VAF Forums

VAF Forums (
-   My RV Build Project (
-   -   RV-3B Dave's in Colorado (

David Paule 09-05-2018 08:54 PM

It's a Canoe
I bought a couple sawhorses and unfastened the fuselage from the steady, reliable Fry jig. Man, that’s a good tool. I lifted the fuselage on to the sawhorses, one by one. The RV-3B fuselage, in this state, is light in the tail and not so much heavy up at the firewall as it is awkward. It was not hard to do it myself,

Folks, I present the canoe.

While I really would have liked to put the main gear on and get it aligned and rigged, there just wasn’t enough room before hitting the jig. If I’d have known, I would have raised the fuselage up from the jig 2.00” at the start of assembly. That would have made some things considerably easier as well as permitting the gear to go on.

If I’d done it, put the gear on, I’d have removed it shortly, too, to facilitate other things. Other things like moving it around on the sawhorses and working around it, since my shop is small. I still think that’s the best time to do rig the main gear.

No, I’m not going to try to add 2” shims and reinstall the fuselage on the jig now. Thought about it and decided against it. You might say that canoe has paddled off.

Heck, yes, there's a backup photo right here.

And yes, the Fry jig is for sale. See the ad in the Classified section. It also fits an RV-4. In fact, it was intended for an RV-4.


RussellT 09-06-2018 08:40 AM

Good stuff Dave, a milestone to enjoy.

David Paule 09-16-2018 08:03 PM

The Fry jig is sold and gone. It will be used by a repeat RV-4 builder in the area.

The concept of installing the main landing gear legs is still on the table, since I could do it with the canoe upside down on the sawhorses. I asked some of the local builders and Eric Lanning volunteered to spray them with Azko Nobel primer for me. I included the tail spring too. Azko Nobel is one of the better primers and frankly, I?d have been better off to have used it myself throughout. It?s very tough, resistant to aircraft chemicals, and epoxy sticks to it. It was ideal for the landing gear legs.

Here?s Eric spraying the legs in his priming booth.

And here they are during the first hour, while they are still soft to the touch.

After that I was able to put them carefully into my car and drive them the 90 miles home.

At this time, the entire right-hand side is unriveted, as is the left side forward of the seat back. The glue is keeping things together, and a few clecos ensure than no edges lift and start unzipping. I was reluctant to do much interior installation or install the main gear until that riveting is done. It?s scheduled for next week. One of the things I could do now was hang the engine mount, which requires trimming the hot air exit fairing at the bottom of the firewall. That?s the titanium belly overlay I added on earlier. The titanium is one of the things needing riveting, though, so I had to be careful not to accidentally rip it loose.

Trimming the exit fairing was relatively simple: fit the engine mount, trim any of the fairing that interfered, repeating both steps as necessary. Took a few hours. Here?s the trimmed fairing.

However, the engine mount has a teensy problem. When it was built, the bottom cross-member was welded on somewhat lower than the engine mount bolt holes to the firewall. The plans show it, drawing 29, as being in line with those holes.

Compare to the actual engine mount. I?ve drawn a thin centerline between the mounting bolt holes.

And observe that the bottom cross tube is a bit forward of the firewall, too. The fairing, you?ll recollect, curves upward there. As a result, there?s interference along the entire bottom cross member as it rests on the fairing.

If I press the engine mount down to the correct position, the fairing bends down around 1/4 inch or so. And that doesn?t include the clearance that I?d need to prevent the mount from chafing on the fairing, due to the three dissimilar metals which would be in contact.

Turns out that this isn?t a new error. Here?s that tube on an RV-3 built 15 to 20 years ago.

Elsewhere on the fuselage, looking towards the top tailcone skin, I found that the aft four bulkheads, F-305, F-306, F-308 and F-309, do not line up well at all. The two middle ones are high and all four are slightly narrow. This isn?t too much of a surprise as they were fitted for fairness along the sides and bottom (the accessible areas whilst in the jig), getting the top be as it would. Still, it was slightly encouraging that the bottom 3? or so is the only areas too narrow.

Stay tuned for the next episode, folks.

Alternate hosting: here, here, here, here, here, and here.


David Paule 09-20-2018 07:19 PM

The canoe was only glued together when I took it off the jig, not yet riveted. Still to be riveted were the entire right-hand side skins, firewall to tail, the left-hand cockpit side skin, and the titanium belly overlay.

With the help of Glenn Potter, I got these done this week. Having the canoe off the jig and on sawhorses really helped as it was possible to rotate it to a convenient position.

The riveting came out decently enough. I'd been worried that I'd screw something up but it looks like we did okay.


David Paule 09-30-2018 08:27 PM

Glue Technique Tip -
As the total thickness builds up, you may need longer rivets. Buy them early and you will have them when you need them.

When that riveting work was done, I really truly had a canoe.

Back when I ordered the finish kit, I ordered a clear Todd?s canopy. More recently, I decided that a light tint would be better. Lance Logan now has canopies available. He took over Todd?s equipment. Lance Logan is the supplier to Spruce, so I had the option of ordering direct or through Spruce. I discussed it with Lance and decided to get it directly from him. The quality of the pre-purchase assistance was terrific.

I chose the 2515 Light Smoke, which blocks 88% of UVa and 99% of UVb. The other easy option was 2094 Gray. Lance sent a sample of both and the decision was pretty easy. The 2094 is pretty dark.

I sent a check and very shortly afterwards, a canopy arrived. And I do mean shortly.

First, here?s the earlier canopy that Todd?s sent, the clear one.

Next, here?s the new one that AeroCanopy sent.

There were some differences in how they were packaged and in a few other things, but all in all, I?m very satisfied with the new one.

Way back when I was assembling the firewall, what I should have done was take the engine mount and the firewall to a machinist friend and have him match the firewall holes to the engine mount. Then I could have used bushings to hold them to the 1/4? hardware on the jig.

But I didn?t. None of the firewall holes were drilled, of course, since this is an RV-3B, but they were center-punched to mark their location. That was nice of them. I determined that the middle two hole locations were somewhat close and that I could use them as a starting place to align and drill the outer four holes, which were, if not located properly, at least on the same firewall.

Drilling them from 1/4? to 3/8? was easy. For the two middle mount holes, I used a Harbor Freight Unibit clone and that left holes just slightly under 3/8?. I finished them with a reamer and those fit the engine mount. Since I lost the AN6-53 bolts that it?ll take to bolt them to the firewall, I bought some hardware store bolts for the job. No worries, they?ll be replaced. This won?t be permanent for a while yet anyway.

The outer four holes were visible through the engine mount?s attachment holes but that?s about all I could say for them. What to do?

I discussed it with the aforementioned machinist friend and decided to get some bushings.

First, I bushed the holes down to 1/4? and used a slightly undersized twist drill to walk the firewall holes closer to center, followed by the 1/4? reamer.

Second, I removed that bushing and the remaining bushing had a 5/16? bore. I did the same thing, using undersized twist drills to get close, and then the 5/16? reamer.

Last, I repeated the process with no bushings, ending up with 3/8? coaxial holes in the firewall and engine mount that the bolts would go into.

How was it? The first bolt with the bushings was very slow but not especially difficult. Learning curve, you know. The remaining bolts went quickly. At the conclusion, the engine mount was temporarily bolted to the fuselage.

Goody - a convenient place to grab when moving the fuselage around.

I rotated the fuselage upside down - geez, that looks familiar - and discovered that with a sawhorse under the upper longerons forward of the panel, I could level it by shifting one sawhorse end fore or aft. When level, I marked the location.

Upon trial fitting the main gear legs, guess what? they don?t go in their sockets. I had to decide whether to reduce the legs or open the sockets. Turns out that the gear legs are 1.375? diameter but the ends of the engine mount sockets are about 1.365?, or .010 smaller for the diameter. I thought that this was due to some excess cadmium plating that migrated into the sockets. The plating company assured me that they?d mask those holes. Looks like they gave me a headache instead. Later I discovered that the plating made a minuscule difference, and that the fit is what the factory sent. No worries, it's an RV-3B, and I signed up for this.

I bought a brake cylinder hone tool. I had never used one of these but heck, it?s a tool right? And I?m a proud member of a toll-using species, right? Plus I have an aptitude for mechanical things. How hard can this be?

Turns out that it wasn?t hard, just tedious. First, the tool.

I dipped the stones in alcohol and ran it on partial speed at about 40 psi, for about 30 seconds at a time. I didn?t want to burn out the stones but I really didn?t want to have the sockets get too loose. Here?s a photo of my doing the honing thing on the top of the right-hand socket. One thing to be careful about is to stay on the internal boss and neither get past it nor remove the tool while the stones are spinning.

It took about fifteen passes to get the top end of the right-hand landing gear strut past the bottom part of the socket. I was still about 1/4? shy from being able to insert the strut all the way. One more thing that I wanted to avoid was getting the strut stuck in the socket so that I couldn?t remove it or adjust it. I made a couple more passes with the hone and then put it all away. I ordered a replacement set of stones and they?re about a week out.

But no worries, there are plenty of things to work on. I started one of those.

Also at Todd's canopy, AeroCanopy canopy, Engine mount attached, hone tool, and actual honing.


David Paule 10-15-2018 11:02 AM

I?ve been sort of roaming around the fuselage, doing things that looked interesting.

I put together the pitch servo mount, following the Dynon RV-4 mount arrangement. It also holds the elevator bell crank. Here it is from the front, looking aft. I need to touch up the primer slightly and then the baggage floor skin fits under the forward end. I used nutplates under the mounting bolts.

It?s looking like Van is a lot bigger man than I am in several ways, and one of those ways is simple body length. The RV-3 has depressions built into the seat ribs for butt room. Neither of the two RV-3s that I?ve sat in demonstrated a need for that recess, so one of the alternatives I?m thinking about is revising the ribs to make them flat on top. I checked with Van?s Support and they say that?s acceptable but wouldn?t specify rivet spacing or type (I'll probably use blind rivets). The major load is the downward crash case, if you?re interested.

The red lines show where the new flange will be.

Another easy fit-up was the aft deck. The orange dots are magnets with mating ones underneath. That does a fine job holding the deck in place. I needed to locate and cut the openings for the stabilizer mounting bars and trim one side to fit, but the stock piece that came with the kit, fit pretty well.

Since the RV-3B isn?t a prepunched kit, there are no holes in the deck yet. Before locating the access holes that are in the plans, or the fastener holes, it seemed prudent to see if the stabilizer actually fits. Remember that the fuselage was built to fit the bulkheads and skins and more or less follow the plans. When people say that all RV-3s are different, that?s not an exaggeration. Let?s just say that there was some question - would the stabilizer fit at all?

The answer, I?m happy to report, is ?yes!?

The stabilizer came down from the rafters, sat in place for ten minutes, posed for some photos, and then went back up where it?s been safe for the last several years.

I marked the front spar and the front spar mounting angle positions and realized that the access holes in the deck will need adjustment. The rear-most hole in the deck allows for the elevator horn to extend down into the fuselage, so I marked the elevator hinge centerline position on the deck. I was going to mark where the aft hole needed to be when I realized that the horn is an RV-4 part. Okay, no problem, I?ll deal with that some other day. I?ll need to bring the stabilizer and both elevators down and assemble them and measure stuff.

With a small bit of epoxy curing somewhere else, it seemed like a great time to leave the project for the night.

Same photos located elsewhere here, here, here and here.


David Paule 11-04-2018 05:51 PM

Once the honing operation on the landing gear sockets was complete, I started to get the gear located. The honing gave me a mild press fit on both legs. I had to use lubricant to get the legs into the sockets. After they were in, I got a tip that on a few engine mounts, the internal sleeves were dislocated top to bottom so I removed them again to check that. Mine seem to be okay, although the internal inserts that were supposed to be 3? long were only an inch long. I?ll try that. And the legs went back in.

It seemed to me that the main thing to check was to get the legs so that the plane is sort of level in roll while on the ground. The toe in/out appears important, too. Camber isn?t under my control. The overall yaw position of the axels isn?t either; nor is the side to side position.

I leveled the plane in roll and marked the side of the fuselage for the sawhorse position. The top longeron, forward of the spar bulkhead, slopes downward. This lets me adjust the roll position simply by moving the sawhorse.

I clamped a 6? angle to the longerons and laid a level on it.

And marked the fuselage so I could repeat it.

Here?s proof that the gear legs fit in the sockets. They aren?t aligned yet.

The top ends can be adjusted +/- 3/16? in or out of the sockets to level the gear, according to the plans. Mine needed more than that.

For the toe in/out business, the manual says make them straight. My mentor says 1/2 degree toe-in. For mine, that appears to be about a .040 shim at the aft outboard end of the axel, which I added.

The photo shows that they are level - a bit later, without having touched anything, they were 1/10 degree off. So there?s some work to do yet.

I was uncomfortable with the edge distance internally with the gear legs in this position, and contacted the factory. After a few emails I talked to Sterling, who was very helpful. I finally decided to accept a small error in the roll alignment in exchange for better edge distance. Sterling sent me a photo of another RV that had much more error than what I should have and commented that it flew fine and that it was unnoticeable to the pilot.

The next step was to drill those holes. I don?t know about you, but drilling long holes in expensive, important bits of heat-treated steel isn?t something I think is trivial. My mentor sent me a very nice drill guide he?d made which apparently works fine for some of the other RVs and perhaps a Rocket. It didn?t fit my RV-3B because some of the engine mount tubes got in the way. I looked around the shop for scrap and after some thinking made this.

I?ll have to make another one of the thicker body parts of it for the RH side. The black line is the Allen wrench for the two set screws which secure the bushing. It fits into a hole I?d drilled to make a holder for it so I didn?t lose it. I would have put the bushing in with the bigger end towards the hole but then I?d have needed to unfasten and realign it with every drill in the sequence; each size drill has a separate bushing.

The two set screws are threaded into the 1/2? plate aluminum.

8-33 x 5/16 long Set screw size
5/64 Allen wrench size
#44 Retention hole for the Allen wrench

So there I was with a drill guide in place and some new drill bits to match the bushings on hand. All I had to do was get up the gumption to start.

Surprisingly, the drilling went well. I?ve only drilled one leg so far. I stopped at the letter N drill and with that one, only went part-way through, so that the B&S #2 reamer would have some material to seat on at the narrow end of the hole. Initially, reaming the hole went fine, but before too long the reamer wouldn?t cut. Apparently this is a known issue, early wear, for the ones from Spruce. It was worn out. I ordered a new one from ATS.

The plan is to follow that with a B&S #3 reamer to final size.

The photos might also be found here, here, here, here and here.


David Paule 11-11-2018 11:01 AM

Tool Woes
While I was waiting for the new reamer, I decided to work on the drill guide for the other side. Partway through cutting the 1/2" thick aluminum plate, the bandsaw broke. The thing was at least 30 years old and had always been cantankerous, so I gave it to a younger neighbor.

After some searching, I bought a Rikon 10-326 bandsaw. I expect it'll be an upgrade but we'll see. I'm assembling it this weekend, so it's too soon for a tool review.


David Paule 11-30-2018 03:31 PM

Engine mount/landing gear to the machinist -
Since I just couldn?t cut a tapered hole in the landing gear mount, I took it to my machinist friend. But I?d gotten far enough so that I could use the tapered pin as a positioning tool. He?ll get to it one of these days. For him, it?s a hobby. He?s a physicist and his company is very busy these days.

ADAHRS bracket -
With the engine mount off, I flipped the canoe and looked around for the next thing to work on. Seems as if that might be the ADAHRS mount. I made a sketch showing where one could go on an RV-3B and found that the place I preferred was just under the top of the F-308 bulkhead.

These brackets let the ADAHRS slide in. I?ll put stops in the slides at the aft end and use nylon screws or tie-wraps or something to secure the front end. They can?t actually get riveted to the bulkhead until the bulkhead is a least clecoed to the turtledeck to skin, and all I?ve done in that direction is screw a wooden frame to the back of the F-308 bulkhead.

Panel -
I laid out the panel and cut it out. The panel, so far a virgin one, fits the fuselage. My fuselage is slightly wider than the plans dimensions, so I had to adjust the width of all the points by a bit over 1%. Not much but it added up. You might think that this is something that I should have laid out in CAD and had water-jet cut, and that would have worked. But this was quicker. From a piece of bent aluminum to drawn in Sharpie to cut out, edges finished and hanging on the longerons only took three hours. I must say that new bandsaw sure is a nice tool.

I made it 1/2? deeper, top to bottom, than the plans call for. The excess height is below the longeron, with the curve above the longeron per the plans. It eases the panel layouts a bit. As I?ve hinted, I?m planning on a Dynon Skyview system with one screen and a few peripheral analog devices. I?ve played with cut-out photos of the various odds and ends that will go on. But I won?t freeze the layout and cut the holes until I get closer. Things change. The panel currently has no holes for the gadgets.

The current plan for the panel includes:

Dynon Skyview, probably the 7?,
11? iPad Pro, unless I use the 10? Skyview,
Trig com,
2 1/4? asi,
2 1/4? 8-day clock,
Slip indicator,
2 ignition switches,
Alternator/master switch,
Fuel pump switch,
Running light/landing light switch,
Nav/strobe light switch,
Perhaps the autopilot disconnect switch.

That?s a lot of stuff to jam into an RV-3B panel, so don?t expect beauty.

Since the stock top of the stick doesn?t poke up very high, there?s gobs of clearance. I measured 3.7? from the top of the stick to the bottom of this panel. My stick might be a slight bit high, and the bottom of the panel is 1/2" low.

The RV-3B panel has a flange around the back side. After some consideration I decided that I could use the F-803CPP strip that Randy Lervold recommended. This is an RV-8 part for the purpose. I?m using a bucking bar to force the strip in place on the panel, and checking into the table to hold it securely. Here?s SK-40 showing how. Got to be careful doing this since some bucking bars have rounded edges that the panel would nest into; a sharp edge is needed for a bucking bar to work.

The center piece of the strip is wider than than the other tabs. I had to shrink that to get it to bend slightly to confirm.

The main issue with the F-803CPP is that the notches are cut so far into the smooth flange that there?s precious little left for the attachment screws. On the RV-8, it uses rivets. The RV-3B uses #8 screws, which need more edge distance. I decided that if I was careful, I could place them away from the notches, at the tabs. It?ll mean 2? spacing instead of 2.5?, no problem there, I think.

The alternate hosted photos are:

ADAHRS Bracket

Panel on Fuselage


Panel flange 1

Panel flange 2


gasman 11-30-2018 08:48 PM


Originally Posted by David Paule (Post 1306058)
Engine mount/landing gear to the machinist -
Since I just couldn?t cut a tapered hole in the landing gear mount, I took it to my machinist friend.

I made it 1/2? deeper, top to bottom, than the plans call for. The excess height is below the longeron, with the curve above the longeron per the plans. It eases the panel layouts a bit. As I?ve hinted, I?m planning on a Dynon Skyview system with one screen and a few peripheral analog devices. I?ve played with cut-out photos of the various odds and ends that will go on. But I won?t freeze the layout and cut the holes until I get closer. Things change. The panel currently has no holes for the gadgets.

The main issue with the F-803CPP is that the notches are cut so far into the smooth flange that there?s precious little left for the attachment screws. On the RV-8, it uses rivets. The RV-3B uses #8 screws, which need more edge distance. I decided that if I was careful, I could place them away from the notches, at the tabs. It?ll mean 2? spacing instead of 2.5?, no problem there, I think.


That's why it is best to build these kind of parts from scratch. I don't have my plans in front of me, but I don't think this part meets the spec's on the RV3 plans.

All times are GMT -6. The time now is 09:20 AM.