Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Recommended interference fit for 3000 RPM Generator couplings 5

Status
Not open for further replies.
edison123

edison123

Electrical
Oct 23, 2002
4,409
0
0
IN
I have a 3000 RPM, 18 MW generator with the shaft dia of 228 mm at the coupling seating area. The rigid coupling (with a flange and a hub) has four circumferential round key pins and four radial round key pins. All the eight key pins equally spaced around the circumference. What would be the recommended bore dia for this coupling?

I have seen interference values from 0.003 to 0.00075 inches per inch of shaft diameter being recommended for keyed couplings. Any tips from end users?

Muthu
 
Replies continue below

Recommended for you

Sort by date Sort by votes
Hi edison123

It might be worth contacting the coupling manufacturer and discuss the application with them.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein
 
Upvote 0 Downvote
If you use 0.003" per inch, you will likely need to cut the hub off to remove it. I would not go below 0.0005" per inch or above 0.001" per inch.

Johnny Pellin
 
Upvote 0 Downvote
I have found the site pretty helpful, I think most of the calculation formula's for the ISO fit come from a book "Precision Machine Design" I don't have a copy but someone I worked with for many years always had his copy around and used it frequently when dealing with larger rotating electric machines when factory info was lacking. For most of my career Machinery's Handbook has done me pretty well.

Regards, MikeL.
 
Upvote 0 Downvote
Mike - I was playing around with that equation in your first link. Changing the stress concentration factor from the default 2 to 1.5, the factor of safety against yielding of the hub changes drastically. At the stress concentration factor of 2, the factor of safety for hub yield never gets to 2 for even 0.1 mm interference on a dia of 228 mm. Should I reduce the stress concentration factor to 1.5?

Also, increasing the hub outside diameter, does not change this factor of safety for the hub yield. I thought a thicker hub will increase this factor.

Muthu
 
Upvote 0 Downvote
If increasing the hub OD does not change the hub's safety factor, you are calculating something incorrectly.

Increased hub OD without any change to hub ID, shaft size, or interference will result in higher hub safety factor and lower shaft safety factor- because the hub will be stiffer, thus experiencing less strain and imparting more strain into the shaft.

This behavior is obvious if you load the calculator that Mike linked- open the webpage, and don't change any values except hub OD and watch how the output values for safety factor change.

With regard to stress concentration factor- this value needs to be a 'real' value. If your goal is a hub design that will have long life and will perform well, you can't just guess.

There are several resources out there that will allow you to determine what the stress concentration factor should be.

I'd start here:

 
Upvote 0 Downvote
The general rule of thumb for interference is fine for shafts ranging from 2" to 5" is diameter. At 228mm you are pushing 9" in diameter. Most often when we encounter hubs of that size we custom fit them to the application. Since you mentioned the machine is very old, I have a sneaking suspicion that shaft is grossly over sized for the application. If it is a high vibratory torque application I would fit the hub for 3x the running torque rate, that may be a very little interference, like only 0.003" total, simply because of the size of the shaft. The key is only using an interference level required, so it makes installation and removal as easy as possible. If you provided the following information I can run a quick calculation for you. Application speed, Application H.P., Required service factor (1.5 smooth running, 3.0 high vibratory torque), Usable shaft length (or desired hub length), and Shaft Size & Tolerance (228mm +/- ? ? ? ?).



When it comes to couplings we are always here to help.
 
Upvote 0 Downvote
Thanks jgKRi. I tried only changing my parameters and didn't see any change on changing the hub OD. But, with the default values in that link, I can see the factors changing if I change only the hub OD. I will to go back and check what I am doing wrong with my acutul parameters.

Thanks CouplingGuru. This is a '80's Japanese machine design, so they could have overdesigned the shaft and the coupling, especially, when they put in four axial keys and four radial keys.

This is a 3000 RPM, 15000 KW generator running smoothly (no vibratory loads), but the load (hence torque) changes be quick and large. The coupling hub length is about 291 mm and the hub OD is about 370 mm in the straight portion and 660 mm in the flange portion. The flange width is about 90 mm. I have uploaded the coupling drawing in the link below. Look forward to your calc.



