Filing the Edges and Corners

Spent about an hour or so tonight filing the edges and the corners of the cutout area of the LE. It was finally warm enough in the garage again to work on the plane. This was basically a repeat of the same effort I used to fabricate the new inspection panel cutout from long ago. The only difference is that the edges are longer and are on a curved surface instead of a straight one. I used the same tools that I used for the access panel job to file down the metal edges of this cutout. I used a standard flat file, a curved rasp, a round rasp, a small jewelers curved file, my small metal ruler to check for edge straightness, and the SafeAir inspection plate that already has the 1/2 inch-radiused corners so that I can use it to test the shape of the corners and adjoining straight edges on either side for a true square fit. Here is the pic of all the tools as I started to file the edges on the top of the LE skin:

I had to once again reposition and re-clamp everything several times so that each edge was supported properly as I started filing. I had to position the edges of the cut section just off the end of the table allow enough clearance for the files moving back and forth, and allow enough room to reposition them as needed. Every time I work with the files I am amazed at how precise you can be with them. The key is to have patience, work slowly, only remove small amounts of material at a time, and CONSTANTLY check your progress against your drawn lines to make sure you are not going to far and removing too much material, which would be devastating at this point. 

I started with the edges and corners on the top side of the LE skin. I worked on those first, so that I could get those shaped properly. Here you can see the "fit" of the inspection plate on the unfinished edges of the LE skin before I started removing material with the files. the edges and the corners all had extra material that needed to be removed. I filed a bit and then kept re-fitting the inspection plate until I was satisfied that both edges were straight, and the 1/2 inch radiused corner was also fitting correctly. Both of the following pics show that there is still some filing to do.

I used the curved rasp to remove the material in each corner by slowly removing the metal and checking the curved line drawn on the skin. The goal was to just meet the sharpee line as well as each edge on either side of the radiused corner, and repeatedly check the fit with the inspection plate until I was satisfied with the fit.

After a few minutes, this side was finally starting to look pretty good:

I have this tendency to try to make everything fit almost perfectly. Some folks would say I have OCD in that regard, but the truth is that I enjoy the craftsmanship part of this project so much that it gives me sense of accomplishment when I am able to form metal with a degree of precision that I used to think could only be accomplished with wood. The reason why it is so important for this to fit together well is that any gaps around the edges will cause the airflow to be disturbed, adding drag to the airplane, and this will just slow me down or rob me of extra performance that could otherwise save me some fuel or allow me to fly faster or to reduce my stall speed considerably. This is where airplane building separates itself from other types of building. If it ain't "perfect," it will most likely deprive you of something, and it may even put your life at risk.

One problem with a "perfect" fit is that I am not allowing much room for primer and paint to be added to the parts later on, so it is very likely that the edges on some of these removable or tight-fitting parts may have to be "reduced" just enough to allow for several mills of that stuff to be applied and still fit together nicely. I'll deal with that later.

Here is the top after most of it has been filed to my satisfaction:

Then I flipped it over and started working on the bottom half. I was able to file the edges out to the tip of the bend so that I would have a good enough reference to match it against line from the other side as much as possible.

Here is a shot of how I would apply the flat file to the edges

Getting closer to completing the bottom corners with the curved rasp

And the last two pics I took of each bottom corner after fitting the access panel to each of them and finally being satisfied with the fit.

I was a little concerned that I may have messed up one of the edges by cutting too close to the edge with the cutoff wheel, but after checking it closely I think everything worked out fine. I used the metal rulers to check the straightness of each of the edges, and even though some optical illusions started messing with my eyes to make me think that the edge was not straight, the ruler sitting flush up against it was proof enough that I should stop filing and walk away before I really messed it up.

SO overall this did not take very long to do - certainly not as long as I though I would, and I am ready for the next step which is to sand all the file marks off of the edges, especially in each of the corners. If I do not manage to remove all the burs and potential stress risers from these edges, they will most certainly start to crack at some point, so proper deburring is an absolute must.  I will sand each corner and each edge until ALL the file marks are gone and the corners are smooth. There are a lot of burs from the Dremel cuts on the inside of the skin, so I will need to make sure that those are removed as well.

