Shaft Design to Avoid Breakage

Maybe G10500, hardened 600°F, from here?

how you chose?

I think it was the highest tensile strength in the table.

you mentioned G10500 is the highest tensile strength, but i think we need high bending strength here,, am i right??orr

They are related. In particular, bending puts the outer curve in tensile stress.

Also worth mentioning: To minimize overhung load, mount the drum as close as possible to the adjacent bearing.

How long was the drum roll when the shaft broke?

these the broken neck......

shaft and wire roll assembly actual position in machine.... the drum length 430mm.and there is a gap between drum assembly and bearing , u can see that in the pic.....

Hi RVZ717,

thx for your valuable comments. giving support on the other end is littlebit prob here boz of space constraints. we are planning to give support at the bottom of roller. 2 support flexible roller at both end of the wire roller.

I blew the picture up and there appears to be a chevron on the broken face.

Ron

Thx KJK, i will discuss on this with my boss.

GA. I agree. Rollers just under the outer edges of the spool.

It would be worth the effort. The lifetime of, not only the shaft, but the motor and bearings would probably be at least doubled.

They make these to support very heavy boats on trailers. Might just do the trick.

yes, we are planning to do the same Del, thx for your good comment.

The cat must get up early! That was my thought, too, but you beat me to it! -- JHF

AISI/SAE 4340 (BS En 24) Hardened and Tempered to 38-45 Rc.

GA - This is the best answer I've seen so far. 4340 is a bit tougher to machine than 1040 or 1050 in the soft state. But the alloy constituents that make it harder to machine also give it better fatigue strength. I'd lean toward keeping the hardness on the low side of that 38-45 Rc number by drawing it at a higher temperature. If you can't find 4340 AISI4140 will work almost as well.

There is another factor to pay attention to. That is surface machining flaws in the smallest diameter part of the shaft nearest to the first bearing. That is where the cyclic bending stresses are highest and stress concentrations from sharp grooves in the surface of the steel are most likely to be initiation points for fatigue cracks. The machining fillet radius where the shaft steps up in size (the shaft shoulder) to the bearing mounting diameter should be as large as possible as long as the wire drums don't dig into it whenmounted for winding. That fillet should be as flaw free as possible and ideally should be polished. This is how we build crankshafts for high performance racing engines.

That said the photo you gave us of the broken shaft seems to show the fracture at a point a few millimeters from the shaft shoulder. This leads me to suspect the flaw where the crack initiated was caused by sharp edges on the end of the center hole of the heavy drums digging into the shaft surface. If this is the case the original shaft design dimensions and material specs may be OK and just some care in the above mentioned areas may cure the problem.

So it may be wise for you to understand what material the original shaft was made from and most importantly whether it was hardened beyond the soft state at which parts like this are machined. An unhardened shaft will dent from the repeated drum installation much more easily than one that is hardened to 35-40Rc (common hardness for AISI 1040 power transmission shafts).

Note that if you choose to make a hardened shaft either carbon steel (1040 or 1050) or an alloy steel like 4140 or 4340 it will probably be necessary to cylindrically grind the bearing mounting diameters to size after heat treating to get a proper bearing fit. Otherwise your maintenance problems with this machine may change from shaft breakage to ball bearing failures.

In my opinion based on a long distance view of your problem I think the best solution to your problem will be to seek out a low alloy medium carbon steel like 4140, carefully machine it to the finished size, polish the abovementioned critical surfaces and do not heat treat it. If you can get even an approximate hardness level on the broken shaft near the location of the break and also a better photo looking in the direction of the shaft centerline we can make a better judgment of the best replacement shaft for this job.

Ed Weldon

Look of the shaft gives an idea that it had served quite a good life. If you are not satisfied with life of the shaft and the shaft size can not be changed, you can try changing material to better specs.

How ever it seems failure is due to bend and stress conc. at the step. If you can remove the step of mounting spool and put a sleeve to locate the spool, it may change location of bending moment stress concentration and shaft failure rate may change.

This is an unusual looking fracture compared with the usual examples in the following links:

http://www.asminternational.org/pdf/spotlights/jfap0502p011.pdf

http://www.plant-maintenance.com/articles/rcfa.shtml

http://cipem.uniandes.edu.co/web/images/stories/fatigue%20failure%20of%20a%20rear%20axle%20shaft%20of%20an%20automobile.pdf

I don't want to present myself as having any expertise in this field; but I suspect the broken shaft is made of an un-hardened plain carbon steel likely something like 1018 or 1020. To me the clue is the barely visible necked down zone in the shaft diameter immediately to the left of the leftmost fracture in the last closeup photo of the two broken pieces. So an improvement in tensile strength, which provides a comparable improvement in fatigue strength is probably a good idea. And I still favor the use of the molybdenum containing AISI 4000 series steels such as the most common, 4140, if it is available. For years we have had available a leaded version of this alloy, Rycut 40, that is significantly more machineable than the standard 4140 alloy and therefore is most suitable for the capabilities of the plant maintenance machinist. I'm pretty sure whatever patents Ryerson had on this alloy have expired and there are similar proprietary steels available from other sources. Here's one:

http://www.speedymetals.com/information/Material42.html

Ed Weldon

There is a groove in the shaft near the shoulder. Where did it come from? It looks like the shaft has been rubbed, probably by the drum. Is the drum slipping on the shaft eg when first running up to speed?

