Rocket Ship 27: An L3 Build Thread

[JLebow]

I’m Joey and for the last four years I’ve been flying with Oregon Rocketry. There was a 25 year gap between when I flew my last Estes model and learned that there was a motor larger than a D. After attending my first Balls in September - my Go Devil 54 went to 18,950 ft on a K700, I’ve been inspired to start my L3 project.


I like to build scratch rockets as depicted in the videos of tfish38 and JohnCoker. I’ve built and flown five of them so far – rolling the airframe on a mandrel, and making nosecone molds from plugs turned on the lathe. This one is a fiberglass clone of mean machine that flies on 29 mm H and I motors.


This carbon fiber rocket did 12,700 ft on an I216:


Last year I build a curing oven because I’m not allowed to let the epoxy curing on the dining room table anymore (wives get funny about certain things) and the epoxy doesn’t cure well at 40 deg F in the garage. The oven build thread is posted here.

My latest rocket had balsa core fins wrapped in carbon fiber, a step up from cutting flat fins out of G10 sheet.


For my L3 project I’m going to scale this up for an Aerotech DMS M1350. Once my taps sign off on the design I will begin documenting the build process.

 

[fyrechaser]

Looking forward to following your build....

 

[JLebow]

These are some details of my previous build, and what I learned and will be changing in my approach for the L3 build.
Image 1: 54 mm airframe and flat profile fin cores.


Image 2: I glued 1/4" balsa sheet to each side of the fin core. I adjusted the angles of some metal rails on each side of the 2x4 base, and with a bar and sticky backed sand paper, I formed a compound double wedge shape on all three fins. I realize the balsa is not in the end grain orientation.


Image 3 and 4: A couple of layers of carbon fiber is vacuum bagged, leaving a lumpy imprint of the breather cloth on the external face of the fin.


Image 5: A bit of filling with epoxy, and sanding smooths the surface quite a bit.


For the L3, I want to make a plug of the fin, and create a mold off the plug. Then when I vacuum bag the fin half, the mold surface will produce a smooth exterior finish, and the bumpy side will be internal to the fin.

 

[David Schwantz]

Last year I build a curing oven because I’m not allowed to let the epoxy curing on the dining room table anymore (wives get funny about certain things)

Boy, and my wife let me build a Byron P-47 in our bed when I was layed up

 

[Mugs914]

Boy, and my wife let me build a Byron P-47 in our bed when I was layed up

 

[grouch]

Sweet thread man. Do you have any more pics of the yellow Mean Machine?

 

[JLebow]

Sweet thread man. Do you have any more pics of the yellow Mean Machine?

I do. When I was a kid I always wanted a Mean Machine, but it was quite a bit more expensive than the other rockets.

This is the rocket corner of the living room.

The clone on the right is nearly to scale. Estes version on the left. The clone is few inches too long and has a pointy 5:1 Von Karman nose. The estes version also doesn't have an eggfinder tracker or dual deploy altimeter.


Here is the plug, mold and nose cone. The PVC plug was a pain to machine. The chips were stringy and would melt and cool in a blob wrapped around everything. The few dollars more for aluminum is worth it.


The original didn't have through the wall fins. The rear closure is diy, too.

 

[grouch]

Very nice!

 

[rharshberger]

Go for it Joey, looks good, hopefully we get to see it at TCR sometime. You still got the Thug, what a screamer it was on a 54mm motor?

 

[JLebow]

Go for it Joey, looks good, hopefully we get to see it at TCR sometime. You still got the Thug, what a screamer it was on a 54mm motor?

Hi, Rich. Thug took some damage at NXRS 2019. I was experimenting with deploy bags, and basically came in drogueless. Next time I will have a pilot chute. It will fly again but needs a little repair at the top of the air frame (similar type of damage to a zipper event).

Thug is a stubby 4 in diameter, 33 in long Binder Designs kit. I wrapped the air frame twice with 6 oz cloth and tip-to-tip reinforced the plywood fins. I also deleted the stock 29 mm motor mount and installed a 54 mm mount. Now it loves to fly on J275 and J415's.

