Level 3 Build Thread -- Velociraptor REX

[rocdoc]

So I’ve decided to go after my Level 3 certification. I certified Level 2 on a Binder Design Velociraptor. It is a relatively complex design by Mike Fisher with split fins and a set of upper “talon” fins. It is also a good flyer. I therefore decided to upscale it by a factor of 2 for my Level 3 project. The result is an 8” diameter, 15 foot tall, rocket with a 98mm motor mount able to accommodate any commercial 98mm motor. I plan to call it the “Velociraptor REX.” Here is a scale drawing:

Airframe is all Performance Rocketry G12 filament wound “Profusion” tubing. Curtis Turner at Performance increased the strength of the Profusion tubing over regular G12 by adding additional wrap angles. It also is laid up using carbon-impregnated epoxy, resulting in a black color instead of the typical green. Nosecone and boat tail are Performance Rocketry 4:1 ogive fiberglass. To get the right thickness for the fins without adding excessive aft mass, I will use 3/8” Nomex honeycomb from Giant Leap Rocketry.

I have a propensity toward avoiding going small in my certification attempts. My Level 1 cert was a dual deploy Giant Leap Crossbow on an I284W. The Velociraptor was also dual deploy on a K550W. In keeping with the “Go big or go home” philosophy, this certification attempt will be on a CTI N4100 Red Lightning (74% N) motor. The red flame will also keep me from having another set of certification photos showing a White Lightning motor.

Electronics will consist of a MARSA4 with a PerfectFlite MAWD as backup. A Featherweight Raven will also be on board as a recording-only altimeter to provide a third data set. There will also be a GPS tracker and radio tracker in the nosecone.

As was the case in my Avenger build thread, I plan to detail each step with high resolution photos. This is the only way that I have seen to adequately show what I am trying to show. I understand it increases loading times and may require scrolling on smaller screens, so I apologize in advance. I am a busy professional and father of three, so this will likely be an extended build thread. Please be patient and I will post as often as possible.

Finally, although I appreciate any insight into the build and my techniques, as is the case with all L3 builds, there are many “correct” techniques. Those shown here will likely raise some eyeballs. I welcome comments, but please understand that final build design and techniques are completely at my discretion under the guidance of my very experienced mentors, Kent Newman and Jim Wilkerson.

All that being said, let’s get it on!

Here are a couple of photos of most of the raw materials needed. A 12” ruler is included for scale. This will end up being a pretty good sized project!:jaw:

The tube on the left is the motor tube. It's 98mm diameter, 48" long. The N4100 will stick out another four inches or so from the forward end of the motor tube. The nosecones are 32" long without the shoulders. Two pieces of the 3/8" honeycomb fin stock are shown on the right side, one from the edge and one from the side.

 

[Rocketman248]

This one is going to be awesome! I fully enjoyed your Avenger thread, so this one will be equally sweet.

Nice choice on motor too. I like "big" certs!

 

[bandman444]

*subscribe* this will be amazing

 

[patelldp]

I love your threads, but there are just too many pictures!!! My computer/Blackberry have seizures trying to open them. Just kidding, should be an awesome build.

 

[pcotcher]

Outstanding - just outstanding - can't wait to watch, I hope to learn as much as I did from the last one!

 

[stickershock23]

Woa.... thats an awesome investment.. this should be fun to watch build..

 

[cjl]

N4100? Very, very nice - that's an awesome motor. I look forward to seeing this.

 

[o1d_dude]

A rocketeer in my club has a Velociraptor that he occasionally launches at our range.

Upscaled 2x? Whoa!

I think you are going to be very happy with the end result.

 

[Cogino]

I am really looking forward to this build!! You have some amazing building techniques.

 

[dixontj93060]

Fun fun! :roll:

 

[rocdoc]

The first step in building this rocket is cutting the fin blanks. The original Velociraptor has 3/16" thick aircraft plywood fins. Upscaling these to 3/8" thick presented a little bit of a problem. 3/8" thick G10 would be extremely heavy. 3/8" aircraft ply would be lighter, but less resistant to damage in the case of a hard landing. In building my Avenger, I used Nomex honeycomb from Giant Leap Rocketry. It is extremely lightweight and stiff. However, it has to have edging applied or it lacks both aerodynamic and aesthetic appeal.

An easy way to edge the material is to apply an appropriately sized dowel to the edge with epoxy. This covers the honeycomb, but can't be beveled like the original Velociraptor fins and also seems like it would pop off easily on a hard landing.

