I’ve been doing a little work on my 801, and I figured it was time to write up what I have done.  

As some of you know, I put my 801 on 2500A amphibs last year, and have been having a fun time flying it around on the water.  I flew it from Knoxville to Mexico MO, and Oshkosh last year on the amphibs.  Well, amphibs are fun, but the performance is such that they will not be full time gear for my 801.  So I decided I wanted to be able to swap to wheels and back.  But I really didn’t want to go back to the 800-6’s that I started with.  I had my eye on Alaska Bush Wheels. There were some things to think about before going from 800’s to ABWs.  How big?  and probably the biggest question is what to do about the nose gear?

So this post is the sharing of my journey from floats to ABWs in my 801.  My purpose in sharing my experiences with modification to my 801 is just to let others know what i have done.  I’m not an engineer.  Don’t copy this.  Do your own homework before modifying your own airplane if you choose to.  

Coming up with a solution….. 

Why redo the nose gear at all?  Why not keep the original?  well…. with big tires on the mains, and an 800 on the nose its going to slope nose down quite a bit on the ground roll.  The nose down slope would reduce prop clearance to less than it had with 800’s on the mains.  Also, what good are big mains if you have a ‘small’ nose tire?  The nose tire would limit the ability to use the advantages of the big mains.  One could make a long nose gear to level the plane, but that would not fix the second problem.  It would not make sense to have main tires that could land on more difficult surfaces if your nose gear could not do so also.

I chose 29” ABWs for the mains, and a 26’ ABW for the nose.  My logic (sound or not) is that the mains will absorb more of the touch down, and I will be going slower when I get the nose down, so the 26 should be enough to give me the full advantage of the 29”s.  The 26” also presents less weight and surface area (in both tire and mounting structure) in front of CG, which would have less ill effects on the flying characteristics.

So my first step was to design a mounting solution for a 26” ABW on the nose gear.  I also needed to eliminate the bungee (see reasoning below) I came up with a design for telescoping tubes of 4130.  An inner-upper 2.25” x 0.120” tube was made free to slide in an outer-lower tube of 2.5”x.120”.  These tubes overlapped sufficiently for strength.  Within these tubes mount holes were made (and reinforced) to allow mounting of a shock absorber.  The shock is a Shock Monster for experimental aircraft by TK1 racing.  I went through the weight and compression requirements with Tony at TK1 racing and he set up a shock to manage what I needed.  I designed the fork section to wrap around the tire with enough clearance to avoid contact with tire deformation.  The fork material is 0.25” 4130.  I did my CAD design in OnShape and put the design through finite element analysis using OnScale (a companion to OnShape - and also free for non professional use).  I made a few adjustments based on the FEA results, basically increasing fork thickness to 0.25 inches, and adding additional material for the axle mounts.  It also let me know to be very very generous with my welds where the tube and fork met - as this is a high stress area.  With this done I felt confident that the fork would work.

But how to mount it to the airplane?

When I went on floats, I also added a C/S prop - and it turns out the governor occupies the same space the nose gear post wants to upon upward displacement.  So right away I knew I would have to either ditch the C/S prop or work a new nose gear mount solution that did not interfere.  Well I love my C/S prop, and the nose gear would require re-design anyhow if I was going to put a big tire up front.  So nose gear mount re-do was chosen.  Fact of the matter is I had anticipated this when I mounted the floats, and had installed the “rubber donut” nose gear mod at time - thinking I would use it when I went back to wheels. So there was extra structure from that mod to use - and I did as I will explain later.

I decided to keep the mounting as close to Mr. Heintz original principles as was feasible.  I kept the lower V-block - simply redone for the larger size telescoping tube I was using.  This required a new larger lower V-block, and larger V-block attach plate.

I designed a new upper bushing that would mount firmly and precisely within the now strengthened (from the donut mod) firewall U-channel.  To do this I made a  mount plate that mated precisely with the U-channel

(using existing bolts), and then a bushing which was
allowed initial alignment play until mounted to the mount plate.  This is very similar to the alignment mechanism initially used to mount the lower V-block.  I made the bushing of the same Nyloil material as the V-block.  

The load bearing surface of the Zenith FLG donut mod also bears the landing load for my setup.  The load is transferred from the fork to the bearing surface through a threaded center post, a large nut, a thrust washer/bearing/washer, a thick aluminum disk and finally a single rubber donut.  The threaded center post continues up above the bearing surface where it is held in alignment by a Nyloil tapered disk which is retained by a top thrust washer/bearing/washer, and nut.  The Nyloil tapered disk keeps the rubber donut centered.  The gear post is also held in alignment by the two lower bushings.