It also has a water droplet paint job. After the base green coat goes on and dries conventionally, I sprayed the rocket with water in a spray bottle to create a bunch of drops on the surface. Then I did a light dusting of black paint at an angle almost parallel to the surface, and a light dusting of white from the opposite direction as the black. The black and white have to be oil based. The droplets of paint stick to the water, and as the water dries, the droplets bond to the green base coat leaving behind a cool pattern. I clear coated the rocket to smooth the surface and lock in the black and white paint.

 

[JLebow]

This post covers the mold making steps and creation of my nose cone. This is my fourth nose cone mold. For the 29 mm, 38 mm, and 54 mm sizes, I turned my own von karman plug. For 76 mm, I'm cheating a bit and using a pre-made 5:1 von Karman nose cone as the plug (the black one).


The mold will be made in two halves. The count for pieces in a mold has a lot to do with the shape of your part and specifically the ease of removing the part from the mold. You can't have your part keyed into the mold because removal of the part will destroy the mold.

The first step is to make a parting board. This board allows half of the nose cone mold to stick up above the board. This parting board has a lifting bolt under the nose cone shoulder, so I used my height gauge to ensure the nose cone was centered in the parting board.

On previous homemade nose cones, my tip is fragile and gets broken off before I complete the rocket. So I made an aluminum cap and hot glued it to the plug. I hope this will allow me to pack a lot more fiber into this area. It will certainly be stronger until just before paint, where I will form this area to a proper radius. If it is still weak at this point, I can cut off the top inch and replace it with a metal tip.

I don't want any cracks or gaps between the plug and the parting board. I used a plastic modeling clay (doesn't dry out) to make a bunch of "snakes" to fill the space. I used an old credit card to scrape the clay flush to the surface of the parting board. This takes a bit of patience. Once the clay is softened and warmed up it tends to scrape cleanly. When the clay is cold, scraping tends to pull the clay out of place.

I also added some side walls to the parting board. This is not required. My first three molds don't have them. They are intended to add a bit of stiffness to the mold to keep it from warping, but it also creates a flat bottom to the mold. The extra stability when trying to lay in the fiberglass is worth the extra effort in making the mold.

Next I waxed the exposed portion of the plug, the parting board and the side walls with Partall #2 mold release paste wax. Apply, let sit for 15 minutes, the wipe to a polished finish. I did three coats. Then I sprayed a couple thin layers of PVA. The surface looks rough like a really dry coat of light spray paint. But after the third coat, the PVA flattens out, and looks really smooth. The trick is to avoid spraying too much and getting pools of PVA. PVA provides a thin water soluble film to prevent epoxy from sticking to the plug and parting board.

Bubbles are a problem, and I've been able to minimize them by spreading a thin layer of epoxy and cabosil on my plug. I try to build up a epoxy in the sharp bends at the intersection of the plug and parting board. Fiberglass tends to bridge up (more bubbles and voids) it you try to push it into too tight of a radius. I let this first application cure to a sticky tack before placing any fiber. Then I place a layer of 1.5 oz. cloth and then 6 oz. cloth, and wet it out with epoxy. Bubble are not desired because they weaken your mold, but at least they are not on the mold surface at this point. The light weight glass is to avoid print through that is possible from the weave of a heavier cloth. I let this layer get to a leather stage (still able to make a good bond at this stage vs. fully cured epoxy, yet may first layers of glass are not going to move around on me). I add two layers of 18 oz cloth (real epoxy hogs) and a final layer of 6 oz. The last layer is probably not required, but I didn't want such a coarse weave on my mold. My fear would be getting cut on a ragged edge of the mold.

 

[JLebow]

Now all the wood can be pulled from the mold. Don't remove the plug. It is tightly fit, and won't go back as well if it is removed. Rough fiberglass at the edges can be trimmed. I didn't have a picture of this step, and the nose cone is back with its owner, so I staged this picture (with the final nose cone standing in as the plug).

