that amount if it's the same diameter and about four times as long. During the infinitesimally small amount of time it takes either shaft to twist that amount, the difference is taken up by a very very small rotation of the planetary differential gears. And I argue it would have no effect on torque steer. Feel free to find fault with any of my arguements.

“It’s not what people don't know that hurts them, it's what they know that ain't true.” ~~ Will Rogers

quote Originally posted by joesfiero: So right there you are not only admitting that a twisting shaft absorbs energy, you are stating it in your own facts.

I agree that a real-world axle shaft will twist slightly (small fractions of a degree) when torsionally loaded and thus temporarily absorb some energy, but it does not "absorb" torque. Torque (i.e. the rotational equivalent of linear force), energy (i.e. torque x angular "distance"), and power (i.e. energy / time) are not the same thing. A torsionally-loaded shaft "absorbs" energy because it deflects (i.e. twists) elastically, not because torque somehow disappears along its length.

quote I never said the [MIT] studies were wrong, just that you were interpreting them wrong. ... What it seems to me they are saying is that their studies are based on the shaft NOT twisting, in which it would show contradictory results of the torque as you move down the shaft.

The source is a textbook, not a research paper. What do you not understand (or have I misinterpreted) about the clear and unambiguous statement that "... the torque does not vary as we move along the axis of the shaft?" The earlier passage that you quoted from the same source was in the context of infinitesimally-thin cross section slices (in this case they would be no more than one atom thick), which is common in scientific and engineering analysis and forms a fundamental basis for integral calculus.

quote So I take it you didnt try the torque wrench thing?

For the record ... I've done numerous similar experiments, using far more precise measuring instruments than a workshop torque wrench, in high school and college physics labs, as well as in real-world engineering situations. I can attest that the fundamental laws of physics and physical mechanics are rigidly enforced. (Quantum mechanics? I'm still not so sure. )

quote Originally posted by Bloozberry: Example 1 ... Example 2 ...

Good examples, well stated.

[This message has been edited by Marvin McInnis (edited 12-16-2009).]

I thought you boys might like to see this one

http://www.specialpatrolgro...y/torque/torque.html

I may be in late on this one, but using an extension I have always been told that it reduces the final torque.

Scooby, the extensions you are talking about change the moment arm and are not at all the same thing as what is being discussed here. We are talking about extensions that move the torque wrench further away from the bolt vertically.

Force x distance = torque.
Does not apply ???????????

Oh God... am I really that bad at expressing myself. Scooby, the link you provided in your email refers to attaching a lever or crows foot, at the end of your torque wrench which effectvely lengthens the handle of the wrench or the moment arm. All of the discussion previous to your post deals with sticking a common extension axially between the head of the wrench and the socket... no leverage increase, no increase in moment, no radial displacement of the head away from the bolt, just plain and simple sticking an axial extension to allow you to reach a bolt that's further down a hole. Capiche? By the way, the equation is not force X distance... it is force X moment arm. An axial extension on a wrench does not change the moment arm, in case you weren't aware.

[This message has been edited by Bloozberry (edited 12-15-2009).]

I am not trying to get in the middle of this one, but torque law is applied to all torque measurements not just the one explained in the link I posted.
My intent was to state that torque law is law whether it is applied to a head bolt or an axel.
I have been watching you guys go at this and thought I’d include this simple torque formula.

Great arguments by the way and good entertainment to boot.

quote Originally posted by Bloozberry:

Example 1.
This is erroneous because a garage door torsion spring is a wound coil spring, it is completely different from a twisting shaft. You are comparing apples to oranges here.

Example 2. Have you tried this? I have physically tried my torque wrench example and have seen the results. I am really tempted to try this out but lack the proper equipment to get 100% accurate results.

Pretty much everything I have been saying thus far is in here.

from my link:
"CAUSE & EFFECT OF TORQUE STEER

The transaxle in a typical front-wheel drive car with a transverse mounted engine is offset to one side to physically accommodate the engine within the engine compartment. This means the position of the differential gears is also off-center, which requires unequal length driveshafts. Unequal length driveshafts, however, have been found to deliver more torque to the wheel with the shorter driveshaft (which is usually the left wheel). Under light loads, the difference is insignificant. But under hard acceleration, the wheel with the shorter shaft gets more torque, causing that wheel to pull harder than the other wheel. The stronger wheel tends to pull ahead of the other wheel, which creates the induced steering pull towards the opposite side. Thus the direction of torque steer in a vehicle with a left mounted transaxle is usually towards the right."