The two telescoping tubes are free to rotate.  The centered thrust bearings above provide the upper pivot surface, and the upper and lower bushings provide additional alignment.  Steering arms were placed to allow the gear to self center due to the downward force now provided by the shock (as the bungee had done before).  The steering arm posts were placed at the same distance from center as previously to keep the steering geometry the same.  This meant shorter steering arms, closer to the larger post.  The steering arms are on the bottom/outer tube, which is directly attached to the nose fork.  Rotating this tube turns the fork and wheel.  The top tube may rotate on its thrust bearings, but is fixed vertically, and does not move up and down.

Main Gear Mods….

I designed spacer blocks to move the 29” ABW main wheels out away from the gear further to prevent any interference with tire distortion, and to allow double calipers to fit without any interference with the main gear. These were easy to design, as they are just a 2 inch extrusion of the Grove axle base surface geometry (I have a grove main gear).

Fabrication…

So design in hand, and feeling at least somewhat confident about the whole thing, I moved on to fabrication.  I have a small bench top mill (Little Machine Shop 5500) that I have used for several years on small projects.  After growing more confident with CAD and 3D printing over the past few years, I felt up to trying to convert this mill to CNC which I did - this is a whole separate story.  Suffice to say that with patience, trial and error, and willingness to purchase extra stock materials, I can now produce a CNC part within some material, shape, and volume constraints.  I used that capability in several parts of this build.

After cutting the tubes to length, the first challenge was to get two tubes that telescoped nicely.   By the ordering spec they should have had about 0.010 of total clearance, or 0.005 circumferential clearance.  That was obviously too tight.  So I made sure the outer surface of my 2.25” tube was nicely uniform and without any marks or defects.  Then I went to work with a cylinder hone and worked and worked and worked (and worked) until I had a nice sliding fit, without any play.  

Next I placed holes for the shock mount bolts in the lower tube, and also a hole in the front bottom of that tube to allow for access to the shock nitrogen fill valve.  These holes would have affected strength, so I welded doublers over all of them.  I then notched out the top tube, and welded bolt attach plates into the recesses so that the top shock attach bolt would fit inside of the outer tube.  I welded three longitudinal support tabs to the top attach post.  Then I notched out three channels in the top tube, and welded a center post (5/8” 4130) into the top tube.  Then it was back to honing, sanding, and fitting until everything slid together nicely again without play.

I made the V-block support from 6061-T6 plate, using the CNC mill, with some hand adjustment for the final shape.  I then match drilled it for the existing mounting holes and attached it to the fuselage.

A new Nyloil V-Block was made also using the CNC mill. This was aligned and mounted just like the original 801 build instructions.

Next I made the upper bushing mount plate (6061-T6) and upper bushing mount (Nyloil) both with the CNC mill.

To mount the upper bushing mount plate, I removed several of the firewall U-channel reinforcement bolts.  I placed the mount plate in place, with the gear tubes, and upper bushing in place and aligned.  I then match drilled the holes through to the mount plate with a center drilled round spacer, to get #30 holes in the mount plate.  Then I removed the plate, opened these up and tapped them for the mount bolts.

The mount pate was then installed, with gear post and bushing.  The bushing was then aligned and #30 pilot holes were drilled from mount plate to bushing.  All was removed and the mount holes were opened and tapped as required for final mounting, and the mount plate holes opened up for clearance of a AN4 bolts.

Fabricating the fork…. Bending Jigs.

I began by creating a bending jig for the 0.25” 4130.  I took a 4 x 36” piece of thin plywood, and cut small grooves in to so that It would be very flexible.  I placed it around the outside of my 26” nose wheel, just as I wanted the fort to fit -  allowing enough room for tire deformation.  Measurements were recorded so that the same curve could be recreated on a flat surface, which was then done.  I placed the curved thin plywood curve on other piece of 3/4” plywood, and reinforced it with 4” sections of 2x4. 

Each 4” length of 2x4 backed up the curve, and prevented it from deforming.  Then I placed another piece of 3/4” plywood on top of the flat curve and 2x4 curve.  Everything was secured with strong construction screws.  Next I filled all gaps and filleted all edges with wax.  Then I poured high strength concrete into the mold, and let it cure.  Once all cured, I removed everything and was left with a nice curve matching my desired inside fork diameter. 

Next I made a steel jig for bending the curve using the concrete “die”.  Round bar end pieces were made parallel and 1/2” (0.25 per side) farther apart than the maximum curve width.  Structure was added to allow adequate depth of die penetration, and to support the two end pieces of round bar.  The important parameter for this is the final position of the end bars relative to the curve.  They must end up exactly 1/4” past the end of the curve once it is forced through the jig.  The other important parameters are of course strength and the ability to fit into a press or some other means of force to bring the cement die down.  

For the fork itself I started with an 18x36 sheet of 0.25 4130, as this was the narrowest 36” piece of 0.25’ 4130 available from spruce, which I cut down to 4x36”.   This was placed into the steel bending jig, and heated as hot as I could get it using two MAP gas torches. The concrete die was placed, and downward pressure applied until the curve was created.  The ends of the fork were clamped, and everything allowed to cool.   If I did this again, i would make the curve a little smaller than I wanted to better account for spring back.  As it was done it took a little post bending adjustment to get the curve just right.  