More parting wax and PVA get applied. I hot glued my fences to the lower mold, but it is not required. I followed the same steps as the first mold half, except the first mold half serves in place of the parting board. Be sure to clear out all of the molding clay. Any clay left behind will result in an undesired depression in the mold. Once the mold cured, I had two mold halves. Also in this picture you can see the $18 badger airbrush I use to apply the green PVA.

The difference in color is a switch in brand of epoxy. I finished up some marine epoxy from Tap Plastics (amber, on the right) and switched to US Composites 635 (on the left).

Next I made a paper pattern. This is made by figuring half the circumference of the cone from tip to shoulder, plus a 1 inch allowance. I cut 10 pieces of 6 oz cloth and 2 pieces of 1.5 oz cloth.


The mold is waxed and coated with PVA. You will eventually get quite good at this step with all of the repetition. I brush a light coating of epoxy into the mold. A light pass of hot air from a heat gun pops a lot of bubbles, and reduces the viscosity of the epoxy to aid with wetting out the fiberglass. I placed a layer of 1.5 oz cloth into each mold half, followed by a layer of 6 oz, then a layer of peel ply - a teflon coated nylon cloth that allows excess epoxy to pool and be peeled away from the part after cure. Saves a ton on sanding time.

After this shell cured, the peel ply was removed, keeping the shell still in the mold. I trimmed off any fiberglass that was above the mold surface with a single edge razor blade. Before I started doing this shell method, I would get bubbles and voids on the surface of my cone because the following steps are difficult to accomplish without disturbing these surface layers.

I then wet out three more layers of the 6 oz cloth into each mold half. Using scissors, I cut this wet blob of cloth flush to the mold surface. More attention is required to work out any bubbles and misalignment from the trimming process. The last 6 oz layer (2 halves) were wet out , both pieces into a single mold, but they were offset from the position of the other layers such that half of the fiberglass piece stuck out above the mold surface on each side of the mold. This excess is not trimmed down. When I join the mold halves, this extra flap is worked over onto the other mold half. This last layer bridges the seam from the mold halves. Sorry, my hands were too dirty to take pictures of this step. I spent the next 20 minutes using a 1/4 dowel to roll out any bubbles trapped in the seam of the mold halves. I used Clico fasteners to clamp the mold haves together. These are like removable rivets. Wing nuts and bolts work just fine too.


Looking down the inside of the layup:

There is a layer of peel ply in there, that in the end wasn't required. I was going to lay the mold on the side, with the peel ply at the bottom to collect the excess resin that would pool. The resin could then be easily removed with the peel ply. I ended up letting the mold cure in an upright position, with excess resin pooling in the tip of the nose cone. Don't do this unless you were fairly efficient with your epoxy usage during layup, or you risk a two inch solid block of resin in the nose. This may be desirable if you already know how much nose weight you want to add. You could drop in shot, and have the resin bond it directly into the structure of the nose.

Here is the part, after removing it from the mold.

 

[Tobor]

Awesome! Thx for posting this process. Very informative.

 

[rharshberger]

Awesome! Thx for posting this process. Very informative.

I concur, the step of making a "shell" first before adding additional layers makes a lot of sense. Do the additional internal layers get added while the shell is still "leather stage/green" so that the epoxy layers chemically bond and not just mechanically?

 

[JLebow]

Awesome! Thx for posting this process. Very informative.

Thanks guys, but I can't take any credit for the techniques. I got all my information on the forum and youtube videos by forum members. In particular, these videos by tfish and John Coker were very helpful:



If you can time it right, applying more layers before the epoxy has cured on the previous layers is supposed to provide a very good (chemical) bond. Next best is supposed to be applying in short time after pulling the peel ply. None of this is known to me firsthand, just what I've read on forums on the internet, so its good to be...

 

[JLebow]

In this post I'll cover making molds for the fins.

I really like triangular fins. I know they aren't as efficient as clipped deltas, but you have to build what you like. As I mentioned a few posts above, I want to make a fin with substantial thickness, so I'm going to make a hollow composite double wedge with 1.5 inches of thickness at the root. Something like this:


First step of making a mold is creating a plug. The plug is exactly the shape of the parts I want to make, but can be made out of any material that is easy to shape. I used balsa, 3 oz fiberglass cloth, bondo filler, and high build primer:


I sanded the two surfaces flat by sticking sand paper to a granite surface plate, and sanding until the surfaces were completely flat. I laminated the balsa with a single layer of 3 oz cloth. Once that cured, I leveled the surface with bondo and more sanding on the granite.

