Drive Shaft Issues

wehave covered this before

universal joints are unsuitable for large angles.

yes it matters how the uni joints are aligned and this was also covered,

did you fit the constant velocity joints as suggested ?

Peter,

You are right. I am looking into buying a CV joint as you had suggested before. That would solve the problem. Meanwhile, i am just trying to find out the root cause of the failure. I would like to convice myself, how the vibration analysis done and the large displacement made in the axial direction, could have caused the bolts to fail in tensile..

I dont think we discussed anything on the tube and spline position in the last discussion. Is there a good practice as to whether we should connect the spline or the tube side of the shaft to the power side, or it doesnt matter?

Thanks for your reply.

Vanuta Please send e-mail adress, I lost it and have to send a scetch of drive shafts.

I would suggest that you give more informations since there are some aspects not totally clear. In the picture you show there is a gear which is not indicated on the principal sketch. To break the bolts there are several possibilities. You say there broke under axial forces but we do not have any informations about the origin of those forces which can be generated by an eccentric force from the gear transmission. Forces could come from the cardan joint but not only, so that as long as the information is not complete an explanation cannot be done. Where goes the shaft through? What is connected via the gear? Where goes the outlet of the gear box? All said till now was based on your information but you bring now new aspects. It is well known that such joints induce a sinus function in the secondary shaft with an amplitude proportional to the cosine of the angle between shafts. If we look at your sketch it is not clear why only this bearing fails since the other bearing presents same angle. Rotational speed variations must not be an important source of forces generating 1mm amplitudes if there are not as well masses which in connection to compliances lead to such vibrations. As said already only a complete information can lead to correct answer.

To Reply your answer...

The gear box:

It is a typical hot-shift PTO that is mounted on a typical transmission PTO pad. The gears you see meshes to the transmission gear when activated. The other end is connected to a shaft ( slip-york type). The PTO is eventually driving a C-pump. There are three shafts ( slip-york) type between the PTO and the C-pump. The Shafts do not carry any other load, except its own weight. I dont know where exactly the axial force is coming and even there is some form of axial forces, how could it reach the bolts?

If the universal joints are not assembled at 90 degrees to each other the angle they can go through is very small if the shaft has been removed by sliding the shaft off the splines, and then reassembled by sliding the splined shaft back on because of the splines the shaft can be assembled in any position relative to the fixed universal joint, So its possible another person has stripped the assembly down and was unaware you have to reassemble so universal joints are at 90 degrees

this would give you the axial rotation that is causeing the problem

as i said in the last thread the shaft if not assembled correctly would instead of rotaing at a constant revolution would for 90 degrees go slow the f90 dg fast then slow etc and this rapidly changing torque would put the shafts under enourmous stress.

i will try and find a drawing to explain better but take it from me what you need is Constant velocity Joints and not universal joints as then you can assemble the shafts in any orientation.

Excellent answer.

I twisted wrenches for over 25 years and cannot tell you how many drivelines I found out of phase. These problems caused more bearing failures that could have been easily prevented is the driveline assemblies had been installed correctly. Many times the problems would come to me straight from the transmission shops. Someone would replace a transmission and then they would have a vibration in the rear axle. I was happy to fix them and many of the customers went back to the transmission shop for a refund of the cost of my bill. The worst one I had was a four wheel drive that had both front and rear shafts install 90° out of phase. You could not keep the thing in the road at 70MPH. That one was from the owner doing his own work. The money he spent on trying to get the vibration fixed, (he had it in 7 shops before mine), after he completed his work was more than the job would have cost if he had brought it to a qualified shop to begin with. CV joins are defiantly a better way to go if it can be done.

Very informative answer, and a GA from me. It is funny how many small details some of us take for granted that others may not know because we do them fairly often and it is almost instinct to align universals in phase. To tell the truth I don't even remember who taught me to align universal joints this way but probably the owner of a welding shop I worked at 25-30 years or so back when I was just getting started good in metal fabrication. We didn't call it "in phase" or have a name for it other than asking: "do you know how to line up them up"? and if they didn't know we would show them and they could pass the correct way on to more people in turn.

I guess I've learned a new word and definition for "line them up the same way" or "turn them the same way" like we said in the younger days.

look at this animation

http://www.youtube.com/watch?v=P2sPkEAt7OA&feature=related

http://www.youtube.com/watch?v=TGvKS4bHgTk&feature=related

The text will give you all technical informations you need:

http://www.csn.eu/fileadmin/user_upload/pdf/katalog2009/Technik2_2009.pdf

It is too long to copy and paste. It has every thing you could use to quantify the bearing loads and see if the bolts were broken due to the shafts or by an other "source".

I would appreciate if you communicate the results of your further investigations.

Hi Nick

I had come across one of the equation in the PDF catalog you sent.It is on chapter 4, pg 12. The equation states the axial force generated due to a slip. I am using a cardan-shaft (slip-yoke and spline type) with universal joints. I have an initial slip length of lets say xmm. After i power up the shaft, lets say the slip increases by ymm. What will be my value of Lu?? Is it just ymm or (x+y)mm. Your answer will be greatly appreciated.

if bolts broke in tensile, then its possible they were overtightened

or because universals were assembled incorectly they were subjected to yawing action of the flange imposed by the rocking action caused by the out of phase of the universial joints.

see my later postings for more information on out of phase

We do not know several aspects: were bolts of right quality? How high was the torque?

Which tool was used for the tightening? How big are the loads from the shaft on the bearings? What is the rpm? How big the transmitted torque? Did same bolt always break first?

If one bolt from the 4 breaks under a combined load of tension AND bending moment on the box the 3 other are loaded in such a way that they will very fast also break. It is for me still unclear what happened. I would like to see the broken sections and also get the information where the bolts were in the flange and how the shaft runs with respect to the flange. For an failure analysis as this one the whole information is still to limited.