I therefore elected to use a similar procedure to the one I used for the Avenger. I will apply edging material inserted between the two layers of G10 after removing the honeycomb. The bevel will take up just about 3/4 of an inch on the edge, so I decided to cut 7/8" from each exposed edge of both sets of fins that make up the split fin design. This would allow me to form a bevel without disrupting the G10 covering the honeycomb.

One of the great features of Rocksim is the ability to print fin templates. The first step for cutting my fin blanks was to print, cut, and tape the templates together for both sets of fins.

Then, I removed 7/8" from all exposed edges.

Once the templates were cut to size, I placed them on the fin stock and marked the edges. I was able to get two of the large forward fins from each 12"x24" sheet of honeycomb with only a small amount of waste.

Here is the honeycomb marked for a forward fin.

This is the setup I used to cut the fin blanks. My wormdrive circular saw is fitted with a fine-toothed specialty fiberglass-cutting blade. The edge of the short side of the saw shoe is 1 1/2" from the edge of the blade. I therefore made marks 1 1/2" from the desired cut line and clamped a section of angle aluminum in place at those marks. I like the Bessey clamps for this because they are flat and can be used to lift the piece off my work bench, allowing me to make a cut right over the bench top. They also don't mar the material like c-clamps can.

Using the angle aluminum as a guide for the circular saw, I can then make a nice straight cut in the fin stock. It should go without saying, but whenever I cut fiberglass, I wear a full respirator, goggles, hearing protection, and long-sleeved clothing. I would also mention the nitrile gloves, but I pretty much wear those any time I am building or launching rockets, so those don't really count as fiberglass-specific protection.

Here, the first cut has been made. This will be the leading edge for both fins cut from this piece of stock.

I needed to cut a notch in the aft end of the fin tab on the forward fins. This notch serves two purposes. First, it provides more even spacing of the centering rings. Second, it allows me to use my beveling jig to bevel the trailing edge of the fins. This will be detailed in a later post.

A circular saw is not much good for notches like this, so I use my jigsaw with a fine-toothed metal cutting blade. It cuts through this material without a problem.

Here the notch has been cut out. This was the last cut for this fin blank.

This photo shows both sets of fin blanks that will make up the split fins.

I had two issues during this step of the build:

1. I shouldn't have wasted my time cutting the 7/8" edge off the fin templates. It would have been quicker to just use the original templates, then measure 2 3/8" from the edge (to account for the 7/8" wide edge and the 1 1/2" wide shoe) and clamp the angle aluminum at that point. The one advantage I had by cutting the template was visual confirmation that the line was being cut in the right spot. But since I cut the templates down, I had to print, cut, and assemble a second set of templates for the next step.

2. An astute observer may notice that I failed to cut a 7/8" strip from the aft edge of the aft fin blank. This would have prevented me from beveling the aft edge. Fortunately, I caught the mistake and cut the proper amount off the fins before it was too late.

 

[rocdoc]

As stated in my previous post, I wanted to ensure secure attachment of the edging material to the fin blanks. Because I will end up with 7/8" of exposed edge material, I did not feel comfortable with only 1/4 or 1/2" of material inserted into the edge of the fin blank. I therefore chose to have a 3/4" tongue of edge material attach the edge to the fin blank.

To do this, I first needed to remove 3/4" of honeycomb material from each edge. I do this on my router table. The table is set up with a 1/4" x 1" double-fluted straight bit set to a height of 3/4" above the table. The fence is then placed so it nearly touches the back of the router bit. There is just enough room for the fiberglass skin to pass between the bit and the fence.

I originally set up my adjustable featherboard to hold the fin material in place against the fence. I later removed the featherboard for two reasons explained below.

Here I have turned the router on and have started feeding the fin blank through. The back fiberglass skin passes between the fence and the bit and the front skin passes in front of the bit.

Here the first pass is almost complete. Notice several things in this picture. First, I am working without a guard over the router bit. This is an unavoidable risk and requires vigilance to prevent injury. Notice that my hands are well away from the bit. Second, I continue to wear my respirator. Third, although you can't see it, my shopvac is attached to the back of the router table. This sucks the cutting debris away from the bit and keeps things much cleaner. Very little dust escapes.

Here is the result of the first pass over the router bit. The side that was between the fence and the bit has been cleaned up quite well, with most of the honeycomb material removed cleanly. The opposite side still has a significant thickness of the honeycomb material remaining.

