What's new
CK5
Register an account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members.

Dealing with thrust loads in drive shafts under high torque

BadDog

BadDog

SOL
Staff member
Super Moderator
Author
Joined
Feb 24, 2001
Posts
8,003
Reaction score
232
Location
Phoenix, AZ
Stephen brought up a very good point about binding in a drive shaft slip joint due to high torque loads. Here is the post where it started on 7/6/04.


Could it be that this is what is breaking many seemingly well mounted cases, adapters, and/or transmission housings? Many times it's obvious why a break occurs. Impact, loose bolts, or bottoming drive shaft and the like. But what about the other times like my resent transmission housing breakage where none of these were at play?

This is likely worse in my case due to running square drive shafts, but does it really matter? I see no signs of wear or galling.

Other than making compromises in a linked suspension geometry to limit shaft length changes, what can be done to reduce this load?

Is there more that can be said? If so, lets put it here instead of taking Neil's post further off topic...
 
Did you have some way to grease your square shafts? I know I've seen pictures but I have no idea how much friction you had there. Everything was greased, right?
 
Nope, the information I got indicates grease is bad on square shafts. I had originally planned on installing zircs for greasing the shafts, but most of the problems I heard related to square shafts, the problems were a direct result of contaminated grease. Since there is no way to seal them, most just run them dry, at least out here in the SW. Also, there is about 18" or more of overlap between the tubes, so there is very little slop or leverage across the coupling. It didn't seem to be a problem, at least not up until I broke the trans...
 
I just looked at it more closely in the daylight. It's smooth as a baby's butt along the wear surface. No galling or signs of wear/stress at all. Though it does show paint rubbed off much further up than I expected. Looks like the axle went about 6" forward when the spring bent. That alone would have bottomed most shafts and probably broken the trans or done other damage, but mine could have taken another 4" or more. And it shows no signs of wear at all.

One thing I did notice is that there is a pretty good impact gouge on the shaft (well, a good scuff really, but usually they hardly mark at all). I never noticed that before. They are tough as nails, but I wonder now if maybe I hit it and didn't realize it while assaulting the waterfall. That might be another explanation for the breakage... Spinning with high torque on the shaft and a side impact just to make things "interesting". <sigh> The post-mortem on my breakage is not providing much in the way of answers. Still it's an interesting topic.
 
Here's a thought:
assume you can put about 3000 ft-lbs of torque on the rear driveshaft
that's 36000 in-lb
At the radius of your drivshaft (assuming 3" tube) you have 24000 lb of force spread over the 4 contact areas.
With a 6000# clamp load at each interface, you're going to have a hard time compressing the driveshaft.

This makes a lot of assumptions but even if the numbers are off by 50%, it seems like it will be hard to compress the driveshaft very far. It would probably be hard to compress any driveshaft, maybe this is a bigger deal than we think?
 
Hmmm, wow... /forums/images/graemlins/eek.gif

But wouldn't that be beyond the yield strength of the mild steel so that you would see galling, indentations, or other deformations?

And assuming (as I am) that this is what broke my trans recently, other than a suspension change which is long overdue, how could it be addressed? Seal and grease? Or is that still going to be too high? Some sort of polished bearing surface (well beyond my means)?

Even with a suspension change it's just a matter of distance changing with no way to eliminate the thrust load. Or is the "soft mount" argument going to help deal with that? I sort of doubt it given the acceleration and inertial resistance.

Also, spline shafts should see similar issues other than the fact they are greased. With the splines at a smaller diameter, the torque load is much higher, and the bearing surface contact length is only a few inches in most cases. They are certainly much closer tolerances and tighter fit with more individual surfaces, but is it really that much better? This exceeds my ability to compare with any mathematical accuracy. Especially when you throw in different numbers of contact surfaces at different offsets/leverage with different load angles, grease, contact length, different finish and tolerance, etc…

My head hurts... /forums/images/graemlins/crazy.gif
 
Here are some numbers that don’t sound quite right. I’m trying to put it together from the various web sites. Anyone please point out any mistakes or incorrect assumptions.

To start we need a torque number. Since most failures are at extreme loads lets do worst case and use 300 lb-ft engine torque (what Chilton lists as peak torque for a TBI 350).

300 lb-ft x 2.5 torque converter reduction x 3.06 first gear x 2 203 low range x 1.98 205 low range = 9088 lb-ft maximum torque on the driveshafts (those gears ratios are from memory, but should be close enough). Now the question is was all the torque going to just one shaft or split equally between the two. Lets do worst case and put all the torque to one shaft.