-Joe

quote Originally posted by joesfiero: Example 1. This is erroneous because a garage door torsion spring is a wound coil spring, it is completely different from a twisting shaft. You are comparing apples to oranges here.

Sorry, but the gloves are coming off this time Joe... you couldn't be more wrong, they behave in exactly the same way... they only have different torsional stiffness.

quote Originally posted by joesfiero: Example 2. Have you tried this? I have physically tried my torque wrench example and have seen the results. I am really tempted to try this out but lack the proper equipment to get 100% accurate results.

I was expecting you to argue the rationale, but seeing that you think comparing the torsional garage door spring to an axle is akin to comparing apples to oranges, it's not surprising that you are still clinging to your misconception. As I've asked before, do you honestly believe that if you had a long enough axle, you could keep any engine of any torque rating from turning using your bare hands by grabbing the other end of the axle? This is what you are technically arguing, with a serious face.

quote Originally posted by joesfiero: Unequal length driveshafts, however, have been found to deliver more torque to the wheel with the shorter driveshaft (which is usually the left wheel). Under light loads, the difference is insignificant. But under hard acceleration, the wheel with the shorter shaft gets more torque, causing that wheel to pull harder than the other wheel. The stronger wheel tends to pull ahead of the other wheel, which creates the induced steering pull towards the opposite side. Thus the direction of torque steer in a vehicle with a left mounted transaxle is usually towards the right."

So you found a web site that promulgates a popular misconception. Bravo. Keep reading your article to the end though, and you'll find that their solutions to solving torque steer are to mount the driveshafts so they run at the same angles. Interesting. Now where have I heard that before? One cure they say is to use an intermediate shaft. Even more interesting because according to your own theory (and theirs that longer axles transfer less torque) this wouldn't cure a thing would it? Putting a joint in an axle doesn't make it shorter or stiffer. The reason it DOES cure torque steer is because torque steer is caused by differently ANGLED axleshafts. The rest of the article talks about suspension bushings which I agree with and have covered in my earlier posts, and off-centerline thrust due to traction issues which they mistakenly attribute as a cause of torque steer. There is such a mix-mash of truths and misconceptions in this article that it's hardly proof of anything more than their ignorance.

But who am I to say... what's twenty two years as an aerospace engineer developing and managing gas turbine engine test facilities for P4 Orions, Hercules, Twin Hueys, Sea Kings, Buffaloes, and CH46's, oh, and then moving on as the senior aircraft maintenance engineer on Canada's military Bombardier Challengers, F-5, T-33, and Snowbird demonstration team aircraft? Tomorrow I'll post the math of torque steer as I should've done at the outset, not necessarily for you... but for anyone in the future reading this post that is interested in an incontrovertable engineering perspective, the laws of which will hold true as long as you're not nearing the edge of a black hole.

How about this paper published in Power Tech Proceedings, 2001 IEEE Porto (IEEE Xplore is a digital library providing full text access to the world's highest quality technical literature in electrical engineering, computer science, and electronics. IEEE Xplore contains full text documents from IEEE journals, transactions, magazines, letters, conference proceedings, standards, and IET (Institution of Engineering and Technology) publications.)

Directly from the text:
"Machine shaft torque is proportional to the shaft twist, which is caused by the angle difference between both ends. However, observing the angle difference in an actual generation system is difficult. Since measuring the rotational speed of the shaft ends is relatively easy, shaft torque is estimated effectively from the rotational speed for shaft torque reduction."

Or how about this patent, it has to do with measuring the relative torque from two different points on a rotating shaft. Now why would they have to do that if the torque were the same in both points???????

Directly from the text:
"A method and apparatus for continuously monitoring torque in a rotating shaft. The invention includes a pair of electromagnetic probes in fixed proximity to the shaft and spaced-apart in the direction of the shaft's longitudinal axis. The probes are operable in two modes. In the first, each probe is excited to cause a fixed magnetic pattern to be induced onto the surface of the shaft along a circumferential line adjacent to the probe. In the second mode, movement of the magnetic pattern due to the shaft rotation induces a signal into each probe. Each signal is indicative of the instantaneous angular velocity of the shaft at the corresponding shaft location. The two induced signals are conveyed to phase detection circuitry which produces a signal indicative of the phase relationship between the two input signals. The electrical phase relationship is a direct measure of the twist in the shaft, proportional to torque."