The fin blank is reversed and run through the router again. This removes most of the remaining honeycomb, but leaves much more scrap honeycomb still attached to the fiberglass skin than there was on the other side

To remove the remaining material, I have found the easiest method is to run the stock over the bit about halfway to remove the majority of the material, then reverse the feed direction, pushing the fin back to the right, then run the stock to near the end and reversing again, nicely cleaning up the fiberglass skin. This is not necessary on the first side, as the first cut is much cleaner than the second. This is the first reason that I removed the featherboard, as it prevents the piece from moving backward on the table.

Here is a photo of a fin edge after completing this step:

The result is a clean 3/4" deep channel between the two fiberglass skins.

The other reason that I removed the featherboard can be seen in the next picture. The fiberglass skins have a tendency to curve inward when the honeycomb is removed from between them. Because the featherboard is only about a half inch tall, it worsens this curving by pushing against the unsupported fiberglass skin.

The photo shows a fin edge after removing the honeycomb compared to the original honeycomb edge. It is obviously a much smaller space. This makes it much harder to feed the fin material over the bit so that the fiberglass skins pass on either side of the router bit. As a result, the fiberglass skins have a tendency to get nicked up by the spinning router bit.

To help alleviate this problem and allow easier feeding of the material over the bit, I wedge a piece of scrap fin stock between the skins above the height of the router bit. This separates the skins and makes it safer to feed the fin stock through.

After removing 3/4" of honeycomb from all edges except for the root edge, I adjusted the bit to a height of 1/2" and passed the rood edges of ths fins through. The resulting gap will be filled with a piece of wood to provide more bonding surface area for the root edge against the motor tube.

This photo shows all fin blanks with edge material removed. It was at this point that I realized my stupid mistake of not accounting for a trailing fin edge on the aft fins. I cut 7/8" from the trailing edges and ran the blanks back through the router. I was pretty disgusted with myself for my stupidity, so I didn't get pictures of the fin blanks after doing this. They looked a lot like this picture, but with the aft fins (on the left in this pictures) a little more narrow.

 

[redsox15]

I am officially subscribed to this thread...wow...this is going to be a great thread.

Matt

 

[DAllen]

Uhhh...

$$$...:y:

op:

 

[rocdoc]

Thanks for the encouraging words, all. Should be a fun build. I always forget how long it takes me to do each step. I always start on a step and think, "I should be able to get this done in the next 10 minutes" then glance up at the clock three hours later and wonder where the time went.:blush:

Fun fun! :roll:

Congratulations on your cert, Tim. :clap:Watched your Hope N Pray thread with great interest over the last several months.

I love your threads, but there are just too many pictures!!! My computer/Blackberry have seizures trying to open them. Just kidding, should be an awesome build.

Sorry. I can prescribe your computer some seizure meds if needed.:roll: Good luck on your own cert attempt. Enjoying your thread as well

Woa.... thats an awesome investment.. this should be fun to watch build..

Uhhh...

$$$...:y:

op:

Yeah, it's a pretty big investment. Please don't tell my wife. :bangpan: We currently are still subscribing to the don't ask--don't tell policy. She's getting ready to start her first HPR rocket, so I'm not sure how much longer I can keep her in the dark.

 

[rocdoc]

As was mentioned in the previous post, the honeycomb material requires edging to cover the honeycomb and to allow for beveling. There are several potential methods for edging the honeycomb. A 3/8” dowel applied to the edge of the honeycomb with epoxy is likely the easiest, but does not allow beveling and I was concerned about popping the dowel off the edge in the case of a less-than-perfect recovery. Giant Leap shows the fin material edged with hardwood or G10. It is shown applied flush with the edge of the honeycomb material with epoxy, which looks better than the dowel, but still has the potential to pop off the edge.

As was the case with the Avenger, I elected to apply the edge material between the fiberglass skins. This holds the material more securely. On the Avenger, I used ¼” Nomex, so I used strips of ¼” aircraft plywood for edging. This worked fine. I was concerned on this build about the ability of plywood to take landing damage on such a heavy rocket. Additionally, I could not find 3/8” aircraft plywood from my usual source. 3/8” thick G10 would provide better durability, but would end up weighing more than the entire rest of the fin. In addition, G10 is somewhat painful to bevel.

I therefore elected to use a sandwich consisting of a middle layer of 3/32” G10 with a layer of 1/8” 5-ply birch aircraft plywood on each side. The middle G10 layer will provide extra impact strength. The aircraft ply decreases overall weight and is easier to bevel than the G10.