Now to convert torque into applied force. Assuming a 3" square inner shaft, the distance from the center of the tubing (center of rotation) to the beginning of the corner radius on the tubing is about 1.956" or 0.163 ft.

So 9088 lb-ft / .163 ft = 55,755 lbs of force pushing the surfaces of the tubing together.

Using the coefficients of friction for steel to steel produces the following:
Dry steel on steel .6 static = 33,453 lbs of thrust
Dry steel on steel .4 sliding = 22,302 lbs of thrust

Lubed steel on steel .1 static = 5,576 lbs thrust
Lubed steel on steel .05 sliding = 2,788 lbs thrust

Thoughts?

Friction links: http://regentsprep.org/Regents/physics/phys01/friction/default.htm
http://www.roymech.co.uk/Useful_Tables/Tribology/co_of_frict.htm#coef

general good info http://www.epi-eng.com/BAS_TOC.htm
 
With the clamp load spread out along the 18" of slip, the point pressure could easily be below galling (even mild steel) but the overall total on friction (caused by the clamp load) would still be greater than the strenght of the mounts, or the actual case. I can see this becoming a big issue when tracion is high along with torque. In effect when you were going up the waterfall the clamp load made your shaft into a single link (you already knew the springs would collapse first) that had the weight of most of your buggy supported on it and the relatively small mounting hardware (tranny mount) for said *link*

This would explain why HD drivelines often use coarse splines (less overall surface area)
 
I think that is certainly possible. Enough friction to cause damage from it not changing lengths, but not quite enough to cause galling of the steel shaft.
 
are we overlooking the fact that 9,000 lb of torque at the shaft is gonna be o, like 46,000 after the diffs (5.13)...
at 46,000 ftlbs... wont the tires spin? 554,000 inlbs/4 tires 138,510 inlbs per tires/ 42"= 3,297... ok, i guess i answered my own question.... if all the weight is on one tire, and the tire is only exerting 3300 lbs force, then the truck wont move... right? /forums/images/graemlins/thinking.gif edit: if you are headed straigt up in the air and under no acceleration.. /forums/images/graemlins/smirk.gif
 
Hmmm, all good points... /forums/images/graemlins/thinking.gif

[ QUOTE ]
are we overlooking the fact that 9,000 lb of torque at the shaft is gonna be o, like 46,000 after the diffs (5.13)...
at 46,000 ftlbs... wont the tires spin? 554,000 inlbs/4 tires 138,510 inlbs per tires/ 42"= 3,297... ok, i guess i answered my own question.... if all the weight is on one tire, and the tire is only exerting 3300 lbs force, then the truck wont move... right? /forums/images/graemlins/thinking.gif edit: if you are headed straigt up in the air and under no acceleration.. /forums/images/graemlins/smirk.gif

[/ QUOTE ]

Straight up? Do we have artificial gravity holding to the wall now? /forums/images/graemlins/laugh.gif /forums/images/graemlins/histerical.gif But that's a good point too. With the gear multiplication, countered by the leverage of large diameter tires, how does that affect everything. The leverage of large tires is what breaks strong axles before their friction can be overcome to spin them. Boy, this is an interesting problem...

One theory:
In my case, coming down after "getting air" and wrapping 4.5 psi 42" TSLs around a variety of rocks with the entire weight of my rig along with deceleration of body mass coming down (inertia) virtually eliminates the possibility of slip/spin, so the torque recoil would have shot back up the drive line looking for an escape. With the torque converter spun up, the transmission case was the "path of least resistance".

Second theory:
All that torque along the drive shaft caused much more "bind" than I ever imagined. But rather than torque directly ripping off the back of the transmission housing, the bound up shaft became a rigid "link" that was pushing (and perhaps alternating pulling) against the transfer case. Add in a spring bending and allowing the axle to move forward further and faster than normal, and something had to give.
 
[ QUOTE ]
The post-mortem on my breakage is not providing much in the way of answers. Still it's an interesting topic.

[/ QUOTE ]

If I may, let me throw a cup of liquid on this fire and see what happens...

You were bouncing your truck pretty much vertically. You've got a Doubler bolted to the back of that T350, lots of heavy cast iron there. The passenger-side drop of the 205 becomes a sort of pendulum when the vehicle is oriented vertically, with the weight of it "pulling" more on the pass. side of the trans housing. With the bouncing, the weight of the Doubler is really shock-loading the tranny case (I'm assuming you don't have some sort of brace on the 205's drop portion like the 208 has) on the landings, trying to wrench the tailhousing off in a clockwise direction (looking at the drivetrain vertically). This, combined with torque loads and maybe a rear driveshaft that's not telescoping as it should, is what caused the transmission housing to split in that kinda-spiral-looking fashion.