I hot glued one of the plug halves to some masonite board, and added masonite strips, 2 inches high as walls around the perimeter. This wall adds quite a bit of stiffness to the mold, and allows it to sit flat, later when I use it to mold up my part. Unfortunately, I don't have pictures of this step, but it is very similar to the nose cone mold process. Three wax coats and a PVA layer go down. After drying for 30 minutes, I applied a thin layer of epoxy thickened with cabosil. The cabosil is a resilient surface for a mold (try sanding epoxy with cabosil - it sucks), and it also allows me to smooth out any abrupt angles on the plug. Fiber glass doesn't do well with tight bends. I find it likes to pull away from my part and create voids. Once the cabosil layer is tacky, I put down fiberglass and unmodified laminating epoxy in the following schedule: 1 layer of 1.5 oz cloth, 1 layer of 6 oz, 2 layers of 18 oz, and 1 layer of 6 oz cloth. The fist mold half looked like this:


Again, leave the first half of the plug in the mold. Once it is removed, it doesn't go back nearly as well as it is when the mold if first made. I hot glued the second half of the plug on top of the first half.

With a hollow plug, it is important to try and seal of the two plugs to each other as well as possible, or laminating epoxy will pool inside, leaving a void in the top mold half. I used a lot of hot glue and plastic modeling clay. I also let the first coat of epoxy and cabosil get leathery before adding the fiber glass layers.

I reused the 2 inch high masonite fence strips and hot glued them in place, sticking up about 1.5 inches above the first mold half. Modeling clay sealed and rounded the fence corners. Repeat the waxing, PVA and composite work, and you get two mold halves.

The cloudy surface of the mold is from cabosil. I'm glad this turned out, because the plug didn't survive the demolding step.

 

[rocketcharlie]

Joey. Will you say more about your choice of fin design? Is strength a factor? Do you feel a thinner fin would function as well? The design does look cool I have to say '

 

[mbeels]

It also has a water droplet paint job.

Whoa, so cool. I'm learning a lot from your posts, you have some interesting techniques!

 

[JLebow]

Joey. Will you say more about your choice of fin design? Is strength a factor? Do you feel a thinner fin would function as well? The design does look cool I have to say '

A flat fin of G10 material is a proven design choice, by all of the kits that are sold and fly, to be strong enough to do the flight I am planning.

The #1 reason I went with a fin that is 1.5 inch thick at the widest part of the root edge is because I like the way it looks. A thinner fin (1/2 inch at the widest point on the root, pictured in post 3 above) worked well on my 54 mm rocket. The stiffness is quite a bit higher than a 3/32" sheet of G10 (5x less deflection for a given force is my wild unsubstantiated guess - it is very stiff). It didn't "look" as dramatic on the rocket as I had hoped, so I'm trying for a more extreme thickness on this rocket.

The trade off for thickness is between stiffness and drag. I will post the process soon, but my fins made with these fin molds weigh about 3 oz (88 g) each, or about the same as a flat fin made from 3/32" thick G10. From my RasAero sim, I'm losing about 3,000 feet of altitude by choosing this thick fin vs G10 with a bevel. But 21,000 feet will still be a decent flight.

So no hard engineering in the decision, just a desire to experiment with composite molds and fin construction.

 

[JLebow]

This is part 1 of making the fins from the molds.

Cut the fabric. The tape does waste some of the precious carbon fiber cloth, but this 3k 2x2 twill more readily unravels than 6 oz fiberglass. Using the paper pattern, I tape out an outline of the piece, and cut the masking tape into two. I laid this pattern out in two different orientations (0 and 90 degrees, and +45 and -45 degrees) I need 5 layers per fin half, or 30 of these triangle pieces to complete the 3 fins.

