Roee Kalinsky's RV-7A Project

2006.01.15: (2.0) Reamed out the bolt holes in the torque tube weldment.  Drilled the bushing blocks.  While the bushing blocks should be "match drilled" to the pre-punched holes in the bulkhead, this is not practical as the bulkhead is already installed.  I drilled the the bushings from measurements, which came out perfect for one and not quite for the other.  I think part of the reason was that I was just holding the blocks down to the drill press base, and the drill tends to grab the plastic.  I'll order a replacement part and go pick up a drill press vise.

2006.01.19: (1.0) Refer to DWG 20.  Electric flaps were an option on the RV-6(A) and are standard on the RV-7(A).  Van's design of the actuation mechanism for the electric flaps is, in principle, simple and effective.  Unfortunately, there are a couple of serious flaws in the details of the design that I find unacceptable.  Flaws that are readily apparent on the plans.  Flaws that could have easily been corrected, but for whatever reason were not.  Annoying...

1. The motor assembly has a gearbox with a jackscrew (or similar) mechanism used to extend or retract the actuator arm.  But when I bench tested it I found that the actuator shaft doesn't extend or retract.  It just spins.  If I hold the actuator shaft to keep it from spinning (with considerable torque), then it extends or retracts.  An query  to the Yahoo group confirmed that my actuator is not defective, that's just how it is.  A good design would have a built-in mechanism to keep it from rotating, but instead Van's design relies on a Mickey Mouse "safety wire" through the actuator shaft and the clevis bolt.  The problem with this is that the "safety wire" is not really a safety wire at all, in that it relies on its overall rigidity, not just in direct tension, to keep the actuator from spinning.  Furthermore, on every cycle of the flaps, the actuator will exert a torque on the wire in either direction.  After who knows how many cycles, the wire will work harden, will fatigue, and eventually fail.  Then the actuator will exert a torque that will only be opposed by the jam nut on the rod end bearing, which will eventually work loose and the rod end will unscrew from the actuator.  Not cool.

I decided to correct this by adding a retention mechanism to the actuator itself to prevent the shaft from rotating.  After considering several design ideas, consulting with Vay, Annie, and the Yahoo group, I decided to go with a simple design using a couple of collars with an offset rod and a guide hole.  This design concept was used successfully by RV builder Lorne Heise, and in fact I decided to shamelessly produce a very similar design to his.  I drew up a design, and now it's up to Annie (an ME/BE grad student and machine shop TA) to go make me one.  Thanks Annie!

2. The actuator shaft and the (weldment) torque tube arm don't line up with each other!  Conveniently, there is no single view on the plans that shows both.  But studying the plans still reveals the problem.  The motor/actuator assembly is asymmetric, with the motor hanging to the right of center and the actuator shaft to the left of center.  The arm on the torque tube weldment though is perfectly centered, yet the two are expected to meet up.  The "nominal" misalignment appears to be about 5/16".

Van's could have fixed this problem by simply welding the arm on the torque tube 5/16" left of center.  Why they didn't, who knows.  Presumably this is a relic from the RV-6(A) manual flaps.  But come on, retooling a welding jig is not that big a deal, and in this case would make the design "correct" and save a lot of trouble for us builders.

Some builders have reported "fixing" this problem by mounting the actuator at a slight angle away from vertical.  Doing this appears to solve the problem in one (and only one) flap position, but now the actuator and the arm will move in slightly different planes, and therefore will try to twist the support structure.  There may be enough slop and flex in the system to get away with this, but that's just plain kludgey.

Carefully studying the dimensions, it looks like I can compensate for some of the misalignment in two places: 1. mount the actuator as far to the right as it can go without causing interference, and 2. connect the rod end bearing to the far left end of the clevis, with all the shimming washers on the right.  Looks like between those two I can make up just over half the misalignment.  Will this be enough to get me home?

2006.02.08: (2.0) Making brackets, etc.

2006.02.10: (8.0) Working on the support structure (did prep, prime, assembly...).

2006.02.21: (1.0) Received another bushing from Van's and drilled it using my new Craftsman drill press vise.  Came out perfect.  Interestingly, this bushing as delivered by Van's was machined to dimensions slightly larger than nominal, and the edges have significant burrs.  The others were dead on dimensions and neatly finished.  Not really a problem, but Van's occasional sloppiness in manufacturing/QA doesn't give me the warm fuzzies.