In fact, torque sensors have been used for many years to measure the torque along two points of a shaft in machinery such as large industrial generators and even gas turbine engines. Are you saying that these engineers have been wasting their time?

So what have we learned here? That there is a difference in torque along two points of a rotating shaft

Listen, I am tired of saying it. I NEVER STATED TORQUE LOSS THROUGH A SHAFT IS THE ONLY REASON FOR TORQUE STEER. Only that the manufacturers of cars have determined that it is a factor.

quote As I've asked before, do you honestly believe that if you had a long enough axle, you could keep any engine of any torque rating from turning using your bare hands by grabbing the other end of the axle? This is what you are technically arguing, with a serious face.

So I take it you didnt try your own example? To answer your question, no, you are thinking of the torque coming from the engine, down the shaft and into my hands which would effectively stop the engine? Are you not in this conversation? My hands would have to be able to hold the force of the shaft twisting, so no, I cant stop a driveshaft from twisting with my bare hands. With a driveshaft made of, say, rubber or plastic, I could effectively stop the end that I am holding until the shaft loaded up too hard for me to hold, or the shaft break. Remember, the more the shaft twists, the more energy it has in it trying to untwist. With the motor spinning, and the end of the driveshaft firmly in my hands, one of the two would happen, do you agree?

So have I answered your question? Have you tried any of the exercises I have recommended? How about the one YOU recommended with the two torque wrenches? I dont think so, so until you have tried them, any conclusions that you may come up with are purely speculation.

-Joe

Thanks for all of your additional examples to support the opposite of what you are saying there Joe. If you had understood what the articles were saying you would've realized that they weren't helping your case.

Your quote: "Now why would they have to do that if the torque were the same in both points???????"

What they are saying is that the amount of torque on a shaft can be calculated using sensors that measure the ANGULAR displacement between any two points on the shaft. Whether those two points are taken at one end of the shaft, in the middle, or at the other end, the amount of twist will be the same between two points that are equally spaced apart, anywhere on the shaft, hence, so will the torque along any point.

Now on to the stopping the engine example...your quote: "My hands would have to be able to hold the force of the shaft twisting, so no, I cant stop a driveshaft from twisting with my bare hands. With a driveshaft made of, say, rubber or plastic, I could effectively stop the end that I am holding until the shaft loaded up too hard for me to hold, or the shaft break."

But I thought you said there are torque losses along the shaft? Theoretically, according to you, there should be a shaft long enough to absorb all the torque from the engine and have nothing at the other end... ever.

quote Originally posted by Bloozberry: Thanks for all of your additional examples to support the opposite of what you are saying there Joe. If you had understood what the articles were saying you would've realized that they weren't helping your case. Your quote: "Now why would they have to do that if the torque were the same in both points???????" What they are saying is that the amount of torque on a shaft can be calculated using sensors that measure the ANGULAR displacement between any two points on the shaft. Whether those two points are taken at one end of the shaft, in the middle, or at the other end, the amount of twist will be the same between two points that are equally spaced apart, anywhere on the shaft, hence, so will the torque along any point. again, from the quoted website "The electrical phase relationship is a direct measure of the twist in the shaft, proportional to torque" Now on to the stopping the engine example...your quote: "My hands would have to be able to hold the force of the shaft twisting, so no, I cant stop a driveshaft from twisting with my bare hands. With a driveshaft made of, say, rubber or plastic, I could effectively stop the end that I am holding until the shaft loaded up too hard for me to hold, or the shaft break." But I thought you said there are torque losses along the shaft? Theoretically, according to you, there should be a shaft long enough to absorb all the torque from the engine and have nothing at the other end... ever.

Maybe you should read those websites again. They are measuring the torque on two points along the longitudinal axis of a shaft to calculate how much the shaft is twisting under load. In these cases they take readings under low load, and readings under a higher load. The reason they do this is to check the condition of the shaft to make sure it is not twisting too far which could eventually break causing damage to the machinery.

