Designing Float Rigging
(or any other aircraft)
There is a wide variation in the design of fittings for attaching floats to an aircraft. I believe that there are a few important rules to follow to achieve well designed rigging. While the S7 above is a very pretty aircraft on Aerocet amphibs with rigging that works just fine, the design of the various components can be improved.
A separate issue is the geometry of the floats relative to the aircraft which will be addressed separately. See Geometry .
With three different sets of floats to try in the spring of 2009, I needed a flexible rigging design. I wanted to get away from fittings that are built with fixed angles.
For example, this is a welded top forward fuselage fitting that was on one of my Rans S7 aircraft on Full Lotus floats. This one had the Cub style gear.
This type of fitting pretty much fixes the fuselage-to-float geometry. So, for example, you would not be able to shorten or lengthen the rear strut to change the angle of the floats to the plane without changing this bracket. Also producing it would require either a very accurate fixture or tack welding it right on the plane which would be positioned over the floats.
Here is a top rear fitting off the S7 on the Aerocet amphib floats above:
It clamps to the vertical tube and is bolted through the bushing Rans welds to all S7 airframes.
It does allow pivoting in a fore/aft plane but again demands one specific float width/height combination because of the welded tab. These floats also used a fixed tab on the floats to attach front and rear struts so, again, no variation in angles allowed.
This is the top forward fitting on that same Aerocet installation shown above. This design is reasonably good in that pivoting is allowed in fore/aft angle of struts and float width but it requires three machined parts, the formed U and a weld.
The lower strut to float fittings are
related to what the float manufacturer provides on the floats. On these 1941
The top front porkchop stub gear used by Cessna, allowed some for/aft pivoting but fixed the float width/height.
In the Aerocet case above, the streamline tubing used for strut material has an inner cross section that allows for 1 square material to be inserted but the inserted fittings must stay parallel to the length of the strut. This material is the equivalent of Rans wing strut tubing.
Many builders then used Rans 1 square clevises to connect strut to brackets like this:
The diagonal strut takes ¾ material
While the requisite pivoting is achieved, it does require the clevis because the strut cross section does not allow any angular difference between strut and inner 1 material.
There are two ways around allowing an angular difference between strut and fitting. One is to design a different strut cross section. The other is to do away with the 1 clevis.
This material shown below the Rans strut material, does allow the strut to mate to a bracket at a wide angular range. The material at the top of the picture was used on the Aerocet above and is the same as Rans wing strut tubing; the bottom material is the other style. Note the thicker centre section which not only adds considerable strength but allows some pivoting. Its disadvantage is additional weight.
The top tubing is 1 inside; the bottom ¾.
For weight saving we want to use ¾ block material. The bottom tube style will bolt directly to the ¾ blocks; the upper tube requires filler at the block.
Here are the various parts that then go together to make up the rigging:
Here is the top front setup on my S7 on Murphy 1500:
And the lower front block on PG floats:
Below is the aft lower attachment
And finally top rear bracket and block:
This is made from 304 stainless and tig welded. The jig shown further down is helpful here to enable tacking the tabs to the bracket with fittings in place to guarantee the strut bolt is parallel to the centre line of the fuselage. All the fingerprints are from the anti seize compound I use.
IMPORTANT POINT Concerning water rudder steering:
With the pull/pull cable system that most of us use for linking the water rudders to the air rudder, make sure that the steering bar (distance between the cables) at the water rudder is close to the length of the bar on the air rudder. For example, on the Rans S7 the air rudder cables are about 6 ¼ apart. On the murphy floats, the water rudder bar is only about 4. This will cause a tightening of the cables at full travel and can restrict movement and add stress to the rudder cables.
The steps are 2 x 2 x 1/8 alum angle sitting on triangular blocks with a ¼ thru bolt which goes through the strut material away from the center thick section.
To use the same type of attachment on Full Lotus floats we have made up stainless steel channels that will bolt to streamline spreader bars (and have unique hardware to mount the spreader bars to the Lotus stiffener tubes.
Of course since the joints can pivot, side to side diagonal bracing is required.
With these brackets in mind and several sets of floats to install, I realized an adjustable jig/fixture was in order. This rig in the next picture enables me to set any float centre to centre width, any fuselage height above the floats, any spreader bar position on the floats, any cg position relative to step and any angle between float and fuselage. It uses the fuselage landing gear sockets and four feet from the red S7 fuselage that is in the first picture (It had an unfortunate death).
The top wood structure allows a plumb bob to hang from a cg point; the bottom wood simulates the float step. The digital level gets things lined up.
Once Ive chosen a specific geometry I can set the jig, bolt the strut end fittings in the jig and then cut and fit the struts without going near the actual aircraft. Here the fuselage stub gear and upper rear brackets are in place.
This is a new fixture used to tack weld the tabs on that upper rear bracket which is done right on an actual fuselage. The intent is to get both front and rear pivot bolts parallel to the aircraft centre line in spite of the upward and inward tapering slope of the fuselage at the rear bracket:
Here is the S7 on the Murphy 1500 floats with rigging built using the jig and the fittings shown above. Float width is 80; fuselage is 24 above float and level line is at 3.5 degrees to float.
