Roee Kalinsky's RV-7A Project

2007.07.22: (0.0) The kit as designed by Van's does not include a parking brake.  In my flying experience, I haven't found much use for a parking brake, except for one situation: getting out of or into the plane when parked on a ramp that isn't quite level.  I always use wheel chocks when parking, but you have to be outside the airplane to insert or remove the wheel chocks.  The parking brake is used to keep the airplane from rolling downhill while you're getting in or out.

I decided to implement a parking brake using the PV-1 Parking Brake Valve from Matco, which has been used successfully by many RV builders.  This valve is available for about $120 from Matco directly, or from Aircraft Spruce and others.  The PV-1 is basically a lever-actuated two-channel hydraulic check valve, and is installed in-line between the master cylinders and slave cylinders.  When the PV-1's lever is in the open position (normal flight position), the valve is "transparent" in both directions, and the pilot has normal brake pedal operation.  When the lever is in the closed position (parking brake position), the PV-1 becomes a check valve.  The pilot applies pressure using the master cylinders (brake pedals), and the PV-1 then holds the pressure to the slave cylinders after the brake pedals have been released.

Matco PV-1 Illustrated Parts List
Matco PV-1 Installation Drawing
Matco Typical Hydraulic Schematic

I elected to mount the parking brake valve on the firewall at the position where the F-6122-1 bracket would normally be installed.  That way, the rest of the plumbing is basically unaffected.  Same hoses, same tubing runs.  One limitation with mounting the valve assembly in this position however is that it protrudes about 2-1/4" aft from the firewall, while F-6122-1 only protruded about 1".  So the rudder pedals will have to be positioned such as to ensure adquate clearance at their maximum forward deflection.  I don't expect this to be a problem, since I'm only 5'9" and therefore will not be mounting the rudder pedals at the forward most position anyway.

To control the valve, I'm using a push-pull cable with a T-handle that will be mounted on the instrument panel or elsewhere within easy reach.  I first bought a cable that locks/unlocks with a 90 degree twist of the handle.  Aircraft Spruce p/n 05-15000, don't know who manufactures it.  Pretty inexpensive at $15, but it was a case of you get what you pay for.  Piece of junk.  I then bought a different cable that unlocks with a spring-loaded push button instead of a twist.  Manufactured by ACS Products (formerly part of Aircraft Spruce) as p/n A-1840, sold by Aircraft Spruce as p/n 05-13596 for about $45.  This is a high quality cable and is the one I will use.  It has an 0.062" diameter stainless steel wire core and a 0.220" outer diameter metal casing with an inner teflon lining for smooth operation.  It is sold in various lengths (I bought 8') but can always be shortened as needed.  Max travel is 3.5", locking force 10 lb.

To fasten the control cable outer casing to the mounting bracket, I bought a control cable clamp, Aircraft Spruce p/n 10464.  Being on the same page of the catalog as the control cable, I assumed it would fit.  Wrong!  It had a much bigger diameter than the cable, so I returned it.  Back to basics, I used a 3/16" cushion wedge type cable clamp (MS21919WDG3), which worked great.

To attach the inner core (the moving wire) of the control cable to the actuator lever on the parking brake valve I used a "cable end B nut" (why this is called a "B nut", I don't know),  Aircraft Spruce p/n 05-16245.  This worked out well.  Finally got something right on the first try!

According to the Matco PV-1 Installation Drawing, the actuator arm needs to be deflected at least 32º above the horizontal (in the orientation I'm mounting it) to ensure it's fully open, and at least 32º below the horizontal to ensure it's fully closed.  For maximum robustness, I decided to incorporate hard stops in both directions as part of my mounting bracket.  I didn't want to rely on the control cable's travel limit because the cable has a fair amount of flex over its length, and may also have a significant amount of thermal expansion/contraction.  I set the hard stops at ±35º, which yields a stop-to-stop cable travel of approx 1.8".  Note that in order to make sure that the valve's actuator lever limits out on the "valve open" stop of the mounting bracket and not on the cable's full forward stop, I will have to connect the cable such that it can't quite reach the cable's full forward stop.  I.e. when limited at the "valve open" stop, the T-handle will still be sticking out a little bit (say about 1/8") from its full forward position.

