Bare Necessities: N5800 C-Star Flying Case

(First off-I decided to create my own account finally instead of using MuddAmateurRocketryClub's general account. I am the current president of MARC, and did this project under their auspices with CarVac.)

This is the "Mudd Team"'s entry into the N5800CS competition. We are flying ours at XPRS (pending no tragedies in the remaining construction process). It was designed to be such that we could build it ourselves (well, by myself mostly, but that's a different story) at Harvey Mudd College over the summer. It uses as many in-house-fabricated parts as possible, as we prefer working with our own processes and materials, especially given the ease with which the N5800 has shrugged off commercially-made nosecones/body tubes/fincans in the past. We are rising Seniors who will be graduating with bachelor's degrees in the spring; CarVac is an engineering major, and I am a physics major. This project represents what we are capable of in terms of design, analytical modeling, engineering judgement, and fabrication talent.

Starting at the bottom: a carbon fiber tailcone made with PTM&W 5712 high-temp high-strength epoxy that will snug over the thrust ring, retaining itself against the end of the nozzle and potentially retain the fincan at the back of the tube.

Fincan: machined aluminum, no welds. Fins machined from .25" 7075-T651 plate, with 8 degree (included angle) bevels. 3.25" span. Fins are heat-shrink-fitted into slots in the fincan, which is machined from a 6061-T6 extruded tube. FEM modeling (including approximations of heating effects) have shown that the shrink-fit will be sufficient to retain the fins against both dynamic pressure loading (axial, primarily) and flutter loading (transverse). In addition, the fins are retained by pressed-in alloy steel 3.32" diameter dowel pins in the fin roots, which (by hand analytical calculations) should also be able to retain the fin against estimated loads, even if there were no interference fit. The fincan will be heat-shrunk onto the motor case, and will be tied in to the tailcone over the thrust ring. The thrust ring of the motor case, however, is the largest diameter on the length of the rocket; there is no material over it.

Forward Anchor: a milled and turned aluminum body press-fits on the forward closure of the motor case. It uses a separate press-fit centering and alignment collet that presses onto the delay well fo the forward closure, and it bolts into the retaining stud on the motor case. It is pocketed out and contains a pair of Raven2 flight computers, independent Li-Poly batteries, a keychain HD camera and it's Li-Poly battery, three Featherweight screw switches, a hardpoint to tie the shock cord to, and a hardpoint that a Rattworks ARRD threads down onto. An aluminum cap screws down onto the top of it, covering the Raven and camera bays, creating an EM-resistant enclosure.

Forward Airframe: the commercially manufactured parts. A 13" long section of Hawk Mountain fiberglass coupler tube, with an internal 29mm tube by the same. They are retained on the aluminum anchor by a veritable panoply of recessed flathead screws in addition to being press-fitted over long shoulders in the aluminum. Contains a fiberglass (G10) av-bay which holds the GPS receivers, a Beeline 70cm (uBlox SIRF III) and a Beeline 900MHz (Trimble Lassen Iq), both with onboard data logging, and their batteries, and a switch to activate them.

(have class now-will add more later. link: https://docs.google.com/document/d/1HYh3MNpeld5GMWMje7R8VcdizjjysCkcC63Ja76DTB0/edit

Very, very interesting. I'll be following this thread closely, since it is similar in several respects to my own design, and I will be very watching to see the flight results. Are the Raven2 altimeters using the stock firmware, and if so, do you have the version without the high altitude bug?

We are not sure. Your mention of it in your thread was the first time I ever saw that there was a firmware revision to fix it. We had initially planned to use timers for apogee, but we might use barometric if it will not experience the bug.

Can the firmware be downloaded somewhere?

CarVac said:

We are not sure. Your mention of it in your thread was the first time I ever saw that there was a firmware revision to fix it. We had initially planned to use timers for apogee, but we might use barometric if it will not experience the bug.

Can the firmware be downloaded somewhere?

Click to expand...

Get with Adrian -- he likely has a list of which units he has updated. If not, he can definitely upgrade the units for you.

-Kevin

Anyway, since we just finished the layup, I'll cover the nosecone next. It's not finished, but the layup is.

Since our rocket's "airframe" is Hawk Mountain coupler tube, our nosecone had to be custom-made to fit, since anything else would have an excessively large diameter.

Filament-wound seemed out of the question: we don't have the machinery to do that. Thus, we decided to lay up cloth in a mold.

