Steel Shaft Design Safety Factors

[coss]

Would like advice on reasonable value for Factor of Safety on a 15' long Steel SAE 1040 Fan Shaft on a 200,000 CFM Fan with a 250 hp Motor driven by a VFD. Fan Sheave is 1,000 lb. ,10 V-belt , 40" Diameter, and the Wheel is 7,000 lb. , 89" Diameter. Shaft is 35 years old and was undercut from 4 3/16" to 3 5/8" along an 8" length and two semi-circular 8" Steel Sleeves were added and welded to the undercut part of the shaft which is inside the bearing. Assuming this is a temporary fix, what factor of safety should be required in this situation and what effect will the welded sleeves have on the stress distribution in the shaft?
Coss

 

[chicopee]

When do you need an answer?

 

[chicopee]

Magnitude of Start up torque, distances between bearing supports with fan and sheave relative placements, loads from rotating belts are variables needed to be known, however a factor of safety between three and five would seem to be reasonable. Was the shaft damaged by bearing failure necessitating steel sleeves?

 

[strokersix]

You might consider another method of attaching the sleeves to avoid weld metallurgical issues. Perhaps the sleeves could be clamped in place? It sounds like this repair is already in place so it's probably too late but if you have to do another I would consider mechanical retention of the sleeves as an alternative to welding.

Mike

 

[coss]

Chicopee
Yes the shaft was damaged by bearing failure. The sheave is overhung by about 16" and the fan wheel is about 60" from the drive bearing. Loads from rotating belts and sheave weight give a resultant force on sheave of about 2,250 lb.
The factor of safety you suggested of at least 3 sounds reasonable to me since the shaft is old, was damaged, and was undercut. Do you think as a rule of thumb that when this type of shaft is originally designed the factor of safety would be at least 2 or 2.5? Thanks for the advice!

Strokersix
What does Mechanical Retention involve?
Thanks a lot.

Coss

 

[beloka]

It all sounds like a very temporary fix. Welded shafts, particularly when the weld or heat affected zone sees any cyclic stress, can be a receipe for disaster. British Standard BS7608 titled "Fatigue design and assessment of steel structures" and possibly BS7910 titled "Guide on methods for assessing the acceptability of flaws in metallic structures" may be of help you. Fillet or incomplete welds subject to cyclic loading have an inbuilt "crack" just waiting to propagete. Imperfect welding is another consideration. I suggest you recalculate the shaft design making due allowance for the stress raisers and diameter reductions made. There is a DRAFT Australian Standard revision of AS1403 based on the German DIN743 Standard. This standard contains dimensional surface and shape factors needed for the calculation of component fatigue strength but does not cater for welded parts. It does have means of calculation for actual and minimum safety factors.
Failure and likely consequential damage, cost ,etc, of a shaft, such as the one described, should be considered. I would be very sceptical regarding ssumptions re the history of the equipment and possible factor of safety used in 35 year old calculations.

 

[strokersix]

It's tough to say exactly what might work without seeing it myself but how about this:

Turn the shaft down in the area of the repair leaving a generous fillet and/or taper at the transition. Polish the transition to reduce stress risers. Then machine the two semi-cylindrical sleeves as before. Instead of welding them in place, use a shaft collar on each end to clamp the repair sleeves in place.

I offer the above only as a suggestion, not a recommendation. Please use caution regarding the risk of shaft failure and bodily harm. A rotating shaft with a cantilever load or deflection is a standard test for material fatigue!

Mike

 

[chicopee]

researching my archives on shaft design as well as two of my references MF Spotts "Design of machine elememts" and Kent "Design & Production", factors of safety range from 1.5 to 2.0 and for shafts with Keyways up to 2.5. Therefore your estimate of 2.5 is reasonable. You may want to do the calculations and MF Spotts has a great section on shaft design.
Now, since the shaft has been repaired you may want to inspect initially by the dye penetrant method the welded connections between shaft and the steel sleeves, the keysways reentrant corners and at transitions into the shaft and any portion of shaft where you have determined high stresses.


s

 

[macmil]

I think you answered your oown question when you wrote" Assuming this is a temporary fix" !
From the description of the fan, it probably is fairly important to the operation. The probable mode of failure will be breakage at the repaired spot. The down side is whether the runaway sheave will take out the motor at that time, or worse, anyone in the vicinity. For my part, I would ensure there was material for a new shaft close by and replace at an early date on a scheduled basis.. We are only talking about 900 lbs or so of steel. Consequential damage could be much more. After 35 years the old shaft does not owe you anything, and it is probably good enough to be reused for smaller shafts, noy using the repaired area.!

 

[coss]

beloka, strokersix, and macmil
I must say you all give great advice. The decision has been been made to replace the five damaged shafts with new ones as the method I first described is considered ill-advised. It seems that the two welds 180 degrees apart introduce a new alternating stress which would result in (macmil's) "breakage at the repaired spot" and (beloka's) "recipe for disaster" and (strokersix's) "use of caution regarding the risk of shaft failure and bodily harm". Every one of you is right on the money and I am quite impressed with the excellent judgement and wisdom. It seems that the 2 to 3 Factor of Safety is good for Solid Shaft Design and does not necessarily apply to Fatigue Analysis for Failure. I must now start a new thread called "Fan Shaft Braking Device" because I am in the middle of the design of Differential Holdback Band Brakes on 48 Fans in which three high-strength bolts will be used to attach a steel plate to the non-drive side of the shafts (now 5 of 48 will be new) and the question will be about shear stress on the three bolts as well as in the shaft, also heat from friction in the brake band. I don't think I'll feel comfortable until you Mechanical Guru's have spoken! I'll begin the new thread on March 15th in Mechanical "Other Topics".
Respectfully, Coss