To create the axle mount points, I began with two short sections of 1.75 4130 bar, in which I had bored a 0.75” center hole.  A 0.25” groove (perpendicular to the bar center axis) was then machined pa way through. 

Welding….

All welding was done using TIG, and 4130 filler rod.  I welded the axle mount points to the end of the forks with 6 passes of weld bead.  I placed long piece of 0.75” rod through the axle mounts in order to maintain alignment.  Once the axle mounts were welded on, I opened them up to 1”.  I also welded the bottom/outer tube to the top of the fork - 6 passes of weld bead.  Then I welded on steering arms with AN5 bolts to capture the rudder pedal nose wheel steering linkage.  My steering arms are welded on to the tube externally, and do not pass through it.  Passing through would have been easier for alignment, but would not have allowed space for the shock.  I used slightly thicker 4130 tubing to allow a little better weld thickness where the steering arm joins the lower/outer tube.  Once all welding was completed, I took the upper tube/top mount and lower tube/fork to a heat treater to be normalized.   When it returned from the heat treater it required a little bit more of the hone to make everything slide nicely again.

Corrosion Protection…

I primed all of the internal and external surfaces (except mating surfaces) with Akzo Noble two part epoxy primer, and top coated the lower/outer tube and fork to match my grey paint.  I usually like to powder coat steel, but I decided to paint these because I want to watch the welds carefully for any cracking (which could be hidden by powder coating).

Final assembly….

I bolted in the V-block plate, V-block, and upper mount plate and upper mount bushing.

Final shock assembly was a little tricky, largely because I chose to install the shock in the inverted position.  This places the fill nipple at the bottom of the gear, and allows access for adjusting the shock.  It also allowed me to use the larger tube on the bottom.  But there are disadvantages…. oil sits at the low point in the shock, which when inverted is also where the fill nipple sits.  This means that if I need to add nitrogen, no problem.   But if I need to remove some - its a problem.  In the future I plan to make a drain/fill sump that will allow me to drain and refill the shock in the inverted position and retain the oil.  But for initial testing my strategy was to start from a low pressure and add nitrogen as needed to achieve my desired results.  In the end it turned out that I filled to near max pressure, so I don’t expect to have to adjust it again in the near future.  

I filled the shock with 95 ml of 100W aeroshell, then compressed it as far as I could without causing oil to spill out. The shock was then held fill nipple up until the fill nipple was eventually capped. I attached to end of the shock piston to the upper/inner tube, and bolted in in place with an AN8 bolt (probably overkill).  The shock/tube combination was held vertically and the lower/outer tube & fork were placed over the shock, and onto the inner/upper tube.  Once aligned, the shock was secured to the fork & tube weldment with one AN-8 bolt.  The fill valve was then installed.  The compressed shock was then placed fork side down and placed into the V-block.  Next the aluminum plate, disk, and rubber donut were placed above the upper bushing mount.  I  placed the 5/8 nut, thrust washer/bearing/washer on the top post.  I then added pressure slowly to the shock while guiding alignment of the inner/upper tube through the upper bushing, and capturing the aluminum plate disk and rubber donut as the top post advanced.  Once the top post protruded far enough above the load bearing surface, I placed the Nyloil cone, thrust washer/bearing/washer and nut onto the 5/8 top post and secured the two 5/8 nuts.

I then added slightly more pressure to the shock while lowering the weight of the nose onto the gear, and adjusting pressure as looked necessary.

I did encounter one problem that I had not planned well for.  That is that with enough upward movement of the lower/outer tube, it was potentially possible to cause the left steering arm to interfere with engine mount.  Fortunately this was at near the end of travel, and I choose to handle this by filling the shock to near max pressure (so it would be unlikely to compress to full travel) and drilling and tapping a hole in the upper/inner tube for an AN4 stop bolt that would prevent further compression before interference could occur.  

Testing…..

My first step was a slow taxi.  I fly from a grass strip, and there was enough uneven ground to give it a fairly good test.  After initial taxi I pulled the cowling and examined everything, which looked fine.  Video footage (from a wing tie down mounted camera) confirmed all looked good in motion. I safety wired the bolt heads, and proceeded to fast taxi testing.  

For fast taxi testing I got going fast enough to lift the nose and then applied full braking.  This gave it a good nose down tilt.  Post flight analysis all looked fine also.

I waited for nice day with light winds, and then took the 801 up round the pattern.  Landing seemed especially soft (not due to pilotage…) and post flight video analysis looked like everything was functioning well.  

So at this point, I’m ready to proceed into normal flying, and will follow on with frequent inspections until I am satisfied with the stability of my new setup.