The cracking follows a line near the shoulder and also near that groove.

You have suggestions about the shoulder already, but you should also prevent that groove forming eg provide a coating on the shaft to cushion the drum. It will need regular replenishing/replacement.

Great. One can never over emphasise the the importance of fillets and surface finish and, sad to say, this is blatantly ignored by people who do not appreciate their value. I have seen a number of shafts fail in fatigue simply due to lack of fillet radius at the base of the key-way

I would agree to the lower hardness for 4140, but I have used 4340 Martempered or Austempered to 50 Rc, with great success, for shafts in Steel Mill, Pinion Stands.

Space limitations have forced me design shafts with DIN 30CrNiMo8 (Ni: 2%; Cr: 2%; Mo: 0.4%) and 30CrMoV9 (Cr: 2.5%; V: 0.2). These have tensile Strengths (UTS) in excess of 180 ksi, with high endurance limit.

Expensive, but great value for money.

I have looked fondly at the aerospace steel, "Ferrium M54" (UTS: 293 ksi !!), but the prohibitive cost has prevented me from using it down on earth.

It gave me great pleasure to read your response.

reaymenon -- Thanks for the data points! I am going to study both your applications in more detail. 4340 martempered to 50Rc may help me in a future project involving a landspeed racecar transaxle. Steel mill pinion stands are a rough use for any steel.

Ed Weldon

The advantage of Martempering & Austempering is that there is very little distortion after heat treatment, as thermal contraction and expansion to Martensite/Bainite occur at different times, unlike oil quench where both take place simultaneously. This results in very low internal/residual stresses, unlike oil/water quenching, and gives higher endurance life and greatly reduces fatigue failure.

If you need exceptional Impact strength and toughness, as your application indicates, go for Austempering. You do not have to temper it again. The final phase is Bainite and not Martensite. You will lose a few hardness points. I can assure you you will be surprised at the results. You will need a salt bath for quenching. Look at a TTT (Time Temp Tranformation) diagram for 4340 and figure out the temps. If you have a problem I can help.

4340 and 4350 are my favourite steels for thru' hardening and Nitriding, and 4320 or 17CrNiMo6/8 for Carburising

Consider changing the setup to look more like a car or truck wheel, with the load centered over the bearing. You might even be able to make the whole thing out of a truck rear axle assembly. You get gear reduction, brakes, and cheap replacement parts out of this plan. I used to make marine winches from truck rear ends. Strong and cheap. If you can get them back, you could even use truck wheels for the wire spools.......

Why not change the center of the spool to a cone shape.

As is most all large magnetic wire rolled,,

Phelps Dodge for shure I have an empty spool,,

filled,this spool would be lighter on one end.

hi,

i couldnt understand what u r trying to say. and what is spool?

The Spool is what you wind the wire on,,

The center is cone shaped.

The center of gravity will be closer to

the load bearing end.

I

I

Dear Sir !!

Nothing wrong in existing design!!

You could improve the shaft quality,by using Spring steel quality (like;51CrMoV4)ground surface(h7)+Heat treated to;444-486 HB - with shot peening performed on its surface (to prevent dislocation cracks movement).Any big size spring steel producer could perform it (hopefully the price will be less time and money consuming)Best Regards: JKL

NB - You should check any turning shaft with "Black Light"-on the spot....and find cracks devlopment immediately

Why not design a shaft that protrudes enough past the edge of the spool to allow for a temporary or movable support bearing. It could be just a 180 degree cradle made from a bearing material that can handle the conditions and load.. The overhang and winding stress is causing the issues. I would try to control the flexing if possible.

Do you have a drawing or better description? 1100 pounds on the end of a 2.25 shaft seems a bit much.

Hello Castkarthick,

Without seeing the actual setup it is hard to make recommendations, but my initial reaction is to go after unloading the shaft by supporting the drum with rollers or tires (perhaps spring loaded).

Jeff

Make the spool such that the axel or starting place for wire to spool,,these (example only)Width in inches of axle as defined above

1.On end having two pillow bearings 6 inches

Width of axle on opposite end of spool 1 inches

comprende amigo,?,,,Turn a styraform cup on its side attach..round plates on each end,,,hole thru the middle...coffee cup = axle

If all else fails go to a wire winding shop (copper) and ask to see an empty spool.

Those figures need turning around,,if i wasnt in a hurry i would paint you a picture,,,it is the wire at the END of your spool that is breaking the rod,

put less wire on that end.Put as much weight as possible NEXT TO the pillow bearings,,this will involve spool change $$$$$ but they sell the spools in hard plastic,Good Luck.

i wiould try a torsion steel axle with self-alghning bearings. the the drum support wheels that has already been mentioned seems to be an excellent idea.