Muthu
www.edison.co.in
 
 http://files.engineering.com/getfile.aspx?folder=a81c946b-7ad8-45b9-8f72-6b7b73eb9cdc&file=Rotor_Coupling_Drawing.pdf
Upvote 0 Downvote
Under electrical fault conditions the peak torque values could be very large. Do you have any electrical data on this machine, reactances etc?
 
Upvote 0 Downvote
Running some quick numbers 15MW = approx 20,115 h.p. That is quite a bit, but at that speed you are "only" looking at 422,500 in-lbs of running torque. Running some basic shaft torsion equations, that means that shaft is only seeing about 3ksi in torsion shear. That is like a 10x safety factor. Based on that I would recommend a transitional fit and let the keys carry most of the load. That may be the reason why there is so many keys on that shaft, they may have wanted the ability to remove it easily so they could get away with out even having a true interference fit. So try to measure the shaft as accurately as possible, then have the hub made at a +0.000/+0.003 tolerance off of minimum measurement. That will probably give you a transitional fit at +0.0015/-0.0015. That will still require heat to install. If that is undesirable, I don't see a reason why a clearance fit won't work it almost looks like they designed it that way originally. But then you would be trading off the difficult heating installing procedure, with finely controlling run-out from the dowels and tapered pin keys.

When it comes to couplings we are always here to help.
 
Upvote 0 Downvote
Not as much information online as I'd expected, but GE's paper GER-3224 references violent torque swings of +3 p.u. to -3 p.u. occuring within a few cycles during reclosing events. A crash synchronisation with the machine out of step with the system would exceed this peak by some margin and, although machines aren't really designed to survive this type of event without sustaining some damage, a coupling fracture isn't an expected outcome.
 
Upvote 0 Downvote
Scotty,

Great points, it seems as if the shaft will easily handle those torque spikes. At 9" in diameter with (4) 37mm diameter keys even with a 1.5 million in-lb torque spike those keys would only see 6410 psi shear. Depending on material, they may be able to handle in excess of 30ksi. That is assuming the dowels provide no help at all.

When it comes to couplings we are always here to help.
 
Upvote 0 Downvote
Thank you, CouplingGuru. The old interference was 0.35 mm and it was a big hassle to remove the coupling. Both the shaft and the coupling got scored in the process despite quick high temperatures reached and the shaft was cooled to about 100 deg C all the time. I am planning now to keep the new interference under 0.23 mm and make sure the coupling is heated to achieve at least 0.5 mm clearance before fitting it back.



Muthu
 
Upvote 0 Downvote
edison123,

0.23mm is a lot of interference, That is more that 0.001" per inch shaft diameter. I would highly recommend cutting that value in half. And I would re-dress the shaft and measure it to verify you can get good uniform contact, especially if it was scored.

When it comes to couplings we are always here to help.
 
Upvote 0 Downvote
Hi CouplingGuru

Yes, the shaft has been turned down and grind finished to 228.50 uniformly from original 230.00 mm per pic below. Small blemishes are ignored since to remove them would mean more dia reduction.

Since the original interference was 0.35mm (for whatever reason) and it's been time tested, I do not want to reduce the interference too much and am going with 0.001" per inch shaft diameter thumb rule.

Thanks for all your valued tips. Much appreciated.

21bu7w7.jpg


Muthu
 
Upvote 0 Downvote
Nice picture ! ! !

Since you did a great job truing up the shaft, did you also take into account the tapered key pins relative location to the hub bore has changed because of the shaft reduction? Since those are tapered pins, may have to compensate for that in the hub.

When it comes to couplings we are always here to help.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

Rodmcm
  • Locked
  • Question
500 RPM 8000 kW hydro generator dimensions
Replies
16
Views
59
FacEngrPE
FacEngrPE
NickParker
  • Question
Diesel Generator in parallel with steam generator
Replies
9
Views
74
NickParker
NickParker
SAEEDGH
  • Locked
  • Question
diffrence between i and rpm 1
Replies
1
Views
22
waross
waross
zaboli
  • Locked
  • Question
Reverse Power Protection for Gas Turbine Generator
Replies
1
Views
12
crshears
crshears
Isaacwjp
  • Locked
  • Question
Rpm vs Max safe speed 1
Replies
6
Views
39
Gr8blu
Gr8blu
Back
Top