Diagonal and aft struts are the temporary telescoping steel tube.
Using the jig, the next set (LAS 1350) were rigged in a day (again, with temporary, angle iron diagonal and rear struts to confirm settings). I ended up being quite happy with the strut length and did not need to adjust.
Here is the final set-up on the 1350 floats:
And a similar set on the 1500 Murphy floats:
Full Lotus 1260s on S-7S
The last Lotus 1260s I weighed were 197 lbs ready to bolt on. This set is only 173 and of that, rigging was 40lbs (If you are interested CG is 5" fwd of the step. To do a proper weight and balance you have to weigh the completed floats and calculate the CG before installing them.
The V bottom on the 1500 makes for much smoother over the water runs on takeoff and touchdown. You can feel even a slight chop thru the Lotus flat bottoms just as you can with 1300s above.
The 1500 gave an indicated 60mph (this is a low actual. I think GPS would have said 70 or more) at 4500 rpm (my economy cruise).
Since everyone says the Lotus being so blunt and bulky are slower, I was surprised to see a 65mph indicated at 4500. This is likely due to the bigger "flat plate area" of the larger 1500's compared to the 1260's.
The video will confirm but I feel that takeoff run is the same or even faster with the Lotus than the 1500's.
Again, the value of my float rigging jig came out. I'd cut and drilled all of the struts and fittings last fall on the jig. Before I took the 1500's off I made up all the diagonal brace cables and assembled everything on the Lotus away from the airplane. Then I raised the plane and switched floats.
The rigging is somewhat unique with streamline spreaders which have been machined with flats where they sit on the stiffener tubes and where brackets mount. The spreaders bolt to the stiffeners with stainless U bolts which also hold the strut fittings. Most parts are either aluminum or stainless.
These floats use some recycled parts. The front and rear struts are off the Aerocets shown above; the diagonals are from an early S7 rear lift struts. Most of the other fittings are my design.
I got the
idea for that additional V brace at the front from a set a guy in
Lotus suggests the fore and aft struts meet the float centered between the two stiffener tubes. This results in them being in the way when walking fore and aft on the floats so I like them closer to the inner stiffener tube. One disadvantage of this mounting point is that on a hard touch down there is a bending moment on the spreader bars tending to curve the mid point down. The tight diagonal braces add to this bending force. The V brace resists this. Here is another installation with V braces on 2 spreader tubes:
Steps above are 2 lengths of stiffener tube.
OK so for the horizontal diagonals which aren't quite so critical I used cheaper turn buckles.
Below the steps are on (formed from alum sheet 040 or 060?).
The plane is loaded
for the trip to the new owner NW of LaRonge
For some background on the geometry of float mounting see: Geometry
My email is peterc which is at pipcom dot com. 705 877 8404
Complete rigging packages for S7 to most floats including Full Lotus using streamline tube spreaders.
Fuselage front and rear fittings and struts for S7 that were on the Aerocet installation. Could easily be adapted to other floats. Some clevises needed (Rans stock part)
LIFTING AN S-7
Traditionally lifting an airframe for installing floats employed four eyes on the top of the cabin near the wing spar attach points. A steel frame the size of the rectangle formed by the eyes allowed stringing cables up to a central hook so that the plane was lifted more or less horizontally.
This simplified keeping the fuselage level for the float installation work.
On the S-7 it is straight forward to get the two front points (see brackets below) and it may be possible to design something to attach to the rear spar to fuselage fittings but because Rans tech support people were uncomfortable with this method I elected to use just the front spar points and the rear fuselage lift handles. In this configuration the weight lifted at the tail is maybe 60 lbs and we have actually installed floats with the rear being held by a person. See movie shoot.
I once used the firewall motor mounts as a lifting point but I do not recommend this. Because it is further forward there is much more weight on the tail and, even more important, the airframe is less stable due to the lower lifting point on the aircraft so that if you let a wing tip drop it doesnt right itself.
The lifting rings used here were made from a 1 ¾ muffler clamp with one half of a long threaded rod nut welded to the U, a 3/8 eye bolt and a sleeve from a piece of 2 x 1/8 wall aluminum pipe to protect the spar carry thru and spread the load.
You can position the U with the nut welded on roughly where it goes on the spar carry thru and mark the windshield underneath. Then, with a long bit drill up thru the lexan and cuff to get a starting point for grinding away an oval hole with a dremel tool.
The eye bolts are a little longer than required and can be shortened if you want. I leave the rings on all year but they could be removed and a small cover put on the holes.
Not shown is a two foot long steel bar with two eyes pointing down and one in the center pointing up to hook onto the lifting cable which , in my case is a 12 volt marine winch.
You must pull up vertically on the fuselage eyes.
My homebuilt 170 pictured above will also be for sale once the recover is finished.