When first designing this project, I originally intended to cut strips of G10 and plywood, sandwich them together, then cut and bevel them to apply to the fin edges. However, I was concerned about the strength of the beveled corners. I therefore elected to cut the G10 as single pieces the shape of each fin. This strengthens the corners. The aircraft ply will then be applied by cutting and beveling (detailed in the next step). Although this design is much stronger, it requires much more G10 than just cutting edging. I ended up using one and a half 12”x48” sheets of G10.

First, I placed the paper template from Rocksim on the G10. I applied masking tape under the template edges. This provided protection from splintering of the G10 and makes it easier to see the pencil line.

After the lines were marked outlining the fin, a spacer was used to make marks on the G10 the same distance from the cut line as the distance between the edge of my circular saw and the blade. I found it is easier to cut a couple of blocks and use them to make consistent marks than it is to measure each time.

This photo shows the fin outline on the full tape and the guidelines on the shorter strips of tape.

A piece of angle aluminum is clamped in place with its edge on the guide lines. This provides a solid guide for the circular saw. The G10 sheet is clamped securely to a sheet of sacrificial plywood.

The circular saw is fitted with a fiberglass cutting blade and the blade depth is set just deeper than the thickness of the G10. The G10 is cut while holding the edge of the saw shoe against the angle aluminum guide.

After cutting the long edges using the circular saw, the short sides are cut using a jigsaw with a metal cutting blade.

One of the lessons I learned during this build is that the templates cut from the Rocksim file are nice, but relatively difficult to use to get exactly the same lines for each fin. I found that the template kept moving and I had a hard time duplicating the results. I therefore elected to use the template to cut one fin, then use the fin as a template for the two other fins. This is much easier and provides more consistent results than reusing the paper template. This picture shows me outlining the second fin using the first as a template.

The next time I use these techniques, I will cut the fiberglass edging first, then use that as a template for cutting the honeycomb as well. This should produce better results.

After both sets of three fins were cut from the G10, they were stacked and sanded to even them up the rest of the way.

I then started cutting out the excess fiberglass from the inside of the fin edging to accommodate the nomex fin stock. Again, I used a block of wood cut to the width of the saw shoe plus an additional 1 9/16", the desired width of the finished edge material, to mark for placement of the angle aluminum guide. This photo shows one of the forward fin blanks with the desired cut line marked on masking tape and the angle aluminum saw guide clamped in place.

In order to provide better support for the saw blade, I clamped a sheet of G10 to the plywood alongside the piece to be cut. This kept the shoe level, keeping the resulting cut straight.

After the first cut is made, the piece is flipped over and the lines are marked for the cut along the leading edge. This allowed me to use the same measurement for the guidelines on both sides, keeping the width more uniform.

The smaller aft fins are more difficult to cut using the circular saw as there are several inside angles. Therefore, the fins were stacked and taped together, then the cut lines were marked and the jigsaw with the metal cutting blade was used to make the first two cuts.

A half-inch drill bit was used to allow insertion of the blade along the inside angles. After cutting the along all lines, the inside edges were cleaned up using a file and 60 grit sandpaper.

This photo shows a set of finished edges.

Here, the edges are dry fit on the fin blanks to show how they fit inside the honeycomb skins.

 

[EMiR]

the Ply/G10 layering, should do the trick nicely,

I'm intrigued though, you have routed the root cords of the fins, are you planing on filling with Ply? or just filling with epoxy?

 

[edwinshap1]

the Ply/G10 layering, should do the trick nicely,

I'm intrigued though, you have routed the root cords of the fins, are you planing on filling with Ply? or just filling with epoxy?

my guess is he's going to spread the root cords out for more surface adhesion to the MMT?

 

[EMiR]

Good chance the Nomex covering would crack doing that, and would give a very weak bond, besides the rear fin's g10 doesn't allow it as it stands

 

[rocdoc]

the Ply/G10 layering, should do the trick nicely,

I'm intrigued though, you have routed the root cords of the fins, are you planing on filling with Ply? or just filling with epoxy?

I plan to fill it with a 3/8"x1/2" piece of ply or hardwood. Wanted more solid adhesion than I would get from the Nomex itself. The plywood will securely bond to the fiberglass skins of the honeycomb material and make the root edge adhere better.

