Roee Kalinsky's RV-7A Project

Rudder / Brake Pedals
Home Up News Specifications

 

Back Up Next

Rudder / Brake Pedals

Start: 2005.08.05, Completion: in progress, Hours: 19.5

 

2005.08.04: (1.0) Prepared and fitted the F-xxxx bracket that holds the brake lines on the firewall.

2005.08.16: (1.0) Started working on the brake pedals, deburring the F-6117BPP brake plates...

2005.08.17: (1.0) More deburring on the brake plates and F-6117A side plates...

2005.08.18: (2.0) Done deburring the brake plates and side plates.

2005.08.20: (1.5) Cut the F-6117C angle angles from stock and match-drilled to the brake plates and side plates.

2005.08.21: (1.0) Trimming and deburring the F-6177C angles...

2005.08.22: (1.0) Trimming and deburring the F-6177C angles...

2005.08.23: (1.0) Done trimming and deburring the F-6177C angles.

2005.08.28: (1.5) Enlarged the holes for the brake pedal hinge bolts to final size.  Note that the two holes in each pedal need to be co-linear so that they can rotate smoothly about the hinge axis.  Since the side plates and reinforcing angles are not quite perpendicular to the hinge axis, I drilled/reamed the holes with the pedal assembled, jigging it so that the drill is perpendicular to the vertical centerline of the pedal.  Did a fit check to the rudder pedal weldments -- all good.

2005.08.29: (1.5) Started fitting the master cylinders in assembly with the rudder pedal weldments and the brake pedals.  Setting the geometry such that in the neutral position, the left and right rudder pedals are angled equally to center up, i.e. forming an isosceles triangle with the plastic bushings.  In this position, the brake pedals are set along the centerline axis of this triangle, i.e. perpendicular to the plastic bushings.  Note that in the airplane the plastic bushings (and the stringer they're mounted to) are not quite parallel to the longitudinal axis of the airplane, so all the geometry described above is relative to that plane and will be tilted by a few degrees when mounted in the airplane.  One pedal done, three to go.

2005.08.30: (3.0) Fitted the master cylinders for the remaining three pedals.  Match-reamed all rivet holes to final size.

2005.09.06: (2.0) Alodined all the pieces of the brake pedals.

2005.09.07: (1.0) Sprayed AKZO on all the pieces of the brake pedals.  [also parts for primary flight controls, but here I'll log 1.0 h attributed to the brakes]

2005.09.07: (1.0) Riveted the brake pedals.

2006.02.07: (0.5) Drilled the bolt holes through one of the plastic bushing blocks.  Unfortunately I used a newly acquired drill press vise from Harbor Freight.  At $7 you don't expect high quality tools, but what could be so bad about a drill press vise?  As it turns out, the jaws aren't quite perpendicular to the base.  So I ended up with bolt holes that weren't quire straight through the bushing block.  So this $7 tool has now cost me $14 dollars, the other $7 being a replacement part from Van's.

2006.02.08: (1.0) Armed with a new drill press vise from Sears ($40), I drilled the other bushing block which came out perfectly straight.  I then used the bushing block as a drill guide for drilling the left side longeron.  I drilled a total of 6 holes for 4 possible fore-aft positions, starting at the most forward location shown on the plans, and back in 1-5/8" increments (half the center to center distance between the holes in the bushing block).

Note: I used a 12" long 3/16" drill bit, squeezed in between the skin and the longeron which are not yet riveted together.

2006.03.31: (1.0) Drilled a new bushing block for the right side.  The dimensions of the new bushing block are measurably bigger than nominal, as was the case with the replacement bushing block I bought for the flaps actuator.  The edges are also not as clean as the original.  What's going on?  Anyway, I made sure the hole spacing is still nominal so it would be interchangeable.  Drilled the longeron (same technique as the left side).

2006.04.02: (1.0) Drilled the center support bracket to the firewall upright.  I carefully clamped it in position, sliding the pedal tubes back and fourth in the bushings along the longerons to ensure that the bottom of the bracket is positioned in the right plane.

There is a comment in the builder's manual about locating the rivet holes so that they don't interfere with the rivets for the firewall recess along the other leg of the upright angle.  It's not clear to me how they would interfere, even if they're right abeam each other.  But nevertheless I offset the holes a little to separate the rivets from each other.  Probably unnecessary.

2006.04.03: (1.5) Drilled the bolt holes in the center support bracket for the plastic bushing block.  I could only do this for the three forward positions that I drilled on the longerons.  The center support bracket wasn't long enough for the fourth (aft) position.  Oh well.

The method I used was this:  Mark the centerline on the bushing block and the corresponding line on the the center support bracket.  For each of the three forward positions: 1) install the pedals in position on the longerons, 2) place the upper half of the center bushing block on the tubes, 3) cleco the center bracket to the firewall, 4) adjust the bushing block and center bracket so they line up properly and clamp them to each other, 5) remove the bracket from the fuselage with the bushing block clamped to it, and finally 6) match-drill using the bushing block as a drill guide.  A lot of steps, but results in a perfect fit in all three positions.

I did run into one snag...  The bolt holes in the center bushing block are 1/4" closer together than the bolt holes in the outboard bushing blocks, resulting in a hole pattern in the center bracket that is different from the longerons.  In the longerons, the aft bolt hole of the first (forward) position doubles as the forward hole of the third position.  But in the center bracket, these are different holes, 1/4" apart center to center.  Being 3/16" holes, this obviously violates normal hole spacing.  But I concluded that in this application there should not be significant stresses that would cause a problem.

2006.04.04: (2.0) Trimmed, cut the lightening holes, and edge-finished the center support bracket.

2007.04.17: (0.0) I've had some reservations about the design of the rudder pedals, and potential for binding.  The problem is that the hinge points for the pedals (as well as the master cylinders) are just two adjacent pieces of metal with a bolt through them.  For the pedals this is potentially problematic because force applied to the pedal will tend to twist the left and right bolts so that they're not co-axial.  Also, just the idea of rotating around a steel bolt as a "bushing" is far from ideal.  Now recently I've read some threads on the VAF forums discussing people actually having problems with the brakes dragging because the pedals don't always return all the way to neutral.  Some people have remedied the problem by adding beefier return springs to the master cylinders, but to me that seems like a band aid and not a cure.  I decided a design change is definitely called for.

First of all, the short bolts on the left and right hinge points will be replaced by a single long bolt (AN3-56).  These are available from Spruce at almost $10 each (ouch!), but so be it.  Then, I want to add a better busing surface between the bolt and the pedal.  I found some plastic flange bushings on McMaster-Carr and at Marshall's Industrial Hardware that might do the job nicely.  Problem is I need to make room for the bushings' flanges by either spreading the mounting flanges on the rudder pedal weldments or by machining down some clearance in the brake pedal flanges.  TBD.

2007.06.01: (0.0) After some experimentation, the best solution seems to be just the long bolts with no plastic bushings.  I still like the idea of the plastic bushings in principle, but none of the bushings I was able to find or make worked very well.  So with metal to metal contact around the bolts, I'll just have to make sure the holes are slightly oversized so they rotate freely, and keep them lubricated to minimize wear.

Back Up Next

 

 

Send mail to roee@kalinskyconsulting.com with questions or comments about this web site.
Copyright © 2003 Roee Kalinsky
Last modified: July 21, 2007

Visit my consulting web site at www.kalinskyconsulting.com for your engineering needs.