I finally decided that instead of sitting here watching people unsuccesfully trying to solve problems with one-way bearings I should stop procrastinating and write a post with drawings and pictures to share with other people what I have learned to make sure the one-way bearings work correctly in my helis.
As illustrative examples I take a look at three examples. The Freya auto hub, the Raptor auto-hub and the infamous KSJ starter wand with the one-way bearing.
This explanation applies to most helis and devices such as starter wands that use one-way bearings. I have had one-way bearing issues with several of my helis in one way or another other helis and there are several posts in here where people keep having problems with their one-way bearings in almost every brand of heli under the sun. In most cases is not as much a problem with their units as it is more of a consequence of a lack of understanding of how they work. Hopefully this post will bring some better understanding of how the one-way bearings work on our helis and it will help us assemble and adjust them correctly.
1) The Freya one-way:
Initially, during the first couple of weeks I had my X-Spec, I blew a few one-way bearings (auto hub one way clutch bearings) making them slip after a few real hard flights. To prevent being grounded by this problem in the middle of a weekend I bought a few of them to keep them in my toolbox just in case.
The problem was that the bearing case developed a micro-fracture (not easy to see to the plain eye) and under torque load the case started to expand so the rollers wouldn't be able to lock against the center shaft sleeve with enough force hence the slipping.
Concerned with why this was happening, I started thinking about it and then I realized what the “why” was. As a matter of fact once I understood it, it became pretty simple to fix.
From past experiences with metal parts I knew micro fractures are usually some problem with metal fatigue induced by repetitive loads in the section that kept developing the micro-fracture.
Looking at the geometry of the one-way bearing I realized that being securely fixed it wouldn't work the way it needed to for this application. A one-way bearing is a self-centering and self aligning device that without the self centering feature won't work correctly. I'm of course refering to the shaft in the middle being able to find its center when rotated inside a one-way bearing. That's what I was used to look for but in this case I had to look at it from a different perspective.
The problem that initally bafled me in this case was that according to what I was looking, the shaft wouldn't be able to find its center because the shaft (or sleeve) is held centered in place respective to the hub by the two ball bearings on top and below it.
Initially I thought that to make it a bit lose as I needed I would need to find a way to introduce some play such as replace the ball bearings with something less precise such as brass or Teflon bushings. I also though of other more involved and complicated solutions but then I realized that in this case the solution was to think of it in a different way and instead of expecting to see the center shaft find its center I should expect to see the outer casing find its center.
This is the reason why:
When you look at the way a one-way bearing works you realize that it's nothing more than a bunch of roller pins that when rotated “climb” the bearing case “walls” pushing the center shaft away from the walls. This “climbing” and consequently the pressure they exert on the center shaft is the reason why they lock. When rotated in the opposite way they “climb-down” releasing the center shaft from the grip.
Now the trick to understand why a secured one-way bearing where neither the outer case or the center shaft are allowed to move won't function or will develop problems is to realize that at the very begining and for a very short period of time until the one-way bearing locks all the roller pins are free and ONLY ONE roller starts the effect by being in contact with both the shaft and the case wall.
When rotated this roller starts “climbing” its own case wall and by doing so it “pushes” the shaft towards the center but you can also see it as the roller pushing the wall or climbing plane away from the shaft. This pushing away makes the whole outer case move away from that first roller pin point of contact and at the same time it brings the opposite side of the outer case closer to the shaft until the opposite roller starts the contact between its wall and its side of the center shaft.
Keep in mind that in the above drawing the outer case off center offset is a greatly exagerated to illustrate the point. In real life the case is off-center only a few thousands of an inch. Also the “climbing planes” are not that steep. They only need to bit a couple of degrees in inclination for them to work. As a matter of fact the lower the inclination the greater the pressure the exert on the center shaft. Also in many cases there's not even a need for a climb, the geometry of an hexagon shaped outer case gives enough climbing effect for it to work so bear with me and look at the drawings as more illustrative than engineeringly accurate.
This produces a sort of “embracing” effect among all the roller pins that makes all the pins climb their respective “walls” and start pushing against the center shaft until there's no more climbing possible at which point the bearing locks into its one-way effect.
Now to understand the problem I was having you need to imagine that either the outer case is not allowed to move freely or it is severely restricted from movement. The result is that the particular roller whose side of the bearing case (wall) is closer to the center shaft will always be the first to be in contact but won’t be able to push the outer case away making it be under a lot more pressure than the other ones. If you repeat this over and over it will subject that single wall spot to excessive and repetitive flexing and consequently it will develop a fatigue crack. This problem is exacerbated when using a metal hub because they're more rigid than the plastic ones. In both cases, whether it is metal or plastic, a rigidly or even semi-rigidly mounted one-way bearing will ultimately develop a crack when flown real hard over and over.
I imagine this is why the design engineers put some play into the fitting of the bearing and I kept misunderstanding it as a lose fit. Then we the american lunatic pilots
started playing pogo stick with our helis bouncing them hard on power pitch takeoffs, piro hovers, piro tic-tocs or any of that crazy pitch pumping we do, so some more freedom of movement of the bearing case is required especially for 91 sized motors and their new higher power levels.
To fix the problem I just had to find a way to let the one-way bearing case move more freely so it could center itself when engaged distributing the load among all roller pins equally and lock in it's ideal center every time. Doing so would prevent one of the rollers from getting the entire load over and over. The solution was amazingly simple.
Those extra loads force the issue of making sure that extra play works under even the more stressful conditions. To do so just pack a lot of grease (I used white tri-flow grease) around the outside of the bearing case so that you guarantee it's lubricated ebough so that it can move freely when it needs to find its center.
The one-way bearing below should be completely surrounded with grease on the outside to allow it to "float" when installed inside the hub.
This problem happened mostly when I was flying it real hard. I don’t have direct experience with the 60 versions but I don't believe that a 60 size engine would have enough torque to break them as often. The 91 engines have considerable more power and consequently we fly them harder and harder subjecting parts to a lot more abuse than before.
This is something I have done and it worked for me. I went from blowing them frequently to never ever again. I also told some friends in other states about this mod with excellent results and a few weeks ago I also told it to one of the guys that I fly with and so far it also seems to have fixed the problem for him. Lucky enough for him I had one of those extra bearings I had bought for myself that I had'nt had any use for again
. I have also found that this assembly tip works both for the metal and the plastic hub.
Now for the perfectionists even though this is not required for it to work you can also use some 400 grit sanding paper to sand the sharp edges of the “fingers” that holds the one-way bearing case in place to make it even easier for the grease to let it move freely.
2) The Raptor one-way bearing:
As an example of a different approach the Raptor also uses a one-way bearing with a sleeve centered by two brass bushings. In the case of this bearing the brass bushings usually have a bit of play already so the lock should be solid from the begining. If it's not a little bit of wear in the bushings gives a bit more play making it more solid.
3) The infamous KSJ starter wand:
Most people that bought the KSJ starter wand have found that after only a few uses the wand stops working and the one-way bearing fails to lock.
In the case of the KSJ wand you have a shaft being held in the middle by two bearings just like the Freya metal hub does but unlike the Freya hub the bearing itself is not allowed any play for self-centering. What happens is that once there is a bit of wear in the shaft the bearings stop working because the roller pins fail to start the centering effect since neither the shaft nor the casing are allowed to move.
Using the concepts explained above and unless it's really worn-out, you can get it back to work. The idea is to allow for some play so that the shaft can move a little bit inside the one-way bearing making it engage into a lock.
The way to do it is to take one of the bearings off and allow the wand to rest on only one of the ball bearings and on the one-way bearing.
Here you can see the back ball bearing and the shaft that goes inside the one-way bearing.
Here's the front bearing. Between the two of them the shaft is securely held in the center of the one-way bearing not allowing it to do the required self-centering to engage into a lock.
Here's the one-way bearing. As it can be seen the one-way bearing is also firmly held in place so unlike the Freya hub no play for the one-way bearing is allowed.
Using a torch to heat the unit it can be easily dissasembled.
The front unit it taken out and now there's enough play in the shaft fo it to move inside the one-way bearing and find it's center. Re assemble the unit and remember that since now there's nothing to prevent the wand to be pushed inwards you need to supply some spacer such as a few washer in front ot the starter motor shaft the so it will hold the shaft in place by pushing it from behind. After this modification the unit now is back in service.
There are several schools of thought as to what system is better. Some prefer ball bearing supported sleeves because they feel that under load the main gear plane has no possibility to lose it's perpendicularity with respect to the main shaft creating misalignment in the gear mesh with the driving pinion. Others preffer the simplicity of the brass bushings. I believe that as long as there's play both systems are good. The reason for my belief is based on the observation that one-way bearings are not only radially self centering as explained above but also axially self-aligning devices. In other words a Hub with only a roller pin based one-way bearing and without either bushings or bearing will not only self-center but also self-align with the center shaft because when engaged all the roller pins being straight cylinders will force the hub to align parallel with the center shaft. Being this the case the wear on the bushings will be limited to the initial wear and not continue to increase afterwards.
This is an extremely over-simplificated explanation. The actual operation of a one-way bearing from a strict engineering point of view is a lot more complicated but the principles explained here are the same. There are several factors such as surface roughness, brinell hardness coeficients, torsional shear and all kind of details that determine additional factors. For instance when I say there's only one pin in contact I don't mean it literaly what I mean is that there's only one that has significant force being applied to it while others have negligible force but they might still be in contact. If I start making all kind of side note explanations the post loses its purpose of being clear to most people. In most cases the pre-loaded pins should engage correctly but as the pre-load wears and becomes less solid the described play becomes more important. There are other issues to consider such as the fact that in most cases there's a pre-load in the roller pins to start the contact, there are also cases where the pins are not circular cilinders but pre-loaded pivoting lips with a progressive curve shape that gives them a transition from fast engagement to high force but in all cases the same principles explained here apply so the post still works as a guideline to understand and rectify problems commonly found.
When faced with a one-way bearing issue understanding the way a one-way bearing works will allow you to make the right diagnostic and it will also give you the knowledge required to devise a solution or a corrrection. Most issues with one-way bearing are easy to resolve. Just think how the concepts explained above apply to your particular problem and you will probably be able to figure out a simple solution.
Avant Aurora Ultimate