And what is it that you dont understand about my explanation of your driveshaft question? The engine produces torque, it is transmitted through the driveshaft. If you have enough clamping force at the other end of the driveshaft, you could stop THAT END from spinning. The longer the shaft/the softer the material, the easier it would be to stop the other end of the shaft from spinning. Of course that would only last long enough for the shaft to load up too much to hold or the shaft break.

And using your own example:
If you do in fact have a shaft made of rubber, attached to an engine on one end and in your hand on the other end, are you telling me that if the engine applies 100/200/300/even 400 ft lbs. of torque that you wouldnt be able to hold the end of the shaft for the time the rubber is twisting until it either loads up to much for you to hold or the shaft breaks? So what you are saying is that the entire 400 ft. lbs. of torque is on the end of the shaft you are holding?

Now, using that example again, if you applied that 400 ft. lbs. of torque to that rubber shaft, which is long enough and soft enough to twist with your bare hands 180 degrees, and only used that 400 ft. lbs. to rotate the one end of the shaft 180 degrees, that you couldnt hold the shaft from spinning? If you can twist it with your bare hands 180 degrees, then why when there is more torque applied at one end, but the same amount of twist can you no longer hold it?

-Joe

You misunderstand that initially, the engine in the rubber band example no matter how powerful, doesn't produce torque until it is loaded. It wouldn't produce 400 lbft of torque initally. It would produce very little torque because it wouldn't feel any resistance at first, just like you holding the other end. As the band twists the engine starts to produce more and more torque which it will happily do until it reaches the max amount of torque the engine is rated for at that RPM. It happens to be the same amount of torque that you need to keep it from spinning at the other end... it gradually takes more and more strength on your behalf as it twists as you know. If it required less torque on your end to hold it, then you would observe that the rubber band is more tightly twisted nearer the engine and less so at your end. But it's not. This is what they are saying in the articles you posted. The more the torque, the more the shaft or rubber band will twist, the more you can measure a change in the angular displacement of any two points on the shaft. The greater the angular difference in the two points on the shaft, the greater the torque on the whole shaft, and NOT the difference in the torque ALONG the shaft.

Anyway, as promised, here is the math of torque steer, after which Joe, you can choose to believe what you want. Thanks for keeping it civil.



Two shafts transmitting a torque T intersect at an angle “a” through a constant velocity joint shown above (the view is from the front of the vehicle). The force “F” is transmitted from one to the other CV joint at the two points of intersection, so for either shaft:

T = 2Fr

But the forces F also have moments about the vertical axis XX at right angles to the wheel hub. This couple “C” about the XX axis is given by:

C = 2Fe

but e = r tan a/2

so C = 2Fr tan a/2

or C = T tan a/2

If the input shaft torque T is clockwise (viewed from the input end) then the upper force F applied to the hub shaft will be pointing out of the screen (or paper), and the lower force will be pointing into the screen. Therefore the couple “C” will be clockwise viewed from below. This couple acts on the wheel about the steering axis and is equal and opposite in nature between the left and right wheels when the angle of the axle shafts is the same. They cancel each other out through the steering rack. When however, the axle shaft angles are unequal, due either to static or dynamic suspension geometry such as when weight is transferred to the rear of the car in acceleration and thereby raising the front of the car and changing the axle angles, the couple on shorter axled wheel will exceed that of the couple on the longer axled wheel. This inequality through the rack results in the steering wheel being pulled in the direction of the longer axle. Viewed from the rear in the picture below, C (left wheel) = T tan a/2 > C (right wheel) = T tan B/2.



Edited for speeeling.

[This message has been edited by Bloozberry (edited 12-17-2009).]

On the 88 rear, I had/have a problem with the long bolt that goes through the knuckle. The bolt was loose...the nut was tight but the bolt was not tightenening up on the bushings. It was stuck in the hole at a certain point. I cleaned the bolt/hole and it is way better. Now I think that the hole in the knuckle is worn, and allows pivoting of the wheel which presents exactly like worn bushings. I suspect this is what you might have if your components are new like mine (Poly on the rear, and "torque steer" with application and removal of throttle). It is now way better than before, but not completely eliminated. If I grabbed the wheel at 3 and 9 and push pulled I could get movement (Really bad) and now I can't feel any, but I know there is some. The change in toe is responsible for the 'steerage'.