In addition to having positive travel stops, I decided to incorporate a snap-action switch (a.k.a. "microswitch") to sense whether or not the lever is at the full open (i.e. normal, not parking) position.  There will be a warning light in the cockpit that turns on any time the lever is not at the full open position (just like the parking brake warning light in the dashboard of my Toyota).  I think this is a worth-while safety feature, since mistakenly taxiing (or worse, taking off or landing) with the parking brake on could really ruin your day (loss of directional control, or having the brake pads catch fire).

[Update 2007.08.30: I just read on AvWeb that an Airbus A319 with 53 passengers aboard landed at Leeds Bradford Airport (UK) with the parking brake on.  Nobody got hurt, but quite a show!  Maybe I should sell my parking brake warning light design to Airbus ;-)]

Snap-action switches are available in many shapes and sizes, and many electrical and mechanical ratings.  This particular application is very low voltage and current DC (sensing only, I will have a separate driver circuit for the warning light), and low number of expected switching cycles (maybe one or two cycles per flight, so say 10,000 over the service life of the airplane), so almost any switch will do.  Still, for ease of future maintenance, I decided to go with a common form factor that is widely available from numerous manufacturers and distributers.  I will also probably use these same switches in other parts of the airplane to sense canopy closure, flap travel limits, etc.  Some example p/n's in this form factor:

C&K (formerly ITT-Cannon?) TM Series: TM-CG-D6-S-T15-40-C (Newark p/n 07F030)
E-Switch LS Series: LS-085-15-02-F075-C1-A (Digi-Key p/n EG4525-ND)
Omron V Series: V-15G2-1C25-K (Digi-Key p/n SW875-ND)
Cherry D4 Series: D41CR1LD (Mouser p/n 540-D41C-R1LD)

The specific switch I selected was the C&K TM series, because they are rated for a wide operating temperature range of -55ºC to 150ºC (-67ºF to 302ºF), which is always a good thing in aerospace, but especially in this instance where the switch is mounted almost directly to the firewall.  Most other switches are typically rated for a standard temperature range of -25ºC to 85ºC (-13ºF to 185ºF), which is probably also adequate, but it's nice to have extra margin.  Some of the other switch manufacturers also offer varieties of their switches that are rated for extended temperature ranges, but the C&K TM series was the only one I was actually able to buy in small quantities (available from Newark at $2.78 each, no minimum quantity).

The switch will be attached to the travel stop flange of my mounting bracket, spaced with two 1/4" high standoffs so that it hits the cable end "B nut" but clears the cable itself.  The standoffs are male-to-female, 4-40 thread, with the male end screwing into my mounting bracket, which is drilled and tapped.  The switch is then attached to the standoffs with 5/8" long 4-40 cap screws and split washers.

Note that the male threaded end of these standoffs is typically about 3/16" long, while the mounting bracket is only 1/8" thick.  The extra 1/16" protruding inwards causes an interference with the valve body on one of the two standoffs, so that standoff's male end has to be "circumcized" down to 1/8" length, easily done with just a little finesse on the bench grinder.

While I was at it, I also drilled and tapped a pair of mounting holes for a second switch to sense the full closed position.  This could be used to have another distinct warning for "valve not fully open, but not fully closed either", i.e. "parking brake valve in transit".  I think this is overkill for this application, and I probably won't do it.  But it was easy enough to make provisions to mount the switch while I'm still working on the mounting bracket on the bench.  Conversely, it would be very difficult to retrofit once the mounting bracket is riveted to the firewall.

The mounting bracket itself is of my own design, pretty simple, fabricated out of a single piece of 1/8" thick 6061-T6 aluminum angle.  It will be riveted to the firewall using all the rivet locations intended for the F-6122-1 bracket, plus one additional rivet.  The leg that gets riveted to the firewall is approx 3-1/4" wide, and the leg that forms the control cable mount and travel stops is approx 1-7/8".  I made it from 2-1/2" x 4" x 1/8" angle stock, which is available from Spruce, but expensive ($15 per linear foot), and I haven't been able to find a second source.  [For future reference: 2-1/2" x 2-1/2" x 1/8" angle stock, which is far more common, could be used instead.  Then, one of the existing columns of rivets in the firewall intended for F-6122-1 would just not be used.]

Since the flush rivet heads are on the forward side of the firewall, I added an 0.063" aluminum shim with clearance holes for the shop heads on the cockpit side of the mounting bracket, so the valve body will lay flat on the shim and not on the shop heads.  When riveting the mounting bracket to the firewall, I'll just have to make sure that none of the shop heads are taller than the shim thickness.  This is fine, since the nominal shop head height is half the rivet diameter, so for 3/32" diameter rivets that's 3/64", or approx 0.047".  The shim also prevents an interference between the sharp corner of the PV-1 valve and the rounded inside corner of the mounting bracket.