So it was a good session tonight with no surprises or issues. Now I'm just working down the remaining list of To-do's I posted previously to get this finished as soon as possible. The part of this that has me nervous is when I fabricate the removable section and start drilling screw holes in that and then start attaching it to the subskin and checking to see f that part will lay completely flush with the rest of the LE skin. Personally I don't think it will for a number of reasons, but I have been pleasantly surprised by my workman before - may be I will again. I sure hope so.

THE BIG CUT!!!!

Well, it almost took two weeks for everything to be "optimal," but wouldn't you know it, on Thanksgiving day, conditions finally allowed me to make the big cut. This is the cut to remove the section of the leading edge that will be replaced by a removable piece of metal that will contain the nacelles for my special project, as well as a plain sheet of aluminum that emulates the section that I am removing. This way I can change the type of LE that I want on the airplane depending on my needs and desires for any given flight.

I started with the bottom most inboard edge cutline, and then cut the outboard edge, moving the Dremel tool from the right to the left, after starting the cut from each of the drilled holes in each corner. the skin was clamped to ensure that it would NOT move at all during the cut, by placing the cutline as close as possible to the edge of the board/table, with just enough clearance for the Dremel cutting wheel to make the cut. The goal was to try to leave no more than 1/16th of an inch of extra material, which has to be filed and sanded away in the final preparation steps for the LE. So you have to take into account the thickness of the cutting wheel itself, and then move it just inboard of the cut line to leave just enough excess for final filing and sanding.

The forward part of the LE was set as closely as possible to the other edge of the table prior to clamping, since I anted to make certain that each edge cut would extend as far forward to the halfway point of the bend in the LE as possible, so that when I flipped the LE over to perform the edge cuts on the other side of the LE, I could easily see the previously cut lines from the other side and align the cutting wheel to meet them so that the entire cut line matches from side to side as much as feasibly possible.

Here is the most inboard cut on the bottom:

And then I unclamped the entire LE, and shifted it outward on the table until the most inboard cutline was just beyond the edge of the table, and re-clamped everything in place again. Then completed the cut in this side in the same manner as the other one:

And here are the two cuts shown together, just after I repositioned the LE again on the table for the rear-most cutline:

And here is the bottom after making the rear-most cut on the bottom of the LE:

Each time I repositioned the LE and re-clamped it, I had to decide how best to re-hang the Dremel tool from the shaft of my bar clamps, and then made a dry run without the tool running to make sure that the flex shaft, electrical cord, etc. was free from any binding or interference that might cause me to completely ruin the LE with a rogue cut. Once I was satisfied that the tool was properly positioned for the cut, I turned it on, started from right to left along the tangent line of one of the predrilled holes, and used two hands on the tool at ALL times. I would periodically stop and remove the tool from the LE if I felt that the cut line was drifting in the wrong direction, or getting too close the pre-drawn Sharpee line. I would reset the cutting wheel as necessary to ensure that each cut stayed within the boundary of each drawn cut line.

Next step was to remove everything, and flip the LE over to perform the same cuts on the top side. Here is the sequence of pics for each of those cuts. I followed the exact same procedure as I did for each cut on the bottom side - doing each of the edges first, and then the final cut on the top rear-most edge.

The last pic above shows the chunk of the LE completely removed and sitting on the floor next to the leg of the saw horse. This last part of the rear cut that would completely free the entire piece of metal was the most worrisome part of the cut, because as that cut moved closer and closer to the final edge, I forgot to account for the fact that the entire part will try to bend itself away as the area holding it in place weakens and the gravity of the part being removed starts to fall. This left a small sliver of metal that had not been cut cleanly that I had to carefully remove with the cutting wheel to completely remove the section from the rest of the LE. Fortunately, EVERY cut went more or less as planned, and the part finally fell cleanly away from the rest of the LE. 