Being organized is a key to composite work. Once the epoxy is mixed, the clock starts, and you don't want to waste time completing tasked that should have been completed during prep. The final product suffers. In the picture you see a layer of 1.5 oz fiberglass cloth. The intent was to have a veil layer on the fin surface, because sanding into carbon fiber weakens it and mars the visual impact that twill carbon for which carbon is famous. The glass did not wet out invisible in what turned out to be a test piece, so I decided to leave it out of the final layup schedule and just sand very carefully.

Here we have the 5 layers of carbon, wet out in the mold with about 3/4" of extra margin spilling out on the leading and trailing edge. I found that briefly blowing a hot air gun after placing each layer popped a lot of bubbles, and reduced the thin epoxy even more, completely wetting out the cloth. This was helpful to avoid bubbles, voids and pinholes. Carbon doesn't have the neat property of fiberglass where it turn translucent letting you know it is saturated. This part has a few spars of balsa as a core material to stiffen the fin. In the end I switched to a foam core (details in a future post).

Next a perforated release film is placed on top of the part. Excess epoxy is pulled from the part through tiny holes on a 0.5 x 0.5 inch grid.

Then a breather (allows vacuum to be distributed to all areas of the part) and bleeder (absorbs excess epoxy) cloth is added. This looks like quilt insulation (batting).

The prepared mold is placed in a vacuum bag, and a vacuum is pulled. My vacuum rig has a pressure switch that maintains pressure at 10 to 12 psi. The vacuum pump runs about 15 seconds every 30 minutes, which is an indication of a fairly small leak rate in my bag and pump system. You can see the excess epoxy seeping out quite a bit around the edges, and minimally in the middle of the part.

Here is a tease for the final part. This is the inside of the fin.

 

[g.pitts]

Here is a tease for the final part. This is the inside of the fin.
View attachment 403793

Looks great, Joey!

 

[JLebow]

Looks great, Joey!

Hey thanks. There's a good chance I saw your L1 flight. From your signature, it looks like you flew at the Rocketober launch in Brothers.

I wasn't happy with the result of the first attempt. The fiberglass veil was too prominent, the added bracing from sandwiched balsa was incredible strong in one direction but there was still quite some flex in the other direction, and the outline of the balsa spar printed though to the exterior of the fin. It is difficult for me to photograph carbon fiber, but the result is quite a bit worse (to my eye) than what shows in this picture.


So I replaced the balsa with 1/4" foam board. I did bevel the edge so that there are no 90 degree bends that really cause problems with bubbles and lift up for composites. The core is cut with a 3/4" margin around the fin edges.

Here is the inside of the fin after bagging and a cure.

To deal with the exterior surface quality, I got rid of the 1.5 oz fiberglass (used as a surface veil), and I did the layup in two steps. I layed up three layers of the carbon, put it in the bag and let the fin cure to the leather stage. This is the same shell technique I described in the post early about making the nose cone. Then I unbagged the mold, and added the core and two more layers of carbon cloth. It was an extra step, but worth the effort.

This is a much better surface than when I tried to vacuum bag my fins with no mold for my 54 mm version of this rocket (post 3 above).

Now I get to remove the mold flash from 6 fin halves. I used a dremel with carbide grinding wheel. The job was not difficult. I left a 1 mm margin and sanded up to my leading and trailing edge. But the job is incredibly dirty. I had gloves, what turned out to be a disposable sweatshirt, respirator, hearing protection, and glasses. I will see about rigging a cardboard box with shop vac and a tyvek body suit with hood next time. This dust is not good.

 

[JLebow]

I fixed some 120 grit to a granite surface plate and did some light clean up of the fin halves. Mostly I focused on the edges, and there was one imperfection in the mold that needed the exact same touch up on the three fins.

I washed the parts, and set them in the molds. A bit of masking tape was enough to hold the parts into the mold. They key in really well.

I mixed up 10 grams of rocketpoxy, and put it into one of those bags used for decorating cakes with icing.

A small bead of adhesive was squeezed into the leading and trailing edges.

Carefully, I joined the mold halves together and let the epoxy cure. I'm quite pleased with the results.

 

[g.pitts]

I fixed some 120 grit to a granite surface plate and did some light clean up of the fin halves. Mostly I focused on the edges, and there was one imperfection in the mold that needed the exact same touch up on the three fins.
View attachment 403851
I washed the parts, and set them in the molds. A bit of masking tape was enough to hold the parts into the mold. They key in really well.
View attachment 403852
I mixed up 10 grams of rocketpoxy, and put it into one of those bags used for decorating cakes with icing.
View attachment 403853
A small bead of adhesive was squeezed into the leading and trailing edges.
View attachment 403854
Carefully, I joined the mold halves together and let the epoxy cure. I'm quite pleased with the results.
View attachment 403855 View attachment 403856 View attachment 403857

You make it look easy! I think I will stick with G10. [emoji23] Are you thinking of any particular launch this year? I’d love to see her in person.

Yes, I launched on Friday at Rocketober last year (only had Friday of that weekend). That was after my first attempt at Fillible’s Follies where a defective altimeter spoiled things and left me draped with chutes and a nose cone just after powering on the electronics. Mr. Murphy ensured that things worked well during all of the ground testing, and delivered his little surprise to me on launch day.

 

[rharshberger]

Looking good so far Joey. Hoping to see this beastie live.

 

[Ez2cDave]

This project is coming along nicely . . . HOWEVER, this pic made me think of a Science Fiction movie - LOL !

Dave F.

 

[William Ellick]

Better science fiction than for a funeral!

I agree 100% with the Tyvek full suit with this stuff. I have worked with fiberglass and epoxy for over 35 years now off and on on a lot of projects (I use epoxy to fix just about everything ). I use a full bodysuit with a full face forced air face mask when I have to do a lot of sanding. I can only imagine how nasty carbon fiber is to sand. Yuch!

Your fins look great BTW. I can't wait to see the finished project. Happy trails with the flying.

 

[JLebow]

Better science fiction than for a funeral!

I agree 100% with the Tyvek full suit with this stuff. I have worked with fiberglass and epoxy for over 35 years now off and on on a lot of projects (I use epoxy to fix just about everything ). I use a full bodysuit with a full face forced air face mask when I have to do a lot of sanding. I can only imagine how nasty carbon fiber is to sand. Yuch!

Your fins look great BTW. I can't wait to see the finished project. Happy trails with the flying.

I picked up a tyvek suit. It was less than $10. No reason not to get it. The carbon fiber composite parts sand more easily than fiberglass (removal rate is higher with same grit sandpaper. The dust is finer and drifts further in the air. Both can embed in your skin and make you itch like crazy. Taking a hot shower makes the itching worse. The dust is bad news. Don't want to think of what it does to the lungs. Use protective gear, for sure.

I'm really happy with how my airframe turned out. Aside from a few pinholes, there will be very minimal sanding and prep required to get a really good surface.

This rocket is going up on an M1350, which has a 76 mm diameter. Turns out this is just a nominal dimension. The measured dimension actually closer to 75.3 mm, or a few thousandths of an inch less than 3". If you order a length of 3" aluminum tube, it is the same diameter as the motor casing, and this is perfect. Tube is specified by the outside diameter and pipe is specified by the inside diameter. I found that out while shopping for a mandrel. I ordered a 6 foot length of 3"aluminum tube with 1/4" wall thickness for around $60. I spent about an hour or two finishing the surface. You want the diameter to be uniform to within 1 or 2 thousandths of an inch (~50 um). If the diameter varies more, then it is OK if the tube is tapered uniformly across the length by up to 0.008" (200 um), but you definitely don't want an hourglass profile or pinched ends with a bulge in the middle. The removal in those cases is destructive to your part and can damage your mandrel and mood. Once the dimensions were corrected using 80 grit, I sanded out the scratches by working through grits 120, 200, 400, 800,1200, 2000, and then an automotive swirl removing polishing cream. Overkill, I know. But like any hobby, it is more fun when you over do it.