Bryan

 

[rocdoc]

After the G10 fiberglass edging was cut, it was time to work on the 1/8" 5-ply birch aircraft plywood edging. This will be applied to each side of the G10 edge base and serves three purposes:

First, it decreases the weight of the edging material over a solid 3/8" G10 edge.

Second, the plywood is easier to bevel than G10.

Third, a very thin rabbet can be applied to the plywood, allowing a better joint between the fiberglass skin on the honeycomb and the edging material.

The first step was to cut strips of appropriate width from the 1/8" plywood sheet. I did this on my table saw with the fence set for 1 9/16" from the blade edge. This allowed 7/8" of exposed edge and 11/16" of material inside the honeycomb core material.

This photo shows the strip of 1/8" plywood being cut from the sheet. It also shows the use of a featherboard to keep the sheet against the fence and a pusher to keep my hands away from the blade.

After cutting the strips, I set up my router table to take a very thin rabbet 11/16" wide from one side of each strip. This rabbet accomodates the thickness of the fiberglass skin on the honeycomb material. Here, I use two featherboards to hold the material against the fence and one on top to hold the material down on the table.

At first, I set up the router with the material between the blade and the fence. DUH! The spiral bit turning in the same direction as the material was feeding grabbed the plywood and shot it across the room. No harm, no foul, and hey, I never said I was a woodworker.:blush:

So I then set up the router table correctly, with the bit set in the gap between the halves of the fence, just sticking out far enough to remove 1/64" or so of material from the plywood.

The fences were set back in place and I started again, this time without the excitement of shooting plywood across the garage like a balista!:roll:

This photo shows the minimal amount of material removed from one side of the plywood. It doesn't even go all the way through one of the 5 layers of the 1/8" plywood. I verified with calipers that indeed the width of the rabbet was the required 11/16"

Finally, it was time to start the cutting. I decided to do both sides of the edges for all three fins at once. This kept things uniform. First, I marked the desired angle using my digital protractor. I then used the reading from the protractor to set up my compound sliding miter saw with laser guide. The laser line was lined up with the pencil line. . .

. . .and the first cut was made. I'm using a 100 tooth finishing blade in the 12" saw.

The angle was then transfered to five other plywood strips; two set identically to the first and three set as mirror images of the first three strips to cover the same edge section on the other side of the fiberglass. By making all six cuts at once, I was assured identical angles for each piece.

The most difficult portion of this step was cutting the very acute angles for the apex of the aft fin. both sides needed to be cut to an angle that was more acute than my saw would allow (it only goes to about 32 degrees from parallel to the material, 58 degrees from a right angle). I needed just a little bit more than this.

First, I measured the angle with my protractor.

Then, I locked the angle with the set screw and transfered it to the plywood material.

I set my saw up for the most extreme angle it would move to and placed the edging material on the saw at an angle so that the line drawn with the protractor matched the line from the laser. This picture is dark, but I wanted to show the laser light lining up with my mark. The photo shows that the piece is secured to the work table at an angle to the fence.

I do not recommend using power tools in any way other than as instructed by the manufacturer, but this worked for me. Because of the extreme angle, even my brand-new finishing blade had a difficult time making a clean cut and not ripping pieces out of the plywood. To help with this and to further hold the piece in place, I used masking tape to support the plywood and lock it down a little more on the work table.

The result is actually quite good, with the bevels coming out almost perfectly.

This photo shows both of the fin edges. The fiberglass center is under the plywood. This is a dry fit and the bottom layer of plywood is not placed. Everything lines up well.

 

[redsox15]

your craftsmenship and eye for every detail is incredible! very nice work so far.

Matt

 

[rocdoc]

With the fins progressing slowly, I spent some time to prepare the material for my centering rings. I will be using fiberglass reinforced 1/2" aircraft plywood for the rings. I had a couple of sheets of 1/4" 12-ply birch aircraft plywood, so I decided to use it for the centering ring stock.

One of the sheets was previously covered with 5.6 oz S-glass, so I only needed to apply the fiberglass to one sheet. I cut an oversized piece of fiberglass and applied it dry to one side of the plywood. I then added West System resin with 206 hardener to the dry fiberglass.

A foam roller was used to thoroughly wet out the fiberglass, starting from the center and working to the edges.

After the fiberglass was thoroughly wet out, the plywood sheet was flipped over and a rotary cutter was used to trim the excess fiberglass from around the plywood sheet. Fiberglass was applied to the other side of the plywood in the same manner.