Now the bottom half of that bay on the LE looks like a very holey piece of Swiss cheese. The full coffee can of clecoes was very important for this entire process as well, since no clamps could easily be placed in the area just behind each cut, the only thing I could do was find something heavy enough to secure the area of the LE skin just behind each cut to keep it as stationary as possible. The heavy can of clecoes worked extremely well for that. SO with each cut I would place the can as close as possible to the LE skin area, just behind each cut area. This provided enough resistance to the skin to prevent it from rising up or buckling while the Dremel was making the cut. So no bowing or bending of the metal was allowed during each cut. Here are some final pics showing the state of the LE skin along the cut lines, and piece of skin that was removed:

The next steps are:

1. File down the newly cut edges of the LE so that they are straight and smooth

2. Finish drilling the remaining #40 nut plate attach rivet holes in the subskin.

3. Reassemble the LE for hopefully the last time and mount it securely back onto the Wing spar.

4. Make the file folder template and mark the positions of the #30 drilled screw holes in the subskin onto the file folder template

5. Remove the file folder template and use it to draw the boundary lines of the new removable section, and match drill the screw hole locations on the top and bottom sides.

6. Cut and trim the new .025 removable LE section and start clecoing each #30 screw hole of the subskin and the new part, starting from the bottom of the LE.

7. Check the alignment of the new part against the edges of the cutout in the LE and make absolutely certain they are correctly aligned along the bottom edges, with no binding, interference, or bowing, etc.  

8. Continue wrapping/forming the removable section of metal around the bend of the LE skin and subskin, maintaining the contour of the rounded LE.

9. Check the alignment of the screw holes and edges along the top side of the LE, and clecoe all #30 nut plate screw holes, working from the most forward holes along each edge of the removable skin to most rearward holes

10. Make certain that the contour of the front of the LE bend on the new removable section matches the contour of the rest of the LE skin. If not, remove the part, determine if screw hole locations or more trimming of the part is needed, make a new template, and start the clecoing process over from the bottom of the LE once again.

11. IF everything aligns with the rest of the LE skin properly, remove the removeable section and make another section on .025 aluminum sheet to make a second part, by using the contour of he edges of the current removable section and the screw hole locations. 

12. replace the original removable section back on the subskin and LE assembly, and match drill all screw holes with a #19 drill bit.

12. Remove the LE from the wing and the removable section and subskin for hopefully the last time.

13. Debur all remaining screw holes in the subskin and removable LE part, then dimple them with a #8 dimple die for a #8 screw.

14. Reassemble the standard LE parts and the subskin in the cradle, and follow Vans instructions for assembling the LE for the final time. Then place the assembly on the wing and permanently attach to to the wing.

So that's it. Fourteen more steps should result in a completed LE on the left wing with the mod that I had originally intended on making so many years ago. 

Tank Attach Nutplates Riveted, LE Skin Cut Preparation

I was able to finish riveting the remaining tank attach nut plates on the subskin. T do so for the forward-most holes I had to change my approach as well as my tool setup from what is was previously. My support blocks were too big to clamp on the inside of the subskin near the bend, so I had to basically reverse the approach.  I moved the support blocks to the outside of the skin, re-clamped the skin to the support blocks from the opposite side from before, switched my flat sets in the squeezer so that the wider set was on the ram and the smaller set was in the stationary part of the yoke, and I had to feed the skin in a slightly different way to maintain tool clearance from the clamps and ensure proper operation of the squeezer.

Here is the pic after I had finished setting all the remaining rivets. Call it being lazy or whatever, but once I had changed the setup to allow me to squeeze the most forward rivets - 2 nutplates on the bottom side and two more on the top, I decided that instead of switching everything back to the previous clamping and setting method, I would just keep going with the new setup. Why is this significant? Well, the change in the setup also meant that instead of pulling the stationary part of the yoke against the manufactured head of the rivet and the skin, and allowing the ram of the squeezer - the moving part - to smash the rivet tail, the process is reversed. So I had to place the moving ram over the manufactured head of each rivet, and try to keep it centered on the rivet head as it moved upward to close the gap with the stationary part of the yoke on the other side.

SO the difference is that when configured one way, you are holding the squeezer stationary against the rivet while the ram moves and sets the rivet. For the other configuration, the ram is moving upward and you have to try to keep the rivet set over the center of the rivet as it moves. THis is how I have messed up so many rivets in the past that I cannot even begin to count. Trying to keep a flat rivet set stationary over a small rivet head while the ram is moving is counter-intuitive, but the problems is that, as is the case here, sometimes you do not have much of choice as to how things need to be configured in order to get the job done. So you have to learn how to rivet using a lot of different techniques and setups - that's just the way that it is. In the above pic you can see how the ram will push on the skin and the rivet head as it moves up, so you end up developing this sense of allowing the ram to move up while the squeezer moves backward a bit, and, if everything stays in alignment, i.e.the rivet set on the ram does not slip off the center of the rivet head, the other flat set will eventually make contact with rivet stem and smash it to the proper size.

