# wheels and axles

Discussion in 'The light stuff area' started by cdlwingnut, Apr 18, 2017.

### Help Support HomeBuiltAirplanes Forum by donating:

1. Apr 18, 2017

### cdlwingnut

#### Well-Known MemberHBA Supporter

Joined:
May 25, 2015
Messages:
406
615
Location:
Iowa
the plans for the ultra-piet call for a 1 inch axle but finding wheels with a 1 inch hub is getting to be a bit of a challenge.
would two of these with a 3/4 inch axle work for a plane that comes in under 600 pounds fully loaded with a landing speed of around 25mph work?
I know side loads aren't great on high wheels, but a. i know how to do a crosswind landing, and b, i'm sure you don't want to subject the wooden gear legs to too much side load anyway.

2. Apr 18, 2017

### proppastie

#### Well-Known Member

Joined:
Feb 19, 2012
Messages:
3,505
1,006
Location:
NJ
two of what, .....what wall thickness of 1 inch or 3/4 inch. not enough information

3. Apr 18, 2017

### cdlwingnut

#### Well-Known MemberHBA Supporter

Joined:
May 25, 2015
Messages:
406
615
Location:
Iowa
4. Apr 18, 2017

### Winginit

#### Well-Known Member

Joined:
Jun 13, 2016
Messages:
811
242
Location:
x
duplicate post

Last edited: Apr 18, 2017
5. Apr 18, 2017

### Winginit

#### Well-Known Member

Joined:
Jun 13, 2016
Messages:
811
242
Location:
x
6. Apr 18, 2017

### akwrencher

#### Well-Known MemberHBA Supporter

Joined:
Oct 16, 2012
Messages:
1,064
359
Location:
Gustavus, AK
They even come with brakes

7. Apr 18, 2017

### proppastie

#### Well-Known Member

Joined:
Feb 19, 2012
Messages:
3,505
1,006
Location:
NJ
That is a fairly strong piece of steel. If you want 3/4 just as strong, need to do some calculations. MY/I is the general stress formula. Need to calculate I for each. Y is 3/8 and .5 for 3/4 and 1 Dia. M= Moment (length x force) (weight x arm) Y= distance from neutral axis to extreme fiber, I= Moment of inertia. Machinery Handbook has sample calculations.

8. Apr 18, 2017

### lr27

#### Well-Known Member

Joined:
Nov 3, 2007
Messages:
3,216
463
If you're going to use 3/4 inch axles, you may need to use stronger material and thicker walls, or even a solid shaft instead of a tube. Or at least that's what I get from the following:
http://jmtusa.com/calculators/section-modulus-calculators/

If the original design was quite a bit stronger than necessary, OTOH, you could get away with it. The 1 inch tube might be about 34 percent stronger than 3/4 inch solid rod, so that's how much stronger the material would need to be to equal it. It might also be possible to change some of the details of the original design to make up for it. For instance, if the original tube is just butt welded onto a plate or something, you could pass it through some kind of bushing instead, so there is no heat affected material at the point of maximum stress. The bushing could be welded to the plate. What is the material specified for the original axle?

The Texas Parasol plans I have show a 1/2" grade 9 bolt as the axle, with a 5/8" OD X 1/2" ID tube over it. The Texas Parasol wheels, or at least those in the pictures, appear to be considerably smaller, but this suggests 1 inch may be oversized. Skypup plans have 1-1/4" axles that are meant to have plastic BMX wheels slide over them.

If finding wheels with appropriate hubs is such a problem, I should think making your own hubs wouldn't be that hard. The holes to press the bearings in would have to be accurate, but the flanges and spoke hole locations shouldn't need to be that accurate. These hubs could be wider so as to tolerate more side force. They could be aluminum. Lacing up and truing wheels is pretty easy. My guess is you could make the hubs and lace up the wheels in a day. A relatively large drill press and some way to cut off tubing square would probably be enough to do the job. You could farm out the outer part of the hubs if you wanted to. 4 simple, identical parts shouldn't be too expensive. Make that two sets of 4 if you want to do it at home.

9. Apr 19, 2017

### vtul

#### Well-Known Member

Joined:
Mar 13, 2017
Messages:
366
91
Location:
Northeast, USA
Looks like bending stress and deflection will be greater with the same load on a 3/4" solid 4130 axle than 1" x .095 4130 tube.

10. Apr 19, 2017

### lr27

#### Well-Known Member

Joined:
Nov 3, 2007
Messages:
3,216
463
Yes. I see now that part of what I wrote was ambiguous. Even a solid shaft would need stronger material.

Just for fun, and because I'm badly in need of CAD practice, I drew up a very rough idea of what a homemade hub could look like. This one is meant for a 1 inch shaft. Materials are aluminum except for the oilite (or whatever) bearings and some roll pins or dowel pins. The holes for the dowel pins could be drilled through both pieces while assembled over a 1 inch shaft, which would reduce the required accuracy. If it turns out the bearing material is strong enough, the hole in the end plate could be enlarged so that it could press onto the aluminum tube also. Either shoulder type bearings or thrust bearings might be a good idea. As shown, the hub weighs about half a pound, but I'm sure this could be reduced. It's 4 1/4" wide to increase the angle of the spokes, but I think wider might be better.