I set the mandrel up on wood blocks, and mounted an empty cardboard tube with 1/2 electrical conduit (use what ever is laying about the shop) behind the mandrel to spool up and feed out the carbon cloth. I did wax the mandrel with mold release wax. It shouldn't be required, but I did it anyway. Then I wrapped the mandrel with two full wraps of 0.003" clear mylar. This gives a really good fit between my motor and the airframe. Mylar doesn't bond with epoxy. When I made my first composite tube, I wrapped the mylar very tightly to the madrel. That tube never came off the madrel (in one piece). Now I wrap the mylar so that when I twist the mylar in the direction it is wound, it twists about a 1/4 turn before binding up, and will twist freely in the direction against the winding direction.

I get about 0.3 mm of thickness per wrap when I make a tube with 3k, 2x2 twill, 5.8 oz carbon fiber. To get an airframe with 1.5 mm wall thickness, I will need a 5 layer wrap, and this works out to a cloth length of 47". The cloth was cut and rolled onto my spare cardboard tube/ fixture. I also cut a piece of peel ply 18" long, to wrap the tube with two layers. I found that two layers holds the shape well while curing, and using less tends to have the layers relax on me with nasty wrinkles all over the airframe.

Fiberglass turns clear when it wets out. Carbon fiber isn't as cooperative, so it is important to pay attention to where you you have and have not yet painted on your epoxy to avoid voids and poor lamination. I like to put a layer of epoxy on the bare mylar before any cloth is placed, and then carefully position the leading edge of my fabric onto the madrel. The straighter you can align the fabric at the start, the less you have to fight the cloth to keep it straight as you roll it up. All the fighting and tugging must impart uneven tension in the cloth that starts to relax before the epoxy cures and causes an abundance of wrinkles. That's my impression of the process, at least.

I spread out some epoxy with a 1 inch disposable chip brush on the top of the mandrel. I then hit the new epoxy briefly (5 seconds max) with a heat gun. I don't want to heat the epoxy too much. I see a ton of tiny bubbles pop during this procedure, and the epoxy really starts to flow into the weave of the cloth. I turn the mandrel a quarter turn to bring new cloth to the top of the mandrel. I use the brush to slightly drag tension down and to the front of the mandrel. I don't want an overly tight wrap, just snug, aligned well, with no loose wrinkles or bubbles. Trying too hard here will cause as many problems as not spending enough attention. Repeat the process - apply epoxy, heat gun, brush the cloth lightly, rotate a 1/4 turn - until you get to the end of the cloth.

When I get to the end of the cloth, the tendency is for the cut edge to unravel. I pull one of the fibers out, a full fiber running the long length on the mandrel. This provides a nice guide to make a clean up cut on an otherwise ragged edge. I make sure the cloth is wet, then I then pull two more neighboring long fibers. Working the cloth with the brush at this point will cause a mess and a really ugly seam. I place the peel ply over the seam, and gently let the peel ply layer stick the frayed edge of the carbon cloth down to the tube. If you do this correctly, the seam of the tube becomes invisible. If you do it wrong, you have to paint a stripe on your rocket to hide the mess.

Here is the tube wrapped up in peel ply, ready for a cure at 85 deg F in my curing oven.


This is the ebay end of the tube. The peel ply leaves a woven texture that is perfect for bonding fins. The peel ply will stay on until I'm ready to bond the fins. The texture will be filled by light sanding with 200 grit sand paper, and using a rag to wipe a thin layer of epoxy across the surface.


Ok so maybe the seam is visible, but it doesn't look too bad to me, and it is better than my previous rocket.


I found the write up that Jim Jarvis compiled on composite techniques to be incredibly helpful. I borrowed a lot of his ideas, but still have a long way to go on attaining the quality of his results.

 

[Ez2cDave]

Pardon a dumb question . . .

Does anyone wet-sand their composites to prevent large amounts of dust and airborne particles ?

Dave F.

 

[JimJarvis50]

You might like the second article also, particularly about pulling tension while rolling the tube.

https://www.raketenmodellbau.org/repository/archive/167793?view=true

One trick that I never published was to use the tip of a chip brush (tap, tap, tap) on top of the peel ply. The pinholes vanish. Not sure this would work as well with two layers of peel ply.

Jim