I then spread more epoxy on the other sheet of plywood, which already had fiberglass on both sides.

This sheet was then placed on top of the other sheet and the entire unit was wrapped in release material and breather. This photo shows the release and breather taped in place.

The entire setup was then placed in a large vacuum bag and a vacuum applied. This gives much more consistent pressure than clamping, especially on a large piece such as this.

This is my RobinAir vacuum pump. It quickly pulls a vacuum and after a few seconds gets to 28-31 in Hg.

After the epoxy cured, I removed the vacuum bag. This photo shows the breather material on the edge of the plywood sandwich. As you can see, excess epoxy was squeezed out from between the two plywood sheets and was absorbed by the breather material.

The result is evenly applied epoxy with appropriate resin for optimal strength to weight.

 

[rocdoc]

Although the router does a decent job of removing the honeycomb material from the fin stock edges, it does leave some of the honeycomb on the fiberglass skin. To remove this, I used a sharpened chisel.

The fin stock for this build has been much more difficult to work with than the honeycomb in the Avenger build. The Avenger fin stock was 1/4" and this build uses 3/8" thick stock, but it seems that this would not cause a huge issue. There were three major issues with using this material:

First, the router had a tendency to catch the trailing edge of the material, tearing out chunks of the fiberglass skin material. I attempted multiple adjustments and two different router bits, none of which resolved the problem. This photo shows one of the worst-damaged fins.

Second, the fiberglass skin seems more brittle than the skin on the fins for the Avenger. This may well have been the cause of the problem with the router as well, but as this photo shows, it was easily gouged by the chisel when removing the remnants of the honeycomb material. I did this once in the entire construction of the Avenger, but had it happen at least once or twice per fin for this build.

Finally, a couple of the sheets of honeycomb stock I received from Giant Leap had holes in the fiberglass skin. This photo shows the worst of the sections. I don't have time to wait another week or two for more honeycomb material, so I used it as is.

The only reason I used the material despite these issues was because it will be covered with a couple of layers of carbon fiber. If I was planning on using it bare, these issues would have been difficult to work around. I don't know if I'll use the honeycomb material again, as it has been much more difficult to work with this time than the last.

 

[cjl]

That's really odd - I've used the Giant Leap 3/8" glass honeycomb before, and I had none of the problems that you are having. Maybe they changed their manufacturing method, since I did get mine in 2006.

 

[rocdoc]

The next step in this build was to assemble the fiberglass and plywood edge materials into a single edging piece. Remember, the edging is made up of a sandwich of 1/8" aircraft ply, 3/32" G10 fiberglass, and 1/8" aircraft ply. The reason for using the 3/32" G10 is that the 3/8" thick Nomex honeycomb is just a little shy of 3/8" thick. This was the closest fit.

First, I thoroughly scuffed the G10 with 40 grit sandpaper.

Next, I cleaned the G10 with acetone to remove dust and oils.

I set up the materials needed for vacuum bagging the edging. In this case, I am not actually using a vacuum bag, however. In order to ensure that the edge material did not warp in a regular vacuum bag, I instead used a scrap piece of Plexiglas as backing material and placed the sealant tape directly on the Plexiglas to act as a vacuum frame. I cut a couple of pieces of release material, some breather, and a sheet of "Stretchlon" film from Fiberglass Supply. This material is flexible and more easily conforms to the contours of the edge material for this application. It is important to get all the supplies ready to go beforehand so you don't get stuck with epoxy curing and no vacuum set up. All the materials are sitting on top of my curing oven that I made from 2" foil-backed foam board. Two light bulbs provide the heat and a Johnson Controls thermostat keeps it tightly in the goal range.

Here I'm getting ready to work on the edging for one of the aft fins. I've laid out a piece of butcher paper and dry fit the plywood on one side of the G10.

I started with one of the two pieces that make up the tip of the fin. This is probably the most cosmetically important point, so I wanted to make sure it fit well. First, I applied masking tape to the plywood and lined it up on the fiberglass.

I then attached the piece on the opposite side of the edging, lining up the two opposing plywood pieces. The other piece that makes up the tip was then taped in place relative to the first side, ensuring a tight fit at the joint.

Here, all the pieces have been attached with tape.

This picture shows all the plywood edging on one side of the fin flipped back along the tape joints.

I then mixed a small batch (one pump) of West System resin with 205 (fast) hardener and applied epoxy to both the G10 and the plywood using a foam brush.

