30 November 2014

Fuselage: Mid fuselage lower structure riveting.

Work on riveting the non-skin related items on the mid fuse lower structure progresses.

There are a couple of ribs, F-01417-L/R, that get F-14100 routing angles (for future wiring, per a recent builder's query to Van's).  It's easy to squeeze the upper and lower rivets, but the middle rivet is just a bit too long for even my 4" deep yoke.  I don't have a rivet set for AN470AD3 rivets, but I do have a squeezer die and a strategically drilled 0.187" hole in a footed bucking bar.  Couple of wacks with the back rivet set and all's well.  Note that so far only 8 of the 10 routing angles are used.  I see no other place in the remaining fuselage plans where the last 2 are used.

The F-00016-L/R step attach weldments represent an improvement over previous models:  The steps are not riveted to the frame but rather are bolted on and removable.  The weldments get riveted to the baggage ribs first.  These rivets are a bit difficult to get to.  The plans call for Cherry blind rivets in the recessed spots and gives the options of Cherrys (ies?) in all spots.

Turns out that they were all accessible with my two tungsten bucking bars (a.k.a., Mr. T) except for the lower, forward-most one.  I have a very small footed bucking bar that could reach it, but it lacked mass and it was clear that the AD4 rivet was reaching incipient work-hardening, so I removed it and went with the Cherry.  This tungsten bucking bar would have made simple work of that one rivet, but for $230, I'll stick with the Cherry.

Using the angled Mr. T to hit the upper inset rivet (left).  And the rectangular Mr. T for the others (right).  The gray primer represents spot priming of the countersunk nutplate holes and final drilled 1/4" bolt holes.

The one Cherry is evident (spot priming is visible and was necessary after removing attempted AD4).

Left weldment's shop heads (left).  Right weldment's shop head (right).  Some spot priming will be necessary where the bucking bars ate the powder coating off the weldments.

The side frame assemblies get riveted to the side frame brackets with the step attach weldments and doublers.  The plans call for using LP4-3 pull rivets in 3 places.

However, I was able to squeeze all 5 rivets on both assemblies using my 4" deep yoke without issue.  Taping the edges of the yoke helps prevent scratching the parts.

The step attach assemblies then get mated to the baggage ribs at the weldments and doublers.

These are tough rivets to set.  In fact, these are the ones that seem most reasonable as Cherry, yet aren't suggested as such in the plans. I opted for squeezing them with my 4" deep yoke.  Five of the six turned out well enough.  One of them was clenched over (not shown).  Ordinarily, I would have replaced it, but given the poor access, I figured I'd mess up the hole and decided to leave it.

Next, the step attach assemblies get riveted to the F-01425-L/R baggage ribs.  Again, tight access.  The plans call for 2 of the rivets on each weldment to be Cherry.  However, I was able to squeeze all of them as AD4s using my longeron yoke.

Completed step attach assemblies (sans nutplates on weldments).

Completed crotch strap assembly.

I'm building a tri-gear so my gear will attach to the forward bulkhead (rather than the firewall as for the tail-dragger).  Thus, the plans require cutting the rectangular holes in the 0.040" forward bottom skins to match those of the 0.025" center bottom skin.  I used a combination of my Dremel with a reinforced cut-off wheel, nibbler, files and sand paper to create those two holes.  The holes aren't perfect, but no one will see them and they'll function just fine as-is.  They were rounded and smoothed down to mitigate cracking.  The Dremel got away from me on the right skin necessitating a notching and subsequent smoothing to reduce skin stress.

29 November 2014

Fuselage: Firewall. Priming.

Nothing exciting here.  Firewall parts were primed.  Had a few touch-ups of mid lower structure parts too.

Setting up HVLP gun.


27 November 2014

Fuselage: Forward mid fuselage bulkheads riveting.

Got nearly all of the forward mid fuselage bulkheads riveted.  Only needed to fashion a way to support the work so I can drive all the rivets solo with my even fancier new tungsten bucking bar (even better than my first one), a.k.a. "Mr. T.".

Left shows good shop heads all around.  Right is the manner in which the bucking bar was used.

Aft center section bulkhead.

Bulkhead bottom channel.

Cover ribs ready to go.

For the forward center section bulkheads there are eight cover ribs.  I wanted to place the manufactured heads on the ribs so I didn't end up with gaps as I did with the wing ribs (before I knew better) and I didn't want to use the evil offset rivet set.  I found that as long as each hole is cleco'd (can use 5/32", #21 clecos for the bolt holes), the flange will sit flat when you deflect the rib to fit the rivet gun with a short straight set.  Left is an example of the holes cleco'd (the uppermost #30 hole is cleco'd from the aft side since it cannot be placed on the forward side, so there's no need to cleco the bolt hole above until later when that uppermost #30 hole is bucked up).  Right shows how much deflection the rib undergoes when placing the gun.

Because of the upper flange impeding access, I decided the only way to rivet the topmost hole was with the manufactured head on the aft side.  I used one of my tungsten bucking bars, Mr. T, in the manner shown below to buck a nice straight shop head.

Here's what each what the rivets look like for the ribs on both sides.

The missing ribs, #3 and #5, are due to bag 3076 being on back order. That bag has the bearings for the bearing bracket assemblies which sits on the missing ribs. Once I get the bearings, I can complete and attach the bearing bracket assemblies to those two missing ribs and rivet them to the spar.

25 November 2014

Fuselage: Firewall. Dimpling and countersinking.

The stainless steel firewall has two cutouts with AD3-sized rivet holes along their perimeters.  The plans specifically state not to dimple those 8 rivet holes on each side.  The callout only references the right side, though the text clearly states both sides (also, you're going to want a small diameter 1/8" female dimple die to avoid distorting the curved flanges on the firewall parts).

That didn't stop me from dimpling the left side anyway.  Whoops.

I skipped ahead in the plans to see what sort of grief and/or ca$h outlay I just caused myself.  Turns out that the vents go there (see page 28-10) and their parts can be easy dimpled or countersunk to accept my special dimples.  So all's well.

The firewall includes a fairly beefy frame as part of supporting the fan up front.  Each one of the angles that compose the frame has to be countersunk where they meet the dimpled firewall.  Here's a layup to find the holes that do not get countersunk.  Covering those holes with tape helps avoid errors.

Turns out there are 196 holes (subject to counting error) that need to be countersink, to accept a 0.025" thick stainless part, some 85%+ of which are for AD4 rivets.  That's a lot of work.  And though the angles are thick, the holes still come down to a knife edge when countersunk, so be sure to provide additional thickness to retain your countersink bit's pilot to avoid elongating holes.  I used a thick aluminum angle with various holes drilled in it so I could keep my work from moving around.

And here are the 196-some countersinks, completed over three days in 2 hour sessions.  Glad that's done.

23 November 2014

Fuselage: Mid fuselage lower structure. Dimpling center bottom skin.

After seeing this thread in VAF, I thought I'd share my approach in the hopes that it's useful.  

The F-01484 center bottom skin is quite large, as it represents one of the defining features of the RV-14:  Cabin width.  So dimpling the holes in the middle of this skin can be a challenge when the throat depth of the DRDT-2 is 22 5/16".  The hardest hole to reach is the one shown below.   The shortest distance to that middle hole is close to 22 9/16".  That's 1/4" too big for the DRDT-2.  Or is it?

Below is my skin following the dimpling of all holes except the center one.  NoteBreak the aft edge of the skin upwards, as called out on page 26-14, Figure 1 (see above), before dimpling is called for on page 26-11, step 6.  I didn't.  I wish I did.

To get to that middle hole, the distance from the edge of the skin to the hole must be somehow reduced by 1/4".  I found I was able to do this by aligning the "notch" in the skin with the DRDT-2's upper arm stand.  With the skin at this angle, this distance is reduced as much as it can be prior to bending the skin.  Here's a verbose description...

Keeping in mind that the distance from the base of the lower dimple die to the bottom of the DRDT-2 upper arm is exactly 3 1/4", then any deflection of the skin upwards in the throat using that space will bring that center hole further into the throat.  So, I pushed the edge of the skin into that upper corner.  It's almost enough to align the hole over the die, but not quite.  Finally, being careful not to permanently distort the skin, I pushed the skin firmly into the DRDT-2 (yes, the skin makes a sort of "pop" sound when you do this) and negotiated the male dimple die into the hole.  Confirming that the skin was flat around the die, the dimple was then applied.  

Following priming, here is the skin with the male dimple die in that hole.  You'll note that the skin is not being held by myself or any other additional artificial means besides the platforms, so with care, this can be done by one person (unfortunately, that's usually me).

A close-up of the approach.  Note the location of the skin's notch in the upper right of the image.  It aligns with the upper arm stand and is pushed upwards toward the upper arm.  Thus, the skin is not laying flat (except around the dimple die, which is important).

Finally, I looked at the Cleaveland and Avery C-frames, and they both claim a 22" throat depth.  So does the DRDT-2, but as I stated above, I found it's actually 22 5/16".  My guess is the technique described above would work for the other C-frames, but I can't be certain.