Fortunately I was able to set all the remaining rivets that way, and they all turned out fantastic! So I did not even have to drill out or reset any rivets. All 14 nutplates for the tank attach edge of the subskin were set successfully and correctly, and I was very pleased with that. it is not often that I engage in a rivet setting session and I do not have drilled our any rivets or fix a problem of some sort. I was glad to get that behind. me. Here are the finished nut plates - all riveted on.

With that done I decided that the next major task was to drill the corner holes in the LE skin to establish each corner radius as closely as possible and provide the relief holes for the straight cuts that I need to make to remove that portion of the LE skin that will now be "removable."

The steps were:

1. Clear off my work bench to allow the LE skin to sit on the top surface

2. DEtermine how high above the work bench I would have to stack other drill boards to account for the depth of the unibit to create a 1 inch wide hole with the necessary 1/2 inch radius that I designed each corner to have.

3. Charge my batteries for my electric cordless drill, which is once again the tool that ALWAYS use when drilling holes in metal with a unibit. NEVER use a high speed air drill with a unibit. I learned this lesson a very long time ago and have heeded it ever since. Air drills turn way to fast, and unibits cut much too quickly and efficiently, and the drills lack sufficient torque, to ensure that the hole will be drilled properly without distortion or without accidentally over-sizing the hole.

4. Install the correct unibit with a 1 inch diameter step in the drill

5. Stack the required number of drill boards on the work bench uner the Le skin to ensure that the unibit will not drill into the work bench top. IN my case I had to ensure that I had at least 2 inches of drill board under the LE skin.

6. Clamp the LE skin over the drill boards. Again, the idea here is that the unibit needs to drill into the thin flimsy metal of the LE skin and then into the wood underneath to keep from marring the hole in the metal and causing substantial burs. This is why you need a high torque drill - because you have to drill through so many layers or thickness of wood, in addition to drilling the hole in the metal.

7. Measure 1/2 inch x 1/2 inch from the cutlines on the rear and sides of the LE skin, mark that location and use the punch to mark that location. This is to be the center point of where the unibit must start drilling in order to achieve the 1/2 inch radius that matches the curvature of the skin that I had already drawn on the surface. I started out  by drilling a #40 pilot hole, so the 1/4 inch wide center point of the unibit would be able to locate a good center without wandering. Then I just inserted the unibit tip in that hole and started drilling slowly to ensure that it was entered correctly. I did NOT drill the hole up to 1/4 inch before using the unibit, as I did not consider that necessary. 

7. Drill the two corner holes on one side of the skin, unclamp, flip the skin over, reclamp, and repeat the steps to locate the drill hole and drill the two holes on the other side of the skin.

Here I am beginning to stack the required drill boards on the bench after determining that I will ned at least 2 inches of wood to prevent drilling into my work bench top.

I decided to use my Dremel tool and an EZ lock reinforced cutoff wheel to make the necessary straight cuts, and for this job I wanted a fresh wheel on the tool, so I changed out the old one which was about down to the shaft, with ab rand new disk:

Another blurry pic, but this is my harbor freight unibit that has a 1 inch wide step in it that is about 2/3 up from the tip. You can just see on the right side where I marked the step just above the 1 inch step with a sharpee so that I could tell which step I was on as the tool advanced through each step in the drill bit. I frequently stopped to clear debris from the drill hole and to check the progress of each step against the lines I had drawn on the LE skin, just as I had one before when I cut out the hole for the new access panel so long ago.

Next are the two holes on the bottom skin after I had drilled as close to the pre-marked lines on the skin as I dared. In fact I got a little too close on the first one.

