Online Community of Zenith Builders and Flyers
Big tires are fun. You can go almost as many places as float flying. Flashback to my original post about going from floats to Alaska Bush Wheels ( https://zenith.aero/forum/topics/801-alaska-bush-wheels?xg_source=a... ). I did a fair bit of work to make the swap. Most of which involved reimagining how to make a bush wheel nose gear work while keeping my constant speed prop (and therefore the governor that sits right above the front gear post). I came out with a pretty functional solution. My son (19 with PPL) has been having a great time flying it, and so have I. The aircraft has been busy, with lots of landings, so that has given me the chance to see how my solution held up.
There were issues. After many many landings, the nose fork developed an offset (as compared to the front post). I am proud to say that my welds held tight, and the nose fork and post remained one unit. But it had to be fixed. I pulled the gear, and separated the fork and post (yes that took a bit of grinding). I adjusted them back to true with a laser level, and clamped them especially tight for welding. I again welded the cr@*& out of them, and added some side bracing support (to prevent future offset). Then off they went for heat treatment, and upon return got painted and reinstalled.
The bracing was not too pretty, I admit, but I was more concerned that it would hold well. It did that…. for a while. It stayed true longer than the first attempt, but eventually began to lean again. So it was clear that a different approach was required.
Looking back on it, I think that the deficiency was essentially only in side load. The 0.25” 4130 was plenty strong in straight compression, but when side loaded, the pieces were just too thin to avoid flexing. And once flexed and a little bit off, that made every landing happen in an increasing ratio of side load to vertical compression. Once the gear deviated from true, it was a downhill slide.
My new approach needed to avoid this problem. That meant a thicker nose fork, and that meant a different material - 4130 would be just too heavy. I decided on 7075 Aluminum, which seemed like a good fit.
I was committed to a design that I could make myself, which gave me some limitations. But fortunately I have been developing my CAD and CNC abilities this year. I’m to the point where I can pretty easily design things that I can’t possibly make. So the main challenge was to keep the parts inside the work envelope of my mill, and to keep the design simple enough that it fell within my shop abilities.
Well there was no way a full nose fork would fit in the work envelope, so right away I knew it would have to be a two piece fork. And even then, each side would have to lay out on the mill table at an angle in order to fit. This was going to be tricky.
I designed the gear in SolidWorks. I made it such that each half of the fork was a mirror of the other (with a few small exceptions) and the front and back of each piece was also a mirror image. That allowed for any jigs I made to be used on both pieces, and avoid the need for extra jig making. I used finite element analysis to predict its performance under load. This time I predicted more side load in my analysis.
Once I had a two part design that I liked, I placed the order for two large 7075 billets of aluminum stock, then sat back and waited for them to arrive. As I did, I asked myself - could I really make two of these gear halves without ruining one? Should I have ordered a third billet of 7075? The large blocks of 7075 were not cheap. There was going to be a lot of operations….. and only one chance to get it right.
When the billets arrived, I loaded one onto the mill table and got it laid out at the necessary angle. Then I double, triple, and quadruple checked my work, clamped it down tight and began cutting the first half of the first piece. It ended up working out just fine. But the challenge level was about to increase. Now I had to flip that piece over, such that the shaped and rounded portion of the gear leg was down against the table. No question but that this would require a jig. So I machined out a negative (0.010 inches oversize ) of the design, and also added locating features onto the jig to establish the correct angle and the correct zero point to cut the pattern just exactly where it was needed - in alignment with the previously cut half. I was a little bit surprised when this actually worked out just right (well….close enough that a little sanding would bring the two halves together smoothly).
With the first half down, I loaded up the second half and repeated the process. It also came out to my satisfaction (a little bit to my surprise.) I now had two halves of my nose fork. But they did not yet have all the necessary features. Additional machining was needed to create the dove tail joint, the slotted top, and the axle holes. These would seem like the easier parts, but the difficulty is getting them aligned correctly and firmly held in place. This would again require some jig making. I make two additional jigs. These jigs allowed for the slotting of the dove tails and the top joint, as well as machining out the axle holes.
Next I made the top plate. This piece is machined out of steel and provides much of the strength for preventing side load deflection. It consists of a top plate with two thick fingers that extend down into the nose fork halves. There is a 3 inch hole in this that goes down to the top of the fingers. The Front gear post slides into this and is welded from above and below. The lower mount of the front shock extends down the tube, and in between the fingers and nose pieces where it is mated to the center mounting bolt - a solid AN-8.
One of the greatest joys of this project was finally placing the nose fork halves into the top plate, along with the shock and seeing all the holes line up and allow passage of the three bolts. It was a tremendous relief to feel the bolts pass through all the holes in perfect alignment. CNC is indeed magical in the precision it can achieve.
There were a couple of other things that nagged at me from the first design, and I took this as a chance to correct them. To begin, I wanted to increase the resistance of the the shock absorbers. The first shock tended to drop the nose down on heavy braking, and then struggle to push the nose all the way back up once stopped. I decided to upsize the shock. This created some challenge, as the larger shock has a larger diameter and would not fit in the same size post. I ended up increasing the front post diameter from 2.5” to 3” to allow for the bigger cylinder. I also improved the shock mounting by using a bolt which joined the fork halves to capture the bottom of the shock, and lining up the top bolt with the steering arms, which I upsized so that I could remove the top shock bolt through the steering arm if it was ever necessary.
Owing to the larger post, I also changed the lower bushing mount. I changed this such that a bushing sleeve is captured by a front cap (secured by four AN-5 bolts). When the front cap is removed, the front gear can be removed from the airplane more easily, and if necessary I can remove the shock and nose fork separately. This makes the whole gear much more serviceable.
Using this new lower bushing mount, I did away with the “V-block” all together. On the first gear, the V-block was too aggressive, and led to very heavy rudder pedals. I could have changed out to a less aggressive V-block, but I thought a more adjustable option would be an improvement. Changing shock forces will change the amount of V-block angle needed, so I wanted something that was easier to change. This time I machined the top of my bushing mount flat, then made a set of relatively thin pieces that function just as the V-block did to push the gear up when rotated. But unlike the V-block they are secured by countersunk screws into holes tapped into the bushing mount. These can be changed out without much difficulty to increase or decrease the aggressiveness of the angle and the rudder forces needed to overcome it.
The method of securing the upper inner post was also changed. The previous method worked, but the amount of movement at the top of the post was more than I liked, especially so close to the governor. Not to mention that it was difficult to install and remove because of the limited space. So I decided to make the top post entirely fixed, and allow all the motion to occur at the interaction of the inner and outer posts - where the shock is. So in the new arrangement the top post is welded to 4130 flanges which bolt to the top and sides of the gear channel. There is a carrier (like a small piston) that interfaces the top post with the lower post and the shock. This carrier has a thrust bearing that allows the lower post to rotate freely.
Taxi and first flight testing were completed successfully. (Video of first landing here: http://youtu.be/G3eoQUp71Gc ) Overall I have to say that I am much more satisfied with the Version 2.0 design. It feels better on the rudder pedals, and there is no movement at the top post near the governor. I believe that the fork will hold up better over time. Of course only time will tell if it will meet expectations. But I will update here as time goes on, in case anyone else considers a bush wheel on the nose. Just remember that my post is just to share my experience. I’m not an engineer. So do your own homework before making any changes to your airplane please.
*footnote - many of the images above are improperly rotate from the upload process. The rotate image feature is not working. I will correct when able. For now, view as if you were in a steep turn.