11. Apr 19, 2017

### proppastie

#### Well-Known Member

Joined:
Feb 19, 2012
Messages:
3,505
1,006
Location:
NJ
That is a nice drawing, but I think it would be easier to remove the bearings on the wheel from Northern and use plain bearings (bronze bearings) in the existing hub. You might even be able to get needle/roller bearings that fit the hub and have 1" ID.

Takes a fair amount of time to figure all this stuff out...but it's a hobby.

12. Apr 19, 2017

### lr27

#### Well-Known Member

Joined:
Nov 3, 2007
Messages:
3,216
463
But this hub is wider than that wheel, and it's very easy to make it even wider. Are needle or roller bearings actually better in this application?

My intuition says that the disks on the end need to be a little thicker. Partly that depends on the spokes, of course.

13. Apr 19, 2017

### Winginit

Joined:
Jun 13, 2016
Messages:
811
242
Location:
x
You are right, but I figured he was looking for inexpensive and he could just toss the brake if he didn't want it. He can buy all the parts including a non-brake hub but it gets pretty expensive that way. Probably looking at $600-800 to buy individual components and assemble them. 14. Apr 19, 2017 ### proppastie ### proppastie #### Well-Known Member Joined: Feb 19, 2012 Messages: 3,505 Likes Received: 1,006 Location: NJ The issue is removing the existing bearing and getting some other bearing to fit into the existing$32 hub. There seems to be some resistance to plain (bronze) bearings so the next idea is a roller bearing with 1 ID fitting the 1 x.095 wall axle. There may not be one out there and yes if you used a needle bearing you would need to have hardened thrust washers each side. You also have to have a hardened ground shaft if the needles ride on the axle. This is not a trivial problem people pay engineers to do this stuff.

15. Apr 19, 2017

### TFF

#### Well-Known Member

Joined:
Apr 28, 2010
Messages:
11,128
3,012
Location:
Memphis, TN
Are the common wheels with larger than 1" axels? buy those wheels and add a bearing spacer.

16. Apr 20, 2017

### lr27

#### Well-Known Member

Joined:
Nov 3, 2007
Messages:
3,216
463
I thought that hub in the picture was chrome plated, not gold! ;-)

What kind of components are you talking about? The original hypothetical plane was just 600 lbs gross, wasn't it?

I may have estimated wrong, but the hub in my drawing was designed to be made with a drill press, a couple of reamers, and a little band saw or something for the outside edge of the disk. Out of what I guessed were standard aluminum sizes. If those sizes aren't standard, and I was serious, I could draw another.

Where do mopeds go to die?

I've seen what appeared to be standard bicycle components hold up to a 375 lb load. (ex pro football player) I was surprised he wasn't popping the spokes on the rear wheel. Remember that the rear wheel on a bike with a rear derailleur is dished over to one side.

17. Apr 20, 2017

### lr27

#### Well-Known Member

Joined:
Nov 3, 2007
Messages:
3,216
463
My question expressed my skepticism, actually. If the plain bearing inserts work well enough, why go for something "better". It would be fairly easy to test the plain bearings. Make a little cart, load it up, and pull it around with the car. Put a thermocouple in the axle. Don't go more than, say, half a mile or whatever distance you expect to taxi.

proppastie likes this.
18. Apr 20, 2017

### Turd Ferguson

#### Well-Known Member

Joined:
Mar 14, 2008
Messages:
4,555
1,553
Location:
Upper midwest in a house
I think you'll be disappointed with a "bushing" type bearing. They are not designed for the kind of speeds your airplane wheels will see. A simple ball bearing will cost very little, operate with very little friction, be able to withstand all the loads applied and last the life of the airplane.

19. Apr 20, 2017

### proppastie

#### Well-Known Member

Joined:
Feb 19, 2012
Messages:
3,505
1,006
Location:
NJ
Yes but the problem seems to be finding one for a 1" dia. axle off the shelf and reasonably priced. AT least that is what the OP said. Now there are lots of truly aircraft wheels 500x5, 600x6 for example which is also an option, and one would not be worried about if they are strong enough.

20. Apr 20, 2017

### lr27

#### Well-Known Member

Joined:
Nov 3, 2007
Messages:
3,216
463
I seem to recall there's some successful history with plain bearings on ultralight wheels. A quick look at McMaster Carr shows an oil impregnated bearing, 6391K423, rated at 2,000 lbs at 90 rpm. It's 1 inch ID and 1 inch long. Someone on one of these threads stated that the ratings are based on heating from friction. If that's true, the heat may be proportional to the product of the load rating and the rpm. The number we get is 180k. In the case of a 20 inch wheel at a taxiing speed of 10 mph, with a 300 lb load, we get 51k. So that doesn't sound too bad to me. I'm assuming that there is enough side loading at the bottom of the wheel to put the full load on one bearing, but there are two, so that may be a pessimistic assumption.

I was poking around the web and found this: