March 2014 Archives

I wasn't able to spend much time at the hangar today, but I did get down there for a bit in the afternoon.  The winds were way too strong to fly (at least until I get more practice in the plane), but I did get a chance to resume work on the gear leg fairings.  I had previously trimmed them to size and put strips of tape along the aft edge as reference points to ensure they're straight.  I then clamped the tailing edge with some cleco clamps and drilled the hinges to the inside of the fairing.  After countersinking, I removed the hinge and deburred it.  I used some of the G/flex epoxy from West Systems to bond the hinge halves to the inside of the fairing and then riveted into place with some soft rivets.  I only had time to do the right fairing today, but hopefully I can knock out the other one tomorrow.

I had been planning on fabricating some jacking points for a while now.  Even though I'm going to lift the plane by the engine mount using the shop crane I bought, I need something under the wings to keep the plane from swinging.  Since I fabricated aircraft jacks awhile ago, using them to support the wings seemed like the best plan.  I bought some 1" galvanized pipe caps and drilled them for a 3/8" bolt.  I then inserted a 2" long 3/8"-16 bolt through and rightened it down with a washer and nut.  This is screwed all the way down to the stop I installed in the wing tie-down point.  I might trim about 1/4" off the bolt to reduce the bending load a little further.

The cap just fits over the end of the ram on the jack.  It has just enough room to move around that I can insert it over the ram without having to be perfectly aligned with it.

I wrapped up the hinge on the other gear leg.  I'm probably going to end up shortening these since they're a bit longer than necessary.  I need to fabricate the clamping pads at the top of the fairings, and then I can align them to the longitudinal axis of the plane.

The clouds lifted enough in the afternoon that I could take the plane up for another flight.  I took the opportunity to calibrate the AOA system.  On the SkyView, this is done by doing a series of stalls in various flight configurations and hitting a button on the EFIS each time.  I also took the plane up to 8,000 ft to measure the true airspeed.  I'm still slow because the wheel pants and gear leg fairings are not installed, but I saw 162 kts.  I expect to be near 175 kts after the fairings are installed.

I also intentionally ran my right tank dry in flight so that I can do the final fuel gauge calculation. The SkyView system annunciated "fuel pressure" just a second or two before the engine started stumbling.  I switched tanks and hit the boost pump, and the engine surged back to life within 3-4 seconds.  The experience was very different than when I've run a tank dry in my Bonanza.  In the Bonanza, the engine power winds smoothly down with no surging or stuttering.  After changing tanks, the power winds smoothly back up.  In the RV, the power loss was much more abrupt and there was quite a bit of stumbling before power was smoothly restored.
Although I had previously calibrated the capacitive fuel senders by setting the empty and full points, I still needed to fill the tank in 2 gallon increments to create the voltage to fuel level mapping.  I started with an empty tank and created the first calibration point.  I thought it was a little odd that the sensor showed 5 volts, but didn't give it too much thought.  I added the first 2 gallons, but the sensor still showed 5 volts.  I was pretty convinced something was wrong at this point, but I added a couple more gallons to confirm.  Sure enough, the sensor still showed 5 volts.  I started to worry that the sender was outputting a bogus value, but I decided to confirm all of the settings first.  Fortunately, I pretty quickly found out that I hadn't configured the correct input pins on the EMS.  I needed to switch from the default (pins 20 and 21) to the ones I wired the senders to (pins 22 and 31) and change the input type to capacitive from resistive.  With that change, I was getting voltage values that made sense for both tanks (2.9 volts from the left tank which is probably around half full and under a volt for the right tank which had the four gallons I just added).  Unfortunately, I now need to empty the right tank again to redo the calibration.  I could try to drain the four gallons out, but it's probably easier to just fly again and run that tank dry.
After burning off the four gallons in the right tank, I calibrated it in the SkyView system.  I'm using a scale that has 1 gram accuracy, and I'm measuring roughly 1 gallon at a time.  Given the accuracy of the scale, I should be able to measure the fuel capacity within one tenth of a gallon over the 21 gallon capacity of the tank.

I added two gallons at a time up through 20 gallons, hitting the add button at each point on the SkyView.  I used the opportunity to calibrate my dip stick at the same time.  Finally, I filled the tank to the very top and measured the amount of additional fuel I added.  I ended up with 20.43 gallons of usable fuel.  The claimed fuel capacity is 21 gallons, so there's only a little over 1/2 gallon of unusable fuel.  That's more than I was expecting, but the flop tube pickups may leave a little more fuel in the tank than the standard pickups.  I still need to do some in-flight checks to determine unusable fuel in all normal flight attitudes including full slips.

Unfortunately, I almost completely drained my battery in the process.  I was down to 10.2 volts before I shut everything down and hooked the plane back up to the trickle charger.  I'm planning on flying again in a couple of days, so I hope the battery is charged enough.  I'm thinking I might need to replace the battery anyway since I killed it at least twice during the build (so dead that the trickle charger wouldn't bring it back).
I got down to the airport mid morning so that I could fix the lower empennage fairing.  I trimmed the extra glass I added and then worked on the rivnut that replaced the ClickBond nut plate that popped off.  I drilled the hole out to 0.182" and then filed a notch for the keyed rivnut.  The keyed variety has a small protrusion under the head to prevent the rivnut from rotating if it ever loosens up.  Finally, I reinstalled the fairing.  I still have some filler to apply, but it's looking much better.

After buttoning everything up, I took the plane out and flew for a little bit.  The right fuel tank calibration looks good and showed a full tank, so I switched to the left tank so I could run it dry in order to calibrate it.  I headed down to Hollister so that I could help Greg with the reassembly of his RV-8.  I got there a little bit before the plane and the crew showed up, so I stayed in the pattern for a bit and practiced wheel landings.  I bounced one, but the other three were very nice.

After helping to unload, we grabbed some lunch and got started on the reassembly.  After some final prep work, we got the wings slotted in and a couple of close tolerance bolts in on each side. We got both ailerons attached and hooked up the fuel and vent lines.

It started getting dark and I needed to be back on the ground at South County before night.  I headed out right around sunset and got back up to South County in just a few minutes.  It was still fairly bright out, so I stayed in the pattern for a bit and did some touch and go's as the sky darkened so that I could get used to the sight picture and get a feel for how bright the landing and taxi lights are.  I stopped about 10 minutes before it got truly dark, but it was good practice and it looks like the lights are plenty bright.
I got down to the airport really early this morning because I wanted to take care of a couple of things and get some flight time before heading back to Hollister to help Greg.  I pulled the cowl and gave everything a thorough inspection.  One of the things I found was that the oil filler tube could be rotated a little bit by hand.  Since the safety wire hadn't moved, I think the fiber washer just compressed a bit.  I used a wrench to snug the tube back down and then redid the safety wire to keep it tight.

I also adjusted the high RPM stop on the governor 1 full turn to increase the maximum RPM a bit.  I've only been seeing a little over 2600 RPM at the stop.  Although I'll likely have to adjust it further, I redid the safety wire to ensure this can't move.

I buttoned everything back up and took off again.  I figured I could get at least an hour of flight time before landing at Hollister again.  I decided to shoot an instrument approach to get some more experience with the autopilot and interaction between the GTN and the SkyView.  I loaded up the GPS 31 approach at CVH with the RUDNY transition and engaged the autopilot.  I duplicated the approach waypoints in the SkyView since they don't yet automatically stay in sync with the GTN (though this is coming).  The GTN did turn anticipation and smoothly guided the autopilot onto the initial leg of the transition.  I programmed an altitude step-down and commanded a 500 fpm descent.  The autopilot turned smoothly to intercept the final approach course and held the descent rate rock solid.  I broke off the approach a couple of miles out and climbed back up to 6,000 ft.

I headed over to the acro practice area and did a couple of rolls to warm up.  I then tried a loop for the first time.  The plane really builds speed fast on the down line, but pulling some G's takes care of that nicely.  Next, I tried a hammerhead on about an 80º upline to keep some positive G on the engine for oil pressure.  The rudder is quite a bit more powerful than anything I've flown before, so the rotation at the top was crisp and quick.

Finally, I decided to do a few spins to learn how the plane recovers.  The plan was to start with an incipient spin entry with an immediate recovery.  I climbed up to 8,000 ft and positioned myself near the South County airport in case I needed it.  I pulled the power smoothly back to idle and started slowing up.  Right at the stall break, I fed in full left rudder and the plane rolled about 90º to the left and the engine stopped cold.  The prop came to an immediate stop straight up.  I had already initiated the spin recovery and I immediately rolled level and got the nose down into a glide.

One of the questions Dave Morss asked me to determine if I was ready for flight testing was how I reacted in an emergency situation.  I've only had a couple of surprises in my years of flying, but I knew that I tend to remain calm and work the problem.  When the prop stopped, my only reaction was to utter the word "interesting".  After establishing the glide, I hit the engine start button and got the propeller windmilling again, but the engine still wasn't producing power.  I still had the mixture pulled way out, so I opened the throttle to give the engine some more air.  After a few seconds, the power came back smoothly.  I climbed back up and decided to try again, but this time over Hollister since they have much bigger runways in case I couldn't get the engine started again.

After arriving over the airport, I again pulled the power to idle and started to slow up.  I slowed more gradually this time so that I could monitor the engine more closely.  As the plane slowed, the engine RPM slowly dropped.  When it got down to around 400 RPM, I knew something was wrong and I added power back.  The engine stumbled a bit as the power came back up, but it ran smoothly again at around 75% power.  I pulled back to around 40-50% power and the engine started running rough again.  I brought the power back in and started examining the engine monitor more closely.  My EGTs were down a bit, so I decided to try leaning the engine further and repeating the approach to stall.  This time, the RPM held steady at about 700 RPM all the way through the stall break.  I didn't repeat the spin entry, but I feel comfortable now that the problem was simply an over rich mixture for the altitude.

I landed at Hollister to help Greg with his RV again.  It took much longer than we expected, but we were able to wrap up all of the spar bolts and get the nuts torqued.  The ailerons and flaps still need to be hooked up, but other than that, the plane looks pretty close to being ready for inspection.

I knew I was pretty close to empty on the left tank, but I wanted to run it dry on the flight back, so I left a little early to ensure I would have sufficient light.  I ended up flying for nearly 20 minutes before the tank ran dry.  I then switched to the right tank and landed at South County.  
I wanted to lay up a thin layer of glass over the rivets holding the hinge material to the gear leg fairings.  I used a couple of strips of the lightest cloth I had on hand and covered them in some Dacron (which is what you see here).  I'll do the other side of these soon and then I can put a thin coat of filler on the fairings.

No pictures tonight since it would look identical to last night, but I glassed the other side of the gear leg fairings.  I also stripped off the peel ply from last night and trimmed the excess epoxy.  I'll need to sand down a few high spots before applying the filler.
I calibrated the left fuel tank tonight.  I ended up adding 20.44 gallons of fuel which is only 0.01 gallons more than the right tank.  That's way less than the margin of error for the fuel measurement I was using, so I'm considering the two tanks identical in capacity and in unusable fuel level.  I filled both tanks all the way to the top so that I can begin calibrating the fuel flow sensor on upcoming flights.

Flight Testing

| | Comments (0) | TrackBacks (0)
I spent the whole day yesterday helping my buddy Greg get ready for his first flight in his RV-8.  He had his airworthiness exam in the morning, then I gave the plane one last thorough inspection before everything was buttoned up for the first flight.  Since the flight went great and there were no squawks to fix, we decided to meet up for lunch in Watsonville with some other friends and then go do some loose formation work.  Greg got several very nice pictures of my plane in flight.

We traded places and I flew formation next to him for awhile.  I'm new to formation flight, and it's definitely hard work flying in tight with another plane.  I was constantly adjusting the controls and varying the engine power.

After 15 minutes or so, we broke formation and I headed out to the central valley and climbed to 10,000 ft for some higher altitude tests.  On one of my earlier flights, I had a brief episode where the fuel flow indication briefly dropped to zero.  The fuel flow itself didn't drop since the engine was continuing to make power and my airspeed didn't vary.  I had just ran a tank dry for the fuel calibration, so I chalked it up to some air bubbles in the fuel line interfering with the turbine in the fuel flow sensor.  Today however, I started getting some fuel flow indication fluctuations in the climb to 10,000 ft, and the flow indication stayed very irregular while I was up there.  When I came back down to 6,000 ft, the fuel flow indications were rock solid.

I'm starting to think the problem might be heating of the fuel flow sensor.  The sensor is after the fuel servo, and below the engine where it is exposed to the heated air that just flowed past the cylinders as well as radiant heat from the exhaust pipes.  The only way the sensor is cooled is by the fuel flowing through it.  Since there is less fuel flowing through at higher altitudes, it can pick up more ambient heat from its environment which can lead to bubbles in the sensor.  I've ordered some extra thick fire sleeve from McMaster Carr, and I'll use that to wrap the sensor and adjacent fittings.  I may also wrap it in some reflective foil wrap to reflect radiant heat from the exhaust pipes.

I landed back at South County with 9.9 hours on the hobbs, so I decided to take the plane down for an oil change and thorough inspection.  While waiting for the oil to drain, I pulled out the interior.

Oil Change

| | Comments (0) | TrackBacks (0)
After letting the oil drain completely from the engine over night, I pulled the oil filter and oil screen.  I really need to figure out a less messy way to pull the filter as I spent more time cleaning up spilt oil than actually removing the filter.

I also pulled the drain cap off of the low point at the oil cooler.  I captured some of this in a cup to get an idea how much oil is in here.  I'm guessing it's at least half a quart.

Here's the oil screen.  I've heard it's not uncommon to find some carbon particles in here, but mine was clean as a whistle.

I reinstalled the screen with a new crush washer and safety wired it shut.  I'm really happy with how easy it is to access all of the parts of the engine necessary for an oil change.

I wrapped up the oil change by installing a steel cap that Greg gave me to replace the aluminum one that I previously had installed.  I installed and torqued it down in order to mark where to drill for safety wire, then removed and drilled it.  I now no longer have any aluminum fittings anywhere forward of the firewall containing flammable liquids (oil, fuel and brake fluid).  I then filled the engine with oil.

While I had the interior out, I decided to install a couple of RAM ball mounts.  While the upper side panels were still in place, I marked and drilled through them and the upper channel here.  I then removed the upper side panels and installed some NAS1329A08K75 rivnuts.

I reinstalled the upper side panels and then attached the RAM ball mounts.  Here's the one on the passenger's side.

Here's the one on the left side.

I've been flying with an iPad mini for ForeFlight since I like it as a nav backup and I keep my POH, avionics owner's manuals, etc. in there.  I was just laying it on the seat, but that's a pain when I want to do some acro.  Mounting it should keep it handy but also keep it from flying around the cockpit when inverted.

I previously sent the backup battery from my left SkyView screen into Dynon because it didn't seem to be charging during the first few flights.  They charged the battery and tested it and it was fine.  They sent it back to me so I installed it tonight and ran the battery test on both screens.  Both passed by powering the screens for 45 minutes (and new batteries should power the screens for more than 1 hour).

While the battery test was running, I replaced a couple of cable clamps in the engine compartment with non-cushion bowden clamps that hold the cable much more securely than the cushion clamps I was using before.  This is the clamp for the oil cooler butterfly control cable.

I also replaced the cable clamp at the forward end of the alternate air door control cable.

Finally, I decided to pull and clean the fuel filter again.  There was a surprising amount of gunk in the filter.  Most of the volume was some sort of fuzzy material, but there was also a lot of dust and even one small metal particle.  I redid the safety wire and installed it back in the airplane.

Adjusted ADAHRS

| | Comments (0) | TrackBacks (0)
The inclinometer on my SkyView has always shown nearly 1/2 ball out to the left, even when the fuselage was absolutely level.  After doing some research, it looked like the only way to adjust this was to shim the ADAHRS until in showed a centered ball with the fuselage level.  I used my shop crane to raise the tail to a level attitude and put my aircraft jacks under the jacking points on the wings to hold the plane rigid in the roll axis and to make minute adjustments to the lateral level of the plane.

After getting the longerons level, I used the wing jacks to adjust the lateral angle until my digital level showed 0.0 when placed across the seat back support.  I then zeroed the smaller digital level I have and placed it on the ADAHRS.  Sure enough, it showed about 0.5º angle with the left edge low (exactly as a left ball would indicate).

I removed the attachment screws holding down the ADAHRS and fabricated some small aluminum shims to fit under the left edge.  After some experimentation, I found that I needed a 0.048" thick shim to center the ball.  Now the ball is nearly perfectly centered.  I'm really surprised how sensitive this is.  Varying the shim thickness by just a few thousandths made a visible difference in the position of the ball.  Given that my digital angle is only accurate to 0.1º, it's really hard to get this accurate since I can now move the ball from one vertical mark to the other while still reading 0.0º on my digital level.

While I had the plane up in the level attitude, I was curious how much ground clearance I have  from the propeller tip.  This is the best case without any further gear compression.  If I were to bounce a wheel landing or raise the tail too far, this clearance could diminish significantly.

Greg also stopped by with his Windows laptop so we could make a few adjustments to my VP-X.  My Windows laptop is acting up (surprise, surprise), so I haven't been able to use it to connect.  We configured the Wig-Wag function, set up intermediate stops on the flaps (and configured the VP-X to use them instead of requiring me to hold the flap switch until they reach the position I want), and turned off current monitoring on the fuel pump.  I still need to adjust the flap airspeeds, but I forgot at the time.
Since I had the aft baggage wall out to adjust the ADAHRS, I took the opportunity to inspect all of the elevator linkages and relubricate them.  I then vacuumed out the entire fuselage.  Despite doing this thoroughly before first flight, there was an amazing amount of debris including a large number of small metal particles from the drilling operation when assembling the fuselage.  I have no idea where these particles were hiding when I earlier cleaned out the fuselage.  I then reinstalled the aft baggage wall and all of the interior.

Before reinstalling the cowl, I adjusted the prop RPM again to hopefully get pretty close to 2700 RPM.  I also wrapped the fuel flow sensor with some extra thick fire sleeve and some safety wire.  It's not the prettiest wrap job, but I'll redo it if works.

Spin Testing

| | Comments (0) | TrackBacks (0)
I snuck in a quick flight after work before it got dark.  I had four goals for the flight: check prop RPM, confirm the fire sleeve fixes the fuel flow at high altitudes, do some spin testing and confirm idle mixture.  On takeoff, the prop RPM came up to 2680, so I'm getting really close on the RPM.  One more small tweak should get me there.  I set up a cruise climb at 2500 RPM and about 120 kts and was still able to reach 10,000 ft by Hollister (less than 15 nm away).  This plane really wants to climb.  I continued the climb up to 12,500 ft and flew around a bit.  At first, the fuel flow was rock solid around 12 gph (still reading high) with none of the dips near 0 gph that I saw earlier at 10,000 ft.  After flying along for a few minutes though, I started getting occasional excursions up to around 16 gph, but they would generally come back down within a few seconds.  It looks like I still have some work to do here.

I descended back down to 8,000 ft near an airport so that I could do some spin testing.  I first slowed an did a couple of stalls to ensure the engine was leaned properly, then climbed back up to start spin testing.  The started with an incipent spin.  As the stall broke, I applied full left rudder and the plane rolled about 90º to the left.  I applied full right rudder and broke the stall, then rolled level and pulled out of the dive.  The recovery was nearly instantaneous, so I climbed back up and tried a second one with the intention of waiting a moment before recovery.  This time, the plane rolled well past 90º and the spin started to accelerate.  I applied full right rudder and was able to stop the spin after about 3/4 turn.  The recovery wasn't quite as instantaneous, but still very positive.  I continued to repeat the testing, letting the spin develop further each time.  The spin really accelerates after the first turn, but stabilizes pretty quickly after that.  It spins quite a bit faster than the Decathlon I have been flying, and fully developed spins seem to take nearly 1 full turn to recover, but recovery is always positive.  After doing a number of spins to the left, I did a handful of spins to the right with roughly the result.  All of these were power off with ailerons neutral.  I'll try power on and in/out spin ailerons on subsequent flights.  Overall, it's a fun plane to spin.  One surprise was that the SkyView really freaks out while spinning.  On one recovery, It showed me upside down after recovering to straight and level.  It fixed itself within just a few seconds (which is substantially better than most competing EFISes).

After landing with a nice warm engine, I taxied back to the hangar and did a slow mixture pull to confirm a small RPM rise.  There was no perceptible rise, so I think I overshot the correct mixture and I'm now running too lean.  I'll adjust it before the next flight.
I sanded down the glass I applied to the gear leg fairings so that the transition is smooth.  I'll apply a couple of skim coats of epoxy, then the outside of these is ready for priming.  I still need to do a little work on the inside for mounting before that though.

One of the builders who stopped by my Airplane Inspection Party back in January is the Editor-in-Chief of Kitplanes magazine, Paul Dye.  We spent some time chatting about my reason for choosing to hold an inspection party and my effort to build my plane in a very public way.  With so many eyes on my plane over the years (both in person and from around the world through the website), countless issues have been caught before they could have caused a problem.

In this month's issue of Kitplanes Magazine, Paul wrote up his visit in the editorial.  He was encouraging others to be as open and willing to take criticism as I was.  Far too many builders are afraid to let others examine their work for fear of exposing their mistakes.  Sadly, those are often the builders who most need another set of eyes on their projects.

Paul was very complimentary of my idea for a party: "Having a large number of experienced builders and opinionated pilots (are there any other kind?) drop in to dissect your many years of careful work takes intestinal fortitude of the highest order--and it shows a commitment to honest, open risk management that should be our standard in the Experimental world."

He was also very complimentary of my work: "Truth be told, it was one of the nicest building jobs I have seen in all my years of working on airplanes and being a tech counselor. When a builder takes the time to lay the flat EGT wires against each other in a perfectly rectangular bundle, you know he has paid attention to details."

I really hope others take Paul's message to heart.  We need to keep an eye out for each other; our lives are on the line.
I spent most of the weekend working (not on the plane), but I did get out a bit to meet up with a prospective RV builder down at Hollister.  He checked out Greg's RV-8 and my RV-7, and we talked quite a bit about what's involved in building.  Greg's son Nicholas then arrived with his transition instructor in another RV-7, so we decided to do some three ship formation work.

We started out with me flying lead since I have the least formation experience.  Greg was off my right wing and Nicholas was off my left.

We then switched positions and I flew off of Nicholas' right wing.  This was the first time I've ever flown this close to another plane, and it was a huge amount of work.  You literally can't take your eyes off the lead plane for a second or you could move substantially out of position.

Greg caught this shot of Nicholas and I in formation.

We then moved into a left echelon formation with Greg taking the lead and me in the middle.  I still have a ton to learn, but this was a blast.

About this Archive

This page is an archive of entries from March 2014 listed from newest to oldest.

February 2014 is the previous archive.

April 2014 is the next archive.

Find recent content on the main index or look in the archives to find all content.



Powered by Movable Type 4.1