The plywood sections were then flipped back into place. The joints were reinforced with extra tape to avoid shifting or gapping.

After repeating the process on the other side of the material, the completed edge unit was placed on a sheet of release material and butcher paper inside the Plexiglas vacuum frame. The second sheet of release material was then placed on top.

A couple of layers of breather material were then applied and a vacuum cup was placed on the breather in the center of the fin area. The Stretchlon material was then pressed onto the sealant tape. I'm using cheap, half-width sealant tape as it is less expensive and quite adequate for this application.

The entire assembly was then placed on the floor of my curing oven. I then attached my vacuum hose that runs directly into the curing oven through a small opening at one end. This allows me to pull a vacuum on a piece while it is warmed in the curing oven. In this photo, I have started the pump and most of the air has been evacuated from the frame.

I then plugged in the oven. I do this before closing the lid so I can ensure both lights are working properly. I also placed a piece of butcher paper with a layer of epoxy on it into the oven. This allows me to open the oven and check the cure of my epoxy before I take the vacuum frame assembly apart.

As I stated before, I am using West System epoxy. This epoxy is designed for room temperature curing and actually is not made stronger by high temp curing or post-curing. In the summer, I could easily leave the piece out to cure, but in the winter the garage is somewhat cool and I use the oven just to keep the temperature up a little and ensure timely curing of the resin. Here is a photo of my Johnson Controls thermostat. It easily keeps the oven between 95 and 98 degrees by cycling the lights on and off. It is attached to one corner of the oven with the temperature probe centered in the oven.

This photo shows an aft fin edge after curing. The excess epoxy has been squeezed out from between the sections and absorbed in the breather material.

Here, the tape has been removed and excess epoxy trimmed off.

This photo shows a dry fit of the edging material on an aft fin. The results are quite good. The root edge still needs a reinforcing layer of hardwood between the fiberglass skins.

 

[rocdoc]

I finally have a couple of minutes to post. Not much progress right now. Seems like one step forward and two steps back.

The next step was to attach the completed edging material. First, I needed to cut the stock for the root edge of the fins. The thickness along the inside of the fin edges was 5/16", so I needed 5/16" x 1/2" stock. I used my planer to cut down 1/2" square hardwood stock to the appropriate dimentions.

All the needed pieces were planed at once, making quick and accurate work of this step.

The pieces were cut to size and mitered for the larger fins on my miter saw. Here is one fin setup with the edging, honeycomb fin stock, and root edge ready to be glued together.

I use Gorilla Glue to attach the edge pieces to the fins. It is strong, relatively easy to work with, and because it expands as it cures, I feel it fills the gaps and provides better attachment of the edge to the honeycomb material.

First, the glue was brushed into the gaps around the fin stock and allowed to sit exposed to air for a few minutes.

Next, plain water was brushed along the edge material.

The edge section was placed and excess glue wiped away. The root edge piece was then placed. Painters tape was applied along the edges to keep things aligned. As I said before, the Gorilla Glue expands as it cures. The best cure is achieved by clamping securely for at least two hours. I decided to use my vacuum pump to give me more consistent pressure along the material. First, I wrapped the fin in release material.

I reused the vacuum bag that I used to press my centering ring stock, so it is oversized. I originally planned to do one pair of fins to see how it would work, but I decided to just do one fin in case I wasn't satisfied with the results. I didn't want to go through the hassle of cutting two more fins and edge assemblies if I screwed up at this point. Here the fin is shown under vacuum. Thanks to Sandy H., I will be building a vacuum press this week (once the parts all come in), so this should be the last time I use my vacuum pump in continuous mode. I'll post some pics of the press once it's completed.

After allowing the glue to fully cure, the fin was removed from the vacuum bag and the tape was cleaned off the edges. Despite all the problems with the honeycomb material, the fin turned out fine. The gorilla glue really stuck the fiberglass skins to the plywood with no gaps or bubbles. This photo shows the root edge with its hardwood reinforcement. This should provide a more secure attachment to the motor tube than the honeycomb would have provided alone.

 

[rocdoc]

NOTE: This post is a little off topic, so please forgive me for including it here. I thought it may be of general interest to members, so have included it in the build thread.

In response to a previous post regarding vacuum bagging, Sandy H. saw my Robinair pump and recommended a "Vacuum Press" system as excellently detailed, including free, step-by-step instructions and links to purchase all necessary parts at https://www.joewoodworker.com. I had never heard of a vacuum press. Perhaps those with more experience have used them for years, but I haven't seen them shown on this forum.