It is also possible that the wall vibrated because a bolt was loose or did it vibrate as soon as the transmission started even if new bolts were in place?....so many possibilities and not enough information.

Yes not enough information, its always the same.

we are trying to diagnose with only a tiny amount of information, Hopefully he will work out whats wrong from all the tips we give him

We try to help but this way is dangerous as well for the OP as for us since we can compromise ourselves by giving an erroneous recommendation due to lack of information.

I sincerely think that it would be better as well for the OP as for us to give a suggestion or solution when the information is at least sufficient for a good understanding of the problem context. This is valid especially in such "expertise" or "forensic analysis" as the one we have been confronted.

That Rube Goldberg contraption is painful to look at. Why not mount the pump directly to the engine and just modify the piping?

The last detail that hasn't been addressed is that the the first output element attaching to the first driven shaft and the input element attached to the last output shaft must be in parallel planes. In order to do this in your diagram, the second support bearing must be raised so that the pto shaft and the pump shaft are parallel.

In addition, as was mentioned before, The first and third drive shafts must be in phase (as defined in the diagrams above) and the center shaft rotated 90 degrees relative to the other rwo.

Assuming that the shaft set has been dynamically balanced as an assembly, then using the proper spline lubricant should result in smooth, relatively vibration free operation.

Bob

Why go through all this? Replace drive shaft with a

hydraulic pump and motor.

Hi 58flh here, most all 2-piece shafts have what you call a carrier bearing, this handles all centrifical, inertia forces, as well as letting the shafts in place! without this part the as soon as you put the shafts together they will drop!!! I am going out on a whim here ,I dont know what kind of vehicle this is for ,or machinery! just trying to shed a little light on the problem. Just trying to help a fellow wrench! 58flh

the above was posted in the wrong place

i have copied it and placed it here

shaft = drive

tube = driven

Looking at your drawing I do not see the angles to be overly extreme for a universal joint to handle and you have the slip yokes installed to add to the flexibility of the shafts. I have to wonder if the quality of the bolts and other materials is correct for your application? Have you routinely greased these universal and slip joints or do you have grease fittings on the universal joint or are they the non grease-able type universal joints? It may be that you just need to go to a higher quality universal in the area you have problems with. Also you may just need to increase the size of the shafts and use a larger heavy duty joint. Most industrial drive shafts are made from chrome moly tubing with heavy duty joints.

We have hundreds of drive shafts on the 2 paper machines where I work and some of them are at greater angles than you show here, We very seldom have one go bad but then again we have a thorough greasing program and we hold the oilers accountable for the items they grease or oil so they have an incentive to do it correctly. When we do have a universal fail it is usually caused from paper or some other debris building up around the joint causing it to over heat and not to flex as it should.

Angles from one shaft to the other have to be the same, other wise you setup harmonics that could lead to your condition.

The 2 connecting joints need to have the same angle, most applications have single shaft, then the shafts the joints are connected to have the same angles or they vibrate.

Hi Vanuta,

The issue you speak about is typical failure due the secondary couple action caused by transmitting torque at an angle. Whenever you transfer torque at an angle there is a bending moment (force) trying to bend the supports at right angles to plane of the joint angle. This mainfests itself in your case by trying to pull out the bolts at the PTO pump case by bending it at the flange as you have seen.

The secondary couple loading is proportional to the angle of the joints via a sin or tan relationship (depending which joint you are looking at - driving or driven) and so at say 6 degree angle the sec couple (bending moment) maybe 10% of the torque then at 12 degrees becomes 21% while at 18 degrees is 32% of the torque.

Additonal to this is vibration due to inertial excitation of the shaft running at non constant velocity. Again at high angles the shaft vibration is proportional to the sin of the shaft angle which grows quickly with high angles. You may have correctly phased the yokes to get cv out of the final shaft but remember the intermediate shaft is running at non-cv and mulitplied by its mass give rise to oscillating loads and vibration.

My recomendation is use a cv joint. take a look at www.thompsoncouplings.com for high angle cv joints such as you are trying to do.

The second couple action is very interesting, could you please send more information about that? or are there any references to this phenomena?

In the document under the link I send to you ALL those informations are present. The way you question is a proof that you do not read what is send.

Why ask if you do not consider to use the informations?

Or do you want to get all done for you?

Nick Name,

I did read the article you sent. I do very much appreciate the help and the time you had spent in sending me the article. The article did not state how the secondary couple is being created and did not state anything about the secondary shaft. I might have overlooked. If you had seen, could you please let me know the page number?

The name secondary does not refer to a second shaft. It is used as a term akin to primary. ie the primary torque or couple is the twisting effort of the driveshaft itself causing motion. The secondary torque or couple is a result of the primary torque going around a corner. You could make a sipmle experiment if you got say a bent shaft of wire and then tries twisting it. There is a tendency then for it to bend at the supports. This is what we call secondary couple or bending moment.

The article was actually written by me years ago and is not published as such in any book. I did it to explain the concept to others in a simple way

Yeah..you are right..i had written as secondary shaft...it was a typo on my part..what i meant was there was not much explaination about secondary couple...It is actually a very good article...

OK you want to know more about secondary couples. I have added some text I wrote a few years ago about this:

It is pretty good explaination....i just have one question...why would the torque output vary over a revolution..the drive shaft is always rotating at the same speed as well as the driven shaft...?

Something like the outside wheel of your car traveling faster than the inside wheel round a corner while the speedometer reads the average speed.

one neads a lot of scethes to explain this!

maybe u can put a torsional vibration dampner to lighten the vibration at the drive shaft, closest to your pto...that might help... it might reduce the vibration due to the secondary couple...

Will a torsion dampner work?