2006.02.24: (2.0) And on the next problem...  With the bushing blocks pushed onto the weldment as far as they go, the distance between them is 3/16" wider than it needs to be to fit the fuselage.  The tube wasn't bottoming out against the back of the bushing, but rather the welds connecting the arms to the tube were the limiting factor.  After studying the measurements, comparing to other builders' measurements, and consulting with Van's, it became apparent that the basic measurements are all correct, but the welds were sloppy and the fillets just extended too far outboard.  Van's suggested drilling the blocks off center to space them out further, but I didn't like reducing the edge margin there.  My solution (which Van's approved) was to bevel out just enough of the bushings to make clearance for weld fillets.  I used a 45º router bit.

2006.02.25: (3.0) With the weldment and outboard bushings now in place, I fitted the flap actuator motor.  While I was already anticipating a misalignment (see 2006.01.19 problem 2), I now found that it's 1/8"  worse because the actuator arm is welded to the tube not quite square.  Of course, it's tilted to the right, to make the misalignment problem worse instead of better.  Anyway, by offsetting every connection as described above (2006.01.19), and mounting the motor at a slight angle away from vertical, I got things to fit.  Despite the slight angle, actuation is reasonably smooth.  I'll have to wait until the whole structure is put together (side covers and all) to see if there's any significant flexing.

2006.02.26: (0.5) Drilled the center bushing block to the floor. 

2006.03.03: (0.5) Of the two holes drilled into the floor structure for the center bushing blocks, only the aft one can be fitted with a standard nut plate.  The forward one cannot because of how the baggage floor rib and the bulkhead overlap at the joint.  I fabricated a little plate that a nut plate can be attached to at a 45 deg angle, and is long enough that it can be riveted to the floor rib past the lap joint with the bulkhead.  Better explained with a couple of pictures: 

2006.03.15: (1.5) Drilled the cover panels to the support channels and installed nut plates in the channels.

2006.03.17: (1.0) Cut out a clearance in the left cover panel for the bolt head.  Deburred both cover panels.

2006.03.19: (2.0) The bolts are spaced pretty far apart per the plans, which allows the edges of the cover panels quite a bit of play between the screws.  The loose edge looks bad, and could easily get caught on something, and could easily vibrate and make noise.  So I decided to double the number of screws attaching the cover panels to the support channels, reducing the screw spacing from about 6" to about 3".  Drilled, installed nut plates, etc.

Note that I positioned one of the screws just above the cutout for the bolt.  This took some careful planning in precisely positioning the screw hole, as well as the nut plate rivets, as to maintain edge distance and not compromise the strength of the reinforcement plate on the inside of the channel.

Ideally I would have liked to add another screw just below the cutout for the bolt, but adding a nut plate there is likely to interfere with the flap motor, so the one on top will have to be good enough.

2006.8.20: (0.0) I finally got around to putting together and testing my contraption for keeping the actuator jackscrew from free-spinning.  Well, this first prototype didn't operate very smoothly.  The problem was that the aluminum rod sliding through a hole in the aluminum collar just tended to bind and grind.  I concluded that aluminum is just too soft or produces too much friction in this kind of interface, with that much force applied.  So I came up with an ECO ("Engineering Change Order", industry jargon for a fix to make it work):

1. I replaced the aluminum rod with a stainless steel rod.  Stainless steel is much harder than aluminum, and tends to hold a smoother surface.

2. On the collar side, I added a brass sleeve that will act as the slide bushing surface instead of the aluminum.  I found some brass tube that had an outer diameter of 9/32" and wall thickness of 0.014", yielding an inner diameter of 0.25325" (just 0.00325" greater than the 0.25" rod).  I drilled out the hole in the aluminum collar to 9/32", which actually made for an interference fit as is.  The brass tube would not slide through it with hand force.  Perfect!  So I put the brass tube in the freezer and the aluminum collar in the oven at 200ºF for half an hour.  With that temperature differential I was able to push the tube in about a millimeter by hand, and then tapped it in the rest of the way with a mallet.  I think a couple of extra taps with the mallet also served to compess the tube longitudenally and expand it radially ever so slightly, for a tighter fit.  In any case, they were never coming apart again, especially after a minute when their temperatures fully equalized.  I then cut off the long tube end and filed everything down flush, leaving a perfect cylindrical insert.  Note that I bought the brass tubing at Marshall's Industrial Hardware, but it's also available at any hobby shop in those displays of small metal stock.  It is "K&S Engineering stock #132".

With that ECO complete, everything operates smooth as silk.  So stainless steel against brass really does make a much better slide bushing interface than aluminum against aluminum.