Unlike most molds, which are created of fiberglass from positives, we decided to CNC profile ours out of solid blocks of aluminum for absolute rigidity and perfect registration using steel pins.

We also wanted to avoid the pitfalls of standard cloth-layup fiberglass nosecones by performing the layup differently: instead of splitting the mold lengthwise into two halves and laying them up separately, we segmented the mold into a forward part and an aft part, each of which was split into two 'conventional' halves for machineabliity. Thus, the seams between pieces of cloth all overlap, leaving no weak spots. Each additional layer of cloth in the forward section gets shorter and shorter, leaving a gentle lead-up on the back so that the aft portion of the layup could bond with all of the layers. All of the aft layers of cloth were the same length, but the outer layers were inserted deeper into the back of the mold so that they stopped in the corner right before the shoulder. Once that height was reached, the next layers forming the shoulder were laid up, preventing any kinks.

The actual material forming up the nosecone is thus: the outer layer is thin woven E-glass for surface finish. Inside of that, I laid up two layers of longitudinally unidirectional S-glass, then 4 or 5 layers of 20-ounce E-glass/basalt hybrid (basalt circumferential for hoop strength), and two more layers of uni S-glass.

The aft portion has actually 6 layers of the hybrid and something like 4 inner layers of S-glass. The E-glass, first two layers of uni S-glass, and the first three layers of basalt hybrid all stop at the corner before the shoulder, and then the outer surface of the shoulder was formed from the next three layers of basalt hybrid. Then, the 4 inner S-glass layers were laid up at varying positions, some deep inside to overlap the joint between the two portions of the layup and some to stiffen the full length of the shoulder.

This "rear insertion" layup is made possible by the fact that the composites start about 2/5 the way back from the tip of the nosecone: the rest of it is made of stainless steel and aluminum. Thus, the skinniest part of the layup is about 1.5 inches wide, which is manageable with tools (dowels). The aluminum part actually is bonded to the layup: it was held in the mold in position so that it would not get glued to the rest of the mold accidentally, but so that the front fibers were laying along a rough aluminum surface for bonding.

Hopefully this ridiculously thick layup is going to be capable of withstanding Mach 4 flight, but just in case, we may be reinforcing it with G10 ribs or foam or both.

This project looks brilliant! Great job guys!

Nice approach. Do you have any pics?

Nice design. If possible, I would replace the hawk tubing with aluminum tube. I've seen a LOT of 98 hawk tubes be the failure point on 98 MD projects with tamer motors. You obviously have machining capabilities so reproducing the ID/OD of the hawk tube in aluminum isn't too difficult and should help to make the rocket bulletproof. It's a short, thin tube so the weight penalty won't be killer.

Todd Harrison

CarVac said:

Anyway, since we just finished the layup, I'll cover the nosecone next. It's not finished, but the layup is.

Since our rocket's "airframe" is Hawk Mountain coupler tube, our nosecone had to be custom-made to fit, since anything else would have an excessively large diameter.

Filament-wound seemed out of the question: we don't have the machinery to do that. Thus, we decided to lay up cloth in a mold.

Unlike most molds, which are created of fiberglass from positives, we decided to CNC profile ours out of solid blocks of aluminum for absolute rigidity and perfect registration using steel pins.

We also wanted to avoid the pitfalls of standard cloth-layup fiberglass nosecones by performing the layup differently: instead of splitting the mold lengthwise into two halves and laying them up separately, we segmented the mold into a forward part and an aft part, each of which was split into two 'conventional' halves for machineabliity. Thus, the seams between pieces of cloth all overlap, leaving no weak spots. Each additional layer of cloth in the forward section gets shorter and shorter, leaving a gentle lead-up on the back so that the aft portion of the layup could bond with all of the layers. All of the aft layers of cloth were the same length, but the outer layers were inserted deeper into the back of the mold so that they stopped in the corner right before the shoulder. Once that height was reached, the next layers forming the shoulder were laid up, preventing any kinks.

The actual material forming up the nosecone is thus: the outer layer is thin woven E-glass for surface finish. Inside of that, I laid up two layers of longitudinally unidirectional S-glass, then 4 or 5 layers of 20-ounce E-glass/basalt hybrid (basalt circumferential for hoop strength), and two more layers of uni S-glass.