Essentially, a vacuum press is a setup that allows the vacuum pump to run only intermittently while maintaining appropriate vacuum on the vaccum bag. It works very similarly to an air compressor, except that the compressor has a tank that stores pressurized air while the vacuum press has a tank that "stores" vacuum pressure. The large volume of the tank allows for small leaks in the vacuum bag without having to keep the pump running the whole time. It is much quieter and more efficient than running a continuous pump.

Here is a photo of the completed system. Joe Woodworker uses homemade tanks from 4" schedule 4 PVC for his press. I elected to use a portable 10 gallon air tank designed to be filled with a compressor and taken where air is needed. I removed all the plumbing from the tank and replaced it with the parts needed for the vacuum press.

I chose to go with an upright system that would have a smaller footprint in my garage space. Here, the upper part of the system is shown. On the left side of the picture are the power switches and the vacuum pressure controller. On the right is the Gast vacuum pump (I wanted to go with an oilless design), the Mac Valve, filter, and subreservior.

The PVC "tank" above the vacuum pump works with the Mac valve to release the vacuum pressure at the pump when the pump turns off. This keeps the pump from having to turn back on while a vacuum is still present, which is harder on the vacuum pump. The filter keeps debris from reaching the pump. The brass fitting on the right side is a muffler that reduces the volume of the pump when it is running.

This is the vacuum pressure controller, which is the brain of the system. It is connected by a hose to the tank manifold and samples the vacuum pressure. When it reaches a set point (-25" Hg on my press), it turns off the vacuum pump. Because no vacuum bag setup is perfectly airtight, air slowly leaks back in the system, decreasing the vacuum pressure. When the controller senses a pressure of about -22" Hg, the pump is turned back on and pressure is returned to the set point. My system pulls down an average sized bag quite quickly. It then cycles on every 20-40 minutes (depending on the airtightness of the bag setup) for a minute or two at a time.

I chose to go with a two-switch setup. The "normal" switch turns the system on and runs it in the mode described above. When the "cont." switch is also turned on, the vacuum pressure controller is bypassed and the pump runs continuously. I may use this when I want to ensure a piece has a constant -25" Hg vacuum without the variation caused by normal mode, but thus far the normal mode is working just fine.

Here is the 10-gallon tank in the lower section. It is recessed far enough into the cabinet to protect the valves, fittings, and gauge. The valve leads to the bag hose. The hose on the left side of the photo leads to the pressure controller, allowing it to sample the tank pressure. The hose on the right leads to the vacuum pump through the Mac valve.

This photo shows the back of the cabinet I built. There are casters near the bottom to allow easy movement of the unit in the garage. The coat hooks keep the power cord and vacuum lines organized.

Here is the last of the large fins for this project under vacuum. I have switched from vacuum cups to the aluminum vacuum coupler shown. The fin edging has been glued to the fin stock with gorilla glue, which works better under clamp pressure. The vacuum bag assures consistent pressure across the entire piece.

Here are the three forward split fin sections showing a complete fin blank, a blank with the tape still on after attaching the edge, and the final fin in the vacuum bag.

As a final modification to my vacuum press, I built a second hose assembly. This assembly attaches to the end of the first hose by way of a quick release valve. It allows independent control of two vacuum bags. This way, one bag can remain under constant vacuum pressure while the second bag is attached and brought down to pressure as well.

I'm quite happy with the design and function of the vacuum press. It should pay for itself by saving me the cost of replacing a vacuum pump every few years. It is convenient to use and is capable of pulling a near-instant full vacuum by turning the unit on and allowing it to come to pressure before attaching it to a vacuum bag.

Thanks again to Sandy for recommending this design and to Joe Woodworker for providing the resource on line. Incidentally, his price on most of the materials is lower than at the local Home Depot and many of the items cannot be readily found locally. He offers a full "kit" that includes all the hard-to-find parts at a very reasonable price. He also has the lowest prices I've seen on Gast vacuum pumps.

Hope this was a useful diversion. The next post will be about rocket building, I promise.:wink:

 

[butalane]

 

[Sandy H.]

The system looks nice. You made a lot of the same modifications I did on mine (valves, continuous mode etc). I hope you find it a good tool for the shop.

Also, I have to thank Paul Phillips for bringing this design to my attention when he hosted a composites workshop for our club over the summer. Good ideas travel far and fast!

Looking forward to more of this awesome build.

Sandy.