@HarryMann:
Known as angular-contact ball bearings
Thanks-- I could not remember the term used.
---------
@Bearfoil:
Lost train, still uncertain how "Splines" would obtain at the Disc? Also, I am thinking the Ball Bearings were chosen for this location due to the need for far greater contact area than Roller? Most of the mechanicals bear on this location?
Bear, roller bearings make line contact, while ball bearings make point contact, so the rollers will take a bigger load (in theory). The contact pressure flattens both the line and the point to a line of some width and a point of some diameter. This effect can be readily calculated, and the repetitive flattening eventually fatigues the surface, so the life is limited even though long.
Oil can actually bear this pressure, to a point which depends on its viscosity (which in turn depends on temperature). The small clearances act as a pump to retain the even greater pressures at the exact otherwise point of contact. The oil pressure (not the supply pressure, but the much greater pressure due to the rolling action) can widen the effective width of line contact, or area of point contact. The oil film must not become thinner than required to keep the asperities of the two metal surfaces apart, hence the need for a very fine finish.
Nice calculation goes pear-shaped in two or three areas: a) the oil film is obvious; b) fatigue life varies about four to one among the same bearing, recorded over several engines-- and in service you can't get past the point of shortest life, so you get no in-service data in quantity; and c) these bearings are really susceptable to mishandling in the field during disassembly and re-assembly, compared to say a smooth-bore bearing. Mishandling ruins the fine finish or brinnells the races, results in too much or too little clamping force; wrong or conterfeit oil...
The use of ball bearings as thrust bearings has puzzled me, given the front fan thrust. My working hypothesis at the moment is that there is room at the aft end of the LP shaft for a more adequate thrust bearing to hold the tension of the shaft. This would be a bearing more like a tapered-roller bearing-- this is usual for high thrust loads. Then this tension would be passed through the helical splines of the separable coupling by the torque of the shaft. The helical angle would be chosen to create the required tension, with some margin. Helical gears (or splines with gear tooth profiles, as one prefers) always create an end thrust along their shaft, you see.
To borrow a phrase from you, I don't actually know yet if this is so. I just think it is worth exploring. I admit to being a little concerned about a tapered-roller bearing turning at 11000 rpm. But there is a lot of radial room back there, together with a couple of lines on the drawing that could be interpreted as one of several (3?) very substantial radial fins supporting this area. There is a largish nook there where a curved-flaring cone appear to divert the hot blast into mixing with the cooler, slower bypass air. I have more thoughts, but I leave it there.
---------
@ Turbine D:
Bearfoil
I agree completely with you post this morning, I am impressed as well with this powerplant. As was suggested in a poster's PM I received last night, I did indeed "Google" to search for a patent I recalled, dealing with the advantages of helical spline shaft couplings and the advantages offered over conventional splines. You can see this patent...
It's commendable that you took the time and effort to look all this up. I haven't even read your reference yet, but I will.
---------
@ firstfloor:
What is still unclear is the method by which the spline joint is kept tight (my money still on torque). I have to say that I do not agree with Old Engineer about his geared coupling description.
I was rushing the post out the door and have no objection to standing corrected. "Geared coupling" is a term I was used to, and that would lead to a lot of pictures and reference material making the principle easy to follow by others. It may be an American term; not sure about "curvic" coupling usage here. But I could agree that curvic coupling is what we are looking for. I don't understand curvic coupling with helical splines on the end; in my view this would be combined into the shape of one set of teeth, but I just don't know and could well be wrong.
I too am working on the assumption that torque is holding the joint together. It reduces the part count. I'm also getting a glimmer that the axial (lengthwise of shaft) force that is generated is in part applied to hold in location the IPT that migrated aft in the event (have I got that part right?). That would also simplify assembly and disassembly, and again reduce part count. I just can't as yet see the boss or whatever (on the isometric engine cutaway) that would do this.
---------
@ mike-wsm:
Old Engineer - Oh wow, a waggly joint carrying how much torque at how many rpm?
Your comment has made me re-think. If only angular shaft displacement is allowed for, the joint collar need not depart from the engine axis (locally, as I assume that the engine axis bends under heavy load). In "geared couplings", one can mate a rigid half with a flexible half, requiring the shaft centerlines to intersect-- this intersection could be maintained by flexing of the long, slender shafts, perhaps. Perhaps this particular "curvic coupling" is built in that manner. Perhaps the reaction boss for the IPT can then be on the collar of the curvic coupling.
---------
I'm wondering if there is a plain-surfaced ball thrust bearing between the ends of the two shaft sections. This incurs more lubrication needs; but the compression force (generated by the screwing-together action of the helical splines) would only be that amount needed for any non-linear relation between torque input at the front fan, and the thrust output there. That would be plus a margin for uncertainty and for various excursions from normal operations, and the IPT load, but could be manageable. It's worth being a working hypothesis.
I'm seeing a lot of potential problems here that don't involve oil fire, given the pictures of damaged splines in RB211-form engines I've seen. I do know something about forces on, and wear and breakage of, gear-tooth section [read spline] elements, if nothing else about this situation.
Footnote: "Rigid coupling", in the King's English, appears to mean a coupling that is rigid in transmitting torque. This is in contrast to say an automotive clutch coupling, which is fitted with torque springs to cushion the application of torque. Anyway, in the official communiques, this regrettably (IMO) seems to have implied this coupling is so simple a device that the required safety margin in the face of spline wear and chipping could easily be determined. Maybe it's just my inference.
OE