Boost: Nice, straight boost. Up Part looked real good.

Recovery: Drogue came out, but the Main did not fully deploy. It ended up landing with the main still folded in half, and still crumpled up.

So, Pato died a horrible death. When all was said and done, Pato was in 4 equal-length sections. R.I.P.

Despite my desire for the interesting, the boring concept of a 3FNC rocket kind of stuck on me, especially in this case. Be it the emotional connection to my L1 rocket, or whatever, I just kind of like the design, and shape of the rocket. The nose cone never fails to impress people that see it, and everyone tells me what a nice job I did on the paint, and such. Things like that are a bit difficult to discount when heard all the time. I like high performance, and I like high and fast, but I also like scale stuff simply because of the challenge. When it came time to start making another rocket, I figured why not remake Pato, and do one better?

The design for Pato was very simple, but there were a couple of things about it that I didn't like. I had few complaints, other than perhaps the durability of the Maple nose cone, and it's uncanny ability to stick the landing wherever it would land. I really liked the shape of the fins, as I thought that while very simple, and to the point, they had just enough flare to them to remain interesting, and seemed to be stout enough to handle just about anything. Prior to it's demise, I got two offers to put a J570 in Pato, just to see what it could do, but I never actually did. I like the 54mm airframe and 38mm motor tube combination, mostly because it gives you just a bit of through the wall structure, which makes a difference, and I liked the boat tail, and the simple, but robust motor retention that I used, but I didn't like it's weight. I was a bit disappointed with the outcome of my altimeter bay, though I think my design had a bit of merit, even if it's execution was not all that it could be. So, I figured that I could rebuild the same design, using new materials, different techniques, and incorporate the best of the old design, while trying a different approach to those things about the old design that I didn't like quite as much.

Few people actually know that the result described herein, is actually Pato III, not Pato II. Prior to the June RocStock, I had done a lot of work, trying to re-build Pato as a 54mm airframe carbon fiber rocket, but I wasn't liking the results enough to want to assemble anything. There were several 'innovations' that I used in the fabrication of parts for that iteration that I later improved upon for this design, but between then and now, Pato had been completely re-designed. I initially used AutoCAD for design, and only designed a couple of sections in 3D. Several of these portions came out with fatal design-flaws, like running a piece of all-thread through the power and switch connectors on my ARTS, or not being able to assemble things because of how they were formed. Incorrect sizes, and the like ran throughout, and I was not happy with the design at all.

So, with Solidworks in-hand, I started from scratch. I had an idea to try to fit everything into a 2.00" ID airframe, because at the time, I was having a bit of difficulty with finding the right size mandrel, and I wasn't liking using full-length couplers to form composite tubes on. With this in mind, the only limiting factor was the altimeter bay. I wanted redundant altimeters, batteries, and switches, and while I wanted at least 2 calibers on either end of the coupler, I didn't want to have to make it 15" long to suit the contents. Again, the altimeter is stacked just on the other side of a central allthread or eye-bolt from the battery. It took a bit of magic, but I finally had it. A 2.00" OD coupler, with an extra layer of carbon fiber, an airframe-sized length in the middle for breather and switch holes, and 5" of coupler on either end. Both altimeters would be mounted to a rigid sled, and then slid into the coupler, and a cap would be screwed on using the allthread that ran the coupler's length. Charge connectors would be sealed from inside the coupler, and an eye-bolt would extend their 'eyes' out of either end. Now that I had that done, I could move on to the rest of the rocket.

I didn't have much luck with the last method I tried for forming the nose cone, so I figured I'd try something I hadn't done before. Because I didn't like the idea of having the entire rocket being made of carbon fiber, and the nose cone being made of anything but, I needed to figure out a way to mold carbon fiber to the shape I was looking for. This proved to be a challenge, but I'll get to that later. I still wanted the long conical cone, with a straight shoulder, but because I intended on being able to fly larger motors, and the last flight of Pato was only marginally stable, I devised a method where in, I can change the CG of the rocket, by adding or subtracting weights in the nose cone, that are firmly attached to a 1/4-20 piece of all-thread hidden in the cone. Based on my initial calculations, this would provide for as much as 4lbs of nose weight to be added if necessary.

Another aspect that I figured I could heavily improve was my fins. Around the time I was designing Pato III (a.k.a. - 'Pato's Revenge') there was a video traveling around, depicting the dreaded flutter that so many people complain about. A really amazing video, and quite a scare if you are a fan of G10, which I am now not. I knew there had to be a better way, so I did some searching, and finally decided on a full composite construction, consisting of a 2-layer core of carbon fiber, with .060" honeycomb on either side, capped on the outer sides with a single layer of carbon fiber, and on the edges with something I could put a bevel in. I thought, and purchased rohacell, which is a high-density aerospace foam, which I still think would be fantastic, but cutting it was a pain (maybe next time?) and finally settled on pine, cut to the right thickness from a piece of 2x2 from Home Depot. After the bevels, a final wrap of CF goes around the outside, and the assembly is trimmed to shape, and installed. After fabbing the fins for Pato II, which I was not happy with, I weighed them vs the G10 that I used on Pato, and they were about 1/3 the weight. I weighed the fins for Pato III, and they were half the weight of the G10, but twice the stiffness of the fins for Pato II. I'll take it!

