What's the number for Locktite's bearing glue? Or, since there's probably a bazillion of them, what's the right one for a steel shaft in an aluminum housing that persists in pulling out?
I assume that you have to assemble with the glue on the shaft, and can't just dribble glue on a pre-made assembly -- correct?
(Note that I think that the correct glue in this application is a shaft that's mechanically constrained from pulling out -- axially loading a press fit doesn't seem wise to me. But I'm not the one asking.)
609 "Retaining Compound" is the most common. Look at Loctite's literature to see if a primer will improve the strength.
290 is a wicking threadlocker meant to be applied after assembly. I've used it with good results to prevent shafts slipping in the inner ring in lightly loaded bearings. I'm not sure how well it would work to carry an axial load. It's about as viscous as alcohol, so is best applied with a syringe or needle oiler if you need to avoid getting it all over the place.
For the highest strength on a press fit, you probably want loctite 648, but they also make other types for looser fits and types that are tolerant of oil contamination. Their web site has a good summary:
formatting link
You normally have to apply these before assembly. However, if there is the slightest thin gap , I have had success using their 7452 accelerator (probably other primers would work too) and super glue. The acccelerator has very low surface tension and gets drawn into the gap. The accelerator then draws in the glue. By the way, this is a good technique to bond broken china. Assemble the pieces first and then apply the accelerator and glue.
I have never tried this technique using their cylindrical retaining compounds (like 648), but if you experiment with it you might want to try the 7471 primer made for those.
Yes. They make a threadlocking compound intended for application "after" but it's not a preferred choice even for threadlocking, and bearing locking is a more demanding job.
Not to mention steel in aluminum - thermal stress / differential expansion will kill most anything, so try to keep it at the same temperature all the time...or just spray it with salt water and wait a couple of weeks for the corrosion to lock it on good.
603, 638, 648, 660... or go pick yourself, or better yet, have your friends pick up the phone and ask them...they have application engineers.
formatting link
Another option would be to actually cut the shaft down, or the hole bigger, and use something non-rigid to bridge the gap and take the differential expansion stresses. The one (I'm sure there are countless other options with varying properties) I happen to have experience with would be Armstrong A-12, which has properties that vary with mix, but in the 1:1 mix gives 5000 psi and 78 D hardness - whether that would hold up to the loads your friends need to handle or not, I don't know. It certainly sounds like they are band-aiding a bad design, so some redesign might be in order...
_I_ think they're band-aiding a bad design myself. It's a premium electric airplane model motor, so by design you're going to (a) hang a propeller on the end of the thing that can lift the aircraft vertically, (b) get things hot and cold and hot and cold, (c) vibrate the hell out of everything, (d) induce huge gyroscopic forces when you do tight turns -- the list goes on and on. I think a shaft with a groove and a circlip, or a shaft with a collar and setscrew (like just about everyone else in the industry sells) would work a lot better.
Here's a link to their site that shows a cut-away of the motor:
formatting link
But I had suggested Locktite half-flippantly, and now I've been asked to clarify...
So, the shaft is pulling out of the aluminum housing at the far end (and through all the bearings...?) Had to sort out this "outrunner" concept to make sense of this - but rather than taking the outrunner to the logical conclusion (bolt prop to motor casing that turns and stick that out front) they are still using a shaft headed the other direction.
Grind threads in the end of the shaft and thread the end housing (labelled shield) might be one way, so long as the motor is only turning one way. Extending the shaft through and circlipping would certainly appear to be simple enough and effective for either direction. Either would be far better than any glue for something that's thermal-cycled as much as this applications appears to do when using dissimilar materials.
... or not enough time and tools. I've done that on a 2" shaft for my backhoe just to get it back into service. If the lathe had been up, I'd have used it, but it wasn't, so...
That's exactly what some manufacturers do. But a lot of people like to have the shaft extend out of the stator part of the motor, and have a fuselage design that's got a firewall in front, with the stator bolted to that, and the prop just on the other side of the firewall.
The picture that I saw was of the motor in question, with the prop, spinner, housing, and shaft lying on the ground in front of the plane
I'm not sure you could maintain runout if the shaft were threaded in -- could you, without using insanely precise threads on both pieces?
The first picture at link below shows an outrunner motor in an obvious application on the nose of a tiny airplane, but it isn't clear how the plettenberg motor at above link would work in such an application. The picture isn't clear enough to show where the wires go in, what the "fixing plate" attaches to, etc.
That's a different sort of brushless motor, where the stator is attached to a tube which is in turn attached to the plane.
Current practice puts some threaded holes in the stator, and the shaft either sticks out through the stator or off the end of the bell. The stator is bolted to a bulkhead (everyone calls it the "firewall"), either to the front with the prop shaft way out on the end of the bell, or to the back of the bulkhead with the prop shaft sticking through.
Thread together with no great concern for precision, mount in lathe, turn the end-bell true (in proper direction to keep threads tight, of course)? I freely admit that the circlip is the simpler solution.
If you were really into micromachining and found customers really into old-school modeling, making the exterior of the rotor into a good facsimile of a rotary piston engine could work well for the "bolt prop to rotor" versions...
I use Vibratite 541 which is equivalent to Loctite 660 and cheaper. I like it a lot. I like that unlike regular glues the mode of failure is not catastrophic: The parts move if force is applied in excess but the same force has to be applied to change position again.
Note that these retaining compounds are less effective on aluminum than on steel (Loctite have nice charts in their technical documents).
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.