The aft portion has actually 6 layers of the hybrid and something like 4 inner layers of S-glass. The E-glass, first two layers of uni S-glass, and the first three layers of basalt hybrid all stop at the corner before the shoulder, and then the outer surface of the shoulder was formed from the next three layers of basalt hybrid. Then, the 4 inner S-glass layers were laid up at varying positions, some deep inside to overlap the joint between the two portions of the layup and some to stiffen the full length of the shoulder.

This "rear insertion" layup is made possible by the fact that the composites start about 2/5 the way back from the tip of the nosecone: the rest of it is made of stainless steel and aluminum. Thus, the skinniest part of the layup is about 1.5 inches wide, which is manageable with tools (dowels). The aluminum part actually is bonded to the layup: it was held in the mold in position so that it would not get glued to the rest of the mold accidentally, but so that the front fibers were laying along a rough aluminum surface for bonding.

Hopefully this ridiculously thick layup is going to be capable of withstanding Mach 4 flight, but just in case, we may be reinforcing it with G10 ribs or foam or both.

Click to expand...

We cannot use an aluminum tube instead of the HME tube because we need radio transparency for GPS lock for the Carmack prize. We have considered (and will now reconsider, on your suggestion) reinforcing the inside.

But there will also be a HME 29mm tube down the center of the rocket, adding an alternate load-bearing pathway from the tip of the nosecone straight to the anchor and the forward closure. It will be attached to the HME 3.9" coupler by some radial sheets of G10, so that will add some strength.

Our concern is RF-transparency for GPS reception; we don't want to replicate Derek Deville's problem with QU8K. Otherwise we'd have used CF in the tubing and the nosecone. Given time, we'd have made our own convolute-wrapped tube; I'd rather trust our own tube, and I know that HME tubing has a not-great reputation for high-stress flights. HME tubing dramatically outlasted Performance rocketry in our own high-temperature stress testing. I might start another thread for the inevitable debate about testing methods and reputations for commercially available tubes.

I should have mentioned, RASAero reliably says we will exceed 110,000ft, so we are a carmack prize contestant, elegible for the entire $10k prize. We announced a few weeks ago.

+1 for purpose-built design engineering-makes this old coot's heart happy! Altho' engineering is not subjective, i have a good feeling about this project. It sounds absolutely sterling. I'm really rootin' for you guys and hope all goes well.

If you have the time, I would like to see pics of the nose cone and the mold. It sounds like an interesting layup process.

Greg

CCotner said:

HME tubing dramatically outlasted Performance rocketry in our own high-temperature stress testing.

Click to expand...

AWESOME. Details would be appreciated -- after you guys fly is fine, y'all must be swamped with prep. But we need more stuff like this

Continuing: (note that there is a link in the google docs to images of the fincan construction).

As CarVac mentioned, the nose cone tip is steel. There is a 6"-ish long aluminum piece, which is bonded into the fiberglass nosecone shell. It has a cutout bored in the bottom which will mount a 29mm hawk mountain fiberglass airframe, which runs the length of the nosecone for carrying the axial load of the metal tip. The true tip is 6.5" of 17-4 stainless (from 0 to .62" diameter), and is solid. It is press-fitted into a 304 stainless mostly-solid cone, which is bored through with a .25" hole (so that the press-fitted tip can be removed with a long .25" steel rod and a hammer). This piece is about 8.3" long, expanding from the .62" of the 17-4 piece out to 1.5". The aluminum piece had a threaded stud onto which the SS nose mounts.

Apogee recovery is accomplished by a yet-to-be-determined trigger on the Raven altimeters. each altimeter and battery will fire an ematch located in the a CD3 device; potted in epoxy and using krytox-lubed orings. A 12-gram canister will fire inside the inner 29mm tube, forcing the nosecone off. A 24" rocketman drogue parachute (box-form, very low CD) will be forced out. At low altitude (2000 feet) the ravens will fire (again, independently) charges in the Rattworks ARRD, which will release, allowing the drogue to pull free the main parachute. The rocket will remain tethered together. The main parachute is a 60" Fruity Chutes Iris Ultra (I believe CarVac mentioned this before), which will give us a descent speed of approximately 22fps. It is black and neon green for visibility, the two colors we have decided are most visible in desert launches based on our experiences launching with the Rocketry Organization of California in Lucerne Valley. A backup (read: hefty) BP charge will be set off underneath anything that remains inside the rocket at 1000'.

WOW Very cool, Subscribed.

daveyfire said:

AWESOME. Details would be appreciated -- after you guys fly is fine, y'all must