Motor retention was another thing that I wanted to lighten up, so I formed my motor mount tube with the same steps that you would find in an assembly of conventional parts, with a slimline boat-tail retainer on it. The same groove, etc was used, and I even used the same snap-ring. This helps to lighten up the back end quite a bit, which then helps to lighten up the front end, because I don't have to balance out that extra weight. Sounds like a win-win to me.

Now, with the design stuff done, the task of making the drawings into parts calls me, and time stands watching over my shoulder with an intimidating scowl. The clock is ticking, and I'm sitting here, watching epoxy cure.

Fins:The original G10 fins were a bit heavy, at 88 grams each. The composite fins that I made for this round turned out to be roughly 43g each. That, in my book is significant weight savings, and more than 135 grams that I don't have to balance out in the front of the rocket, which reduces the total weight by at least 270 grams. That's nearly half a pound! You see how that works? Yeah, good stuff. The first method I tried to use to attach the honeycomb to the panels is by using a veil layer, saturated with epoxy to provide a bond line between the panel and the honeycomb. The *real* way to do this is with a thermosetting adhesive sheet, which you cure in an oven, under pressure for a period of time, and it actually melts into a thin layer of epoxy, and forms fantastic little fillets around every wall of the honeycomb, if you do it right. The problem with this technique is that the thermoset adhesive is not only expensive, but it also has a storage shelf life, and I need to use adhesives for the flat panels within this assembly that can tolerate the elevated temperatures and durations of the curing cycle of the thermoset adhesive without reaching the glass transition temperature of the epoxy, turning the whole thing to mush. While I would like to try to make a couple of panels that way, I opted to just paint the bonding faces of the CF panels with epoxy, and then throw everything in a vacuum bag. The fins came out a bit heavier than the previous run, but I think that because of that excess epoxy, they actually got the fillets that I was looking for, and turned out stronger than the first try, and with a post-cure, they became rock-solid. As a rough guess, I would equate their strength to an equivalent shape of 1/2" Baltic Birch plywood, and at only .210" thick, and weighing less than half of the same shape in birch, you get the aerodynamics advantage of a thinner panel, the weight savings of composite materials, and the same strength.

Motor Retention: The boat-tail in the original Pato was one of the heaviest single components in the assembly, and was stuck right on the aft-end of the rocket. The retainer has a very sound method of retention, using a groove into which a snap ring is inserted that retains the motor. This, I think is a fantastic design, and is very positive retention for ejection. I really like this system, but I can't stand the extra weight on the rear-end. Just the retainer weighs more than the lower composite airframe, and all of that weight is concentrated in the last inch and a half of length on the rocket! The motor mount tube will have two inside diameter changes in it, and the composite boat tail will be formed to match the lower airframe, and the outside of the motor mount tube. A final piece will be epoxied into place to create the groove, and once assembled and epoxied together, with the centering rings, they will be trimmed to an appropriate length, making the final assembly. The function of the retainer will be exactly the same as the aluminum version, but will be significantly lighter, and will be a very stout assembly of composite parts.

Altimeter Bay: If one is good, then two is better, right? Okay, well, maybe not. But redundant electronics is an easy way to make sure that it comes back. Pato had an alt bay for only one altimeter, with limited choices due to my arrangement, and sizing of the bay. I have since then, acquired an Ozark ARTS, which I would like to fly in Pato II as a primary, with the original Perfectflite MAWD as a backup. Once I am confident in the ARTS, then I can fly that solo, or keep the perfectflite in it for redundant deployment, and data collection. Among the changes to purely composite construction in Pato II, and since my failure to meet the RocStock deadline, and my tragic design flaw in my altimeter bay, which ran a piece of allthread through the connectors on the ARTS, and sticks them through the airframe, I decided to make matters worse, by trying to make the same volume of crap fit in a 2" ID tube. Well, I succeeded. It is a very tight fit, but the assistance of laser-cut parts, and Solidworks 3D modeling (which saved me several times) I made sure that I wasn't overlooking anything. So, with 2 altimeters, 2 batteries, 2 power switches, 2 sets of connectors, standoffs, screws, et al. that I am able to fit in the 2" OD coupler, thanks in whole to the laser-cut parts, Pato II will fly it's 2" ID airframe with two altimeters in a coupler that is 11" long. If I may say so, it really is a work of art.

Nose Cone: The Maple nose cone for Pato was fantastic, but not without it's own design, and construction challenges. Going to a full composite airframe, and fins, I just couldn't stand to not do the nose cone, and boat tail as well. The first design of Pato II called for the boat-tail to be formed as the aft-end of the lower airframe, which turned out to be a MAJOR pain in the rear (*rimshot*). So, in the re-design between RocStock, and Plaster Blaster, I opted to make the boat tail the same way I decided to make the nose cone: With fiberglass molds.

Now, this is much easier said than done. There are a couple of resources for it on the internet, including the Vastaas site, and John Coker's ASP nose cone, but none of these were able to prepare me for what was to come. In general terms, I understood the process, and grasped the concepts involved, but like one of my Rocket Buddies' sudden strike of fear once he begins to mix epoxy, when you're knee-deep in it, all you can do is 'wing it' when you come across something that is not what you had expected. The process is fairly straight-forward: Make a plug. I opted to turn mine from wood (Poplar, as a less-expensive alternative to my favored Maple), Finish the plug. Sanding sealer, I would suggest. Polyurethane gloss clear coat, I would not. Wax the plug. Meguiar's No 88. Cut a parting board. Basically, a hole in the middle of a piece of wood that matches the profile of your plug. Use clay to seal plug to parting board. Wax everything again. Support your plug and parting board, I very highly recommend it. And if there is any doubt, wax everything again. I used what is called PVA as a release layer between my plug, and my mold, which worked like a charm. Apply gel coat, add reinforcement, add glass mat for bulk, and let it all cure. remove parting board, and wax everything again. More PVA. Gel-coat the other side, more reinforcement, and more glass mat for bulk. Once it cures, drill some bolt/alignment holes, and pry it apart. Wax and release your mold face, lay up your cloth in it, and bag it. At the green-cured stage, de-bag, trim your parts, join your molds, and add reinforcement along the seams with a bit more epoxy, and cure the whole mess. Sound easy, right? Keep reading.

I tried to turn the nose cone, and tail cone for Pato II by hand, but it turns out that I kind of suck with the straight lines thing on a lathe. One of the issues, is that as you get close to finishing, the tip of the nose cone becomes really quite brittle, and you can't take off as much material with each pass as you would like, so you get a bit of a bulge where this takes place. So, I looked into finding some sort of duplicating, or copying tool for my Dad's wood lathe. After a long, drawn-out search, I managed to find something that works fairly well, and it has it's own quirks, but it gets the job done. I made templates for all of my plug parts, and continued with my processing. It was right about this time, that I figured out that the GBU was going to be taking up about half of my processing time, so, I decided to shelve it, but not before making the plugs for it (which, again, I would have done differently).

I sealed my plugs, and covered the four that I was going to use for this round of rockets with what I believed to be a decent clear-coat for this sort of thing, and either it wasn't fully cured, or it just doesn't suit the job, but much of the clear-coat ended up on the mold face when I removed the plug, instead of still attached to the plug. Obviously not my intention. I applied several coats of wax to this clear coat, and to my parting board, and then fit the parting boards to my plugs, using non-drying clay. This in itself, I don't believe to be a mistake, but perhaps it stemmed from the loose fit of the parting boards, and my blind confidence in the non-drying clay to resist the 'stippling' of a brush, making sure that there was resin everywhere, but I would make darn certain that my parting board were as close as humanly possible to the same profile that the plug is, and use a minimal amount of that clay, only to touch-up, not create a mold surface. My parting line in my molds turned out to be just a little bit crooked, which for the most-part is alright, but I wouldn't use them for any kind of production. The molds also lacked reinforcement in some of the fine corners, like the shoulder, and the change from the round part to the flat of the parting line. I would definately change this, were I able to do it all over again. The first mold halves that I made turned out like crap, but I had learned a bit by the time I got to the second halves, so those turned out quite a bit better, but there are still things that I would do differently.

Now that I have molds, it's time to mold parts. My first try (like many things I do), I bit off more than I could chew, by trying to lay-up, and bag 8 mold halves in the same vacuum bag. It's not that the volume was too large, but rather, your attention is spread too thinly among 8 parts, when you are drawing a vacuum, and sealing the bag, and making sure that everything on the inside of the bag is sitting just right, and just the way you want it, that things move without you knowing it, or how it will affect your final part's outcome. Let's just say that this wasn't going to be the only time I made these parts before the rockets flew.

After that expensive mistake, I was just about ready to throw in the towel, when my Dad helped me realize that I had simply fallen victim to the learning curve, and that my frustration, while warranted, wasn't the end of the world, and the only thing I had to do to solve that, was to examine my processes, and my outcomes, and envision methods for improving one or the other, and flesh out the details. I managed to identify no less than 5 things that went wrong with the parts, and 5 things that I would do quite a bit differently on the molds next time to help ensure a better outcome. So, with that information in hand, I give it another shot, and it turned out MUCH better. Shoulders were defined, and the steps to the outer airframe diameter looked like steps, and not slides. I actually had fabric in the tip of the cone for Pato, despite it being only .125" dia at the tip, and is in fact purely carbon tows aligned with the long-axis of the cone, making the tip nearly all carbon. It's little things like that that make the big difference.