Comments? /forums/images/graemlins/ears.gif Flames? /forums/images/graemlins/doah.gif Problem solved? (yeah right... /forums/images/graemlins/rolleyes.gif)
 
Yeah, these points have also been considered at one point or another.

And most of the inertial torque load from bouncing vertically would be parallel to the (normally) horizontal plane roughly defined by the frame rails. Still something to consider, but as a bending load on the transmission case rather than as a torque load as it would be if the truck were, say, bouncing along a bolder field or whoops.

Also, the doubler weight distribution and loading characteristics should be (I think) roughly the same as a 203.

So it may have contributed over the long haul with fatigue to the case, but I don't think it directly caused the, break, but that's just an opinion...

FWIW, the actual catastrophic failure which left me stranded occurred on the obstacle after the big waterfall where I bent the spring, and the event that produced it was backing off (and dropping the front off of) the small fall to take a different line. That would have been an inertial torque load perpendicular to the drive train just as you describe…


And another thing to add...

It just dawned on me that the last time I ripped the transfer case off the back of the trans by stripping the bolts, I was climbing a nasty waterfall then too, AND I bent a spring! /forums/images/graemlins/eek.gif At that time I didn't have on a skid, and I saw evidence of impact on the front drive side of the case, so I "assumed* (yeah, I know) that was the culprit. It didn't even start leaking till I got home, so who knows. But now I wonder if this drive shaft bind and higher loads due to acceleration from bending a spring are the culprit in that break too. I starting think my initial post mortem on the stipped bolts was off base...
 
If you impacted the square shaft, you'd also need to figure any difference in circular resistance due to it's shape.

Ryan
 
[ QUOTE ]
Some sort of polished bearing surface (well beyond my means)?

[/ QUOTE ]

kind of....
what about a dry film lubricant?
you say you are getting wear in areas that you can see due to paint loss. a spray on dry film is like metal once cured properly. it even wears like metal. flakes off after time, but inexpensive in the long run. it can also take a bit of slop out of a slip fit setup suck as a square drive shaft. its kind of like lead based graphite in a way.
just a thought and some input.
Grant
 
Russ... if you impacted your shaft, and its square, wouldnt that make an incredible amount of side load? Possibly one of the corners caught the rock, and with the torque on it at the time, it tried to 'walk' itself sideways, but broke something instead? A round shaft woundlt produce any side loads like that at least.

Just a thought from this unedumaticated Canuck! /forums/images/graemlins/smile.gif I may fix F-18s, but I didnt design them hehe.
I also dont know how exactly the break happened.

Mike
 
[Edit] Just realized I made a mistake when reading the comments on "impact". I first read it to mean that you thought it had distorted, which it didn't. But I guess you really meant it's "impact" on the torque loads (pun intended)...

The "impact" really left nothing more than a good "scratch" with no visible deformation, though it might be to small to see. And the shaft has a small tendency to "walk" on the square edges as demonstrated once after sliding off a line and landing on it with the rear tires effectively in the air. But I don't think that's what caused this problem. At this point, after several threads discussing different options, I'm also leaning toward "option 2" above, at least for the moment.

Grant:

That sounds interesting. I've seen stuff like that, basically paint with graphite in it and the like. Do you know of a particularly good brand or good value?

Anyway the "wear areas" have not real "wear" visible at all. It's just that the paint is gone. And that actually happened the first time before I ever installed them just putting them together on the bench. Messed up my perty paint. /forums/images/graemlins/rolleyes.gif So I sprayed them again after I got them in place, which is how I could tell how far they had compressed. Anyway, I had been watching that "wear area" and thought all was good because it seemed to barely keep it rust free with no signs of wear. Never even thought about the distribution mentioned earlier causing significant bind without visible wear. /forums/images/graemlins/doah.gif
 
What about something like a dry Moly lube. It is a big thing for reducing friction in rifle bores and might work in a driveshaft if you could get it to adhere to the inner tubes walls. I don't know how long it would last though. Just a thought.

Harley
 
As an afterthought to my above post what about having the outside of the 2" inner tube hard chromed and use a dry lube on it.

I know chroming weakens the metal, but is .250 wall so it shouldn't matter.

Do you run the 2" tube to the tcase yoke or the pinion?

Just an afterthought.

Harley
 
You must log in or register to reply here.

Latest Posts

Top Bottom