I also had a ride height issue. I had to increase it to allow for clearance between the altenator bracket and the the pass side shaft...it was rubbing. I think that because my engine mounts are weak that the engine pivots when on throttle and contacts the shaft, thus robbing a bit of torque, et voila.

I have yet to test drive it because there is a snow on the ground!

C

Doublec4, I sincerely apologize for crumbing up your thread with this back and forth garbage. You made a thread looking for an answer about a specific topic and somewhere along the line my post to your original topic was misconstrued and it has been one big downhill tangent from there.

I would prefer to just stay out of the rest of this thread and hope that there are others out there who can post correct information because neither I, Marvin, or Blooz are going to change our minds anytime soon. I hope that you have at least learned a little bit about your original topic and you can take or leave whatever opinion you want.

-Joe

Edit to add: I think I have found the source of the confusion regarding torque.
Blooz and Marvin argued that the torque along a shaft is the same, this is correct if you only take the torque of the shaft itself.
Take a shaft and hold it fixed at one end, then rotate the other end 90 degrees, say it takes 10 ft. lbs. to do so. The torque along the shaft is going to be 10 ft. lbs. along the entire thing which is why they twist evenly and not bunched up at one end.

Now, if you apply 300 ft. lbs to that shaft with a load on the other end and the shaft twists that same 90 degrees, using 10 ft. lbs. of torque to do so, you will have an outputted torque of 290 ft. lbs. from the shaft. Hence, what I was saying before about having different amounts of torque on the hubs. Not by much, but different.

Torque in - torque used to twist a shaft = torque out.

[This message has been edited by joesfiero (edited 12-18-2009).]

Bloozberry, thanks for posting that. I'm going to try understanding it during the day when I'm alert. "+"

You guys should go get a Mechanical Engineering degree.
That's what I did. (Hence the name)

So, after beating this topic to death did we ever answer the question? I say; YES, the Fiero can exhibit some form of torque steer. However the effect is so small compared to FWD that it is not a concern, and it is due to the slight difference in torque delivered to the rear wheels overcoming the caster in the front. In theory, it should cause the car to pull slightly left on acceleration, only if the axles aren't straight when you squat the rear. If you are holding the steering wheel at all, you shouldn't notice much.

Now, if your rear suspension is messed up, such as the infamous 88's rear long bolt, or your 84-87's rear ball joints, all bets are off. The rear suddenly does its own steering on acceleration. The funny part is how it doesn't actually effect your steering wheel, but does change the direction of the car, making you need to steer out of it. My 88 did this after installing polly, which promptly caused my long bolt to loosen itself. I wound up double nutting the bolt, and never had a problem with it since.

quote Originally posted by Fierobsessed: So, after beating this topic to death did we ever answer the question? I say; YES, the Fiero can exhibit some form of torque steer. However the effect is so small compared to FWD that it is not a concern, and it is due to the slight difference in torque delivered to the rear wheels overcoming the caster in the front. In theory, it should cause the car to pull slightly left on acceleration, only if the axles aren't straight when you squat the rear. If you are holding the steering wheel at all, you shouldn't notice much. Now, if your rear suspension is messed up, such as the infamous 88's rear long bolt, or your 84-87's rear ball joints, all bets are off. The rear suddenly does its own steering on acceleration. The funny part is how it doesn't actually effect your steering wheel, but does change the direction of the car, making you need to steer out of it. My 88 did this after installing polly, which promptly caused my long bolt to loosen itself. I wound up double nutting the bolt, and never had a problem with it since.

"Torque steer", strictly speaking, is only relevant to front wheel drive vehicles. Torque steer refers to the unbalanced forces transmitted to the steering rack under torque. The driver must hold (fight with) the steering wheel to resist these unbalanced forces. That's why the discussion led to FWD.

So for a simple answer, no, a Fiero cannot suffer from torque steer.

Any moments around the kingpin axis are resisted by the tie-rods, which are solidly attached to the cradle. The rack is completely not related to the rear suspension.

The rear wheels might be subject to toe changes due to compliance in the wishbone bushings. But strictly speaking, that is not torque steer.