And here is my clamping setup. I ended up using a 3/4 inch particle board, 2 2 x4s, and a 2x6 board on top of my work bench to provide the depth that I needed to keep from drilling into my work bench. Used a very haeavy can of clecoes and my clamps to secure the skin on the workbench. The top of the skin draped over the front of the work bench just fine while the bottom skin layed relatively flat. This setup worked great.

And here is the top of the LE skin after flipping it over. Of course when you do this the side of the skin that you are drilling on is reversed, so you have to reposition drill boards and clamps accordingly.

And finally a pic of all four corner holes drilled in the LE skin. I had hoped to also perform the big cut on the remove the section of the Le skin, but I was exhausted and needed to get some sleep first. Unfortunately the weather is going to change and get cold and snowy for the next two days, and I will NOT perform the cuts of the LE skin unless EVERYTHING is optimal. I have come too far to screw this up now, so I am not taking any chances. SO the best I could do for now was setup the LE sin for the next step, which is to FINALLY cut out the section that will ultimately be removable.

Here is the sequence of pics showing how I did that:

Subskin Nutplates Part 3

Here is a pic after dimpling the #40 rivet holes for each nut plate with my pneumatic squeezer, using the reduced radius female die and the standard sized #40 male die. Setting dimples was pretty easy.

The next pic is blurry but shows the process and positioning of the nut plate on the dies for setting the dimples on each nut plate flange. This also went quickly. There are 14 nut plates for the fuel tank attach portion of the subskin.

The nut plates sat a little proud on the dimples in the subskin, but they actually riveted on quite nicely I think.

With the required dimples completed, everything was clecoed in place:

And here was the creative clamping structure for this round:

I used the pneumatic squeezer to set these rivets. I switched out the dimples dies on the same flange yoke for flat sets, one if which was the same reduced radius die as the female dimple die. I always get nervous when I have to use such a small radius die for setting rivets, especially when the smaller one is the one that must be used to smash the rivet tail down. Everything has to be centered pretty well, ad there is not a lot of room for error. it is very easy to slip off target and cause the rivet to become deformed. Fortunately all the rivets I set were completed successfully, with out the need to drill any of them out, so I was quite pleased with that.

The clamping structure was set up to allow me to pull on the wider top flat set to apply pressure to force the manufactured head of the rivet against the dimple in the skin, and close any gaps between the skin and the nut plate flange on the other side as shown below:

The trick for alignment is to place the 1/3 inch wide set just below the rivet head and slide it up over the head making sure that it is centered as much as possible. Then you make certain that the flat set is truly flat against the rivet head and the skin. One of the biggest problems with this set on an air squeezer is trying to ensure that the set is truly flat. I found that you have to take into account any angle to the skin that might exist ir is created when you apply pressure to the skin and the rivet head. once you know that the rivet set is flat and centered over the rivet, then you are ready to engage the trigger and set the rivet. I always try to position everything so that I am pulling on the rivet head as the plunger goes up from the other side to set the rivet stem.

One other trick that use religiously is to set the air pressure going the squeezer to 75-80 psi max. The instructions with the tool always say to use 90 psi, but I figured a long time ago that this causes the ram to actuate much too quickly and can cause the rivet sets to move out of position just enough to ruin the rivet or the parts. I found that when I reduced the air pressure it was much smoother and less jerky, and the sets tend to stay in position better and you can watch the ram as it comes up to smash the rivet and make sure that it s positioned correctly before you commit to the full smash. The ram still has plenty of force to set the rivet, even though it moves a little slower.

First the rivet goes in the hole:

One down and one to go.....

And one nut plate riveted, 13 more to go:

This also turned into a creative clamping exercise because after each rivet was set the clamps had to be repositions and the part moved into position to allow enough clearance for the squeezer to set the next rivet. I was able to do 5 of the rivets and had to stop when  I got to the two most forward ones:

It was too cold tonight to set the remaining rivets, but I'll get this done as soon as possible. Unfortunately it looks like when I get the tank attach nut plates riveted I am going to need to cut the skin on the LE and then reassemble everything one last time before final riveting of the entire LE so that I can make the file folder template to mark the hole locations for the removable LE skin, and THEN I can drill the remaining nut plate holes to final size, debur the holes, dimple them, and rivet them to the subskin. So the to-do list is getting shorter, but it will still take a lot of time to get to the finish point.