The valve and shim are finally fastened to the mounting bracket and firewall with two AN3H-13A bolts.  I decided to use plate nuts on the forward side of the firewall so that the valve can be installed or removed entirely from the cockpit side of the firewall.  This again is an ease-of-maintenance consideration.  The stackup is as follows, aft to forward (bolt head to plate nut):

AN960-10 washer (0.063" thick)
PV-1 valve (0.750" thick)
shim (0.063" thick)
mounting bracket (0.125" thick)
firewall (0.020")

That's a total stackup thickness of 1.021".  The bolt grip length is 1.000".  Note that I'm calling out an AN3H-13A, the "H" signifying a drilled head, so that the bolts can be safety-wired to each other.

2007.08.18: (1.0) Riveted the bracket to the firewall with Garnet's help (Garnet on the gun, me on the bar).  We were unable to rivet the nut plates though, because my rubber-cushioned swivel flush rivet set was too wide, and interfered with the shop heads of the other rivets we just set from the other side.  Should've seen that one coming.  I see two options: 1. Get a 1" non-swiveling flush rivet set and give it a shot, or 2. admit defeat and use some pop rivets.  Stay tuned.

2007.09.27: (1.0) Got a 1" non-swiveling flush rivet set, and with Greg Larson's help riveted on the nut plates.  I had to drill out the first two attempts because the flush heads stuck out.  I think this was made difficult because the firewall has a fair amount of flex.  But eventually we figured out that the trick is just very very light pressure on the bucking bar, and the rest went well.

2007.10.30: (12.0) [summay of several evenings] Brought out the output ports of the parking brake valve at right angles using AN822-4D elbow fittings.  I needed to offset the left and right sides vertically so they don't interfere with each other, so I made a downwords extension for the left using an AN911-1D male-to-male and an AN910-1D female-to-female pipe thread coupling.

I then completed the brake lines from the parking brake valve to the bulkhead fittings in my modified F-782B cover support ribs.  The brake lines (centers of the tubes) cross the F-601N-L firewall stiffener angle at 3" and 3.75" above the floor (F-772), and 1" aft of the firewall (F-601A-2).  They are secured with MS21919-WDG4 cushioned clamps to a little bracket I made that attaches to the F-601N-L firewall stiffener.  The brake lines then jog down and run aft along the floor at at approx 1.25" and 2.0" inboard of the F-601N-L firewall stiffener and F-772B-L floor stiffener.  The tube centers are 0.375" above the floor, i.e. there is 0.25" between the tubes and the floor.  They then ascend at 0.25" from tube center to the center section bulkhead, i.e. 0.125" between tube and bulkhead, and are secured to the bulkhead with cushioned clamps.  (For future reference, it works out that the nominal center-to-center run length of the tubes along the floor is approx 26.375".)  They then take a diagonal jog forward/up, and then a bend outboard to pass through the F-783B cover support ribs.  Keeping the brake lines as far aft as possible this way keeps them well out of the way of the fuel selector valve, fuel pump assembly, and fuel lines.  Note that instead of using the center hole in F-783B that Van's intended for the brake lines, I'm using the upper hole, which Van's intended for electrical wiring (I'll cut another hole for wiring).  This hole lines up almost perfectly with the center hole in my modified F-782B cover support ribs (still a very slight jog required to pass through the hole centers perpendicular to the ribs), and also makes it easier to avoid interference between the brake lines and fuel lines.  I also enlarged this hole to 0.75", which allows me to pass the brake line's flared fitting nuts through, and therefore the completed brake line tubing assemblies can be removed and replaced at will.  I'm also using MS35489-7 (AN931-4-12) rubber grommets that'll slide into this hole (I put the grommets around the tube before flaring), although there would be sufficient clearance between the 1/4" tubing and 3/4" hole that I think it would be fine without a grommet.

Note that Van's original design requires the builder to bend and flare the tubing in place, which would be damn-near impossible to do cleanly, and then the tubing, support ribs, and landing gear mounts would be impossible to remove without cutting the tubing.  I didn't like that.  I wanted to be able to do a neat job of it, and be able to R&R the tubing, support ribs, and gear mounts easily and without destroying them, at least during construction, so I don't end up with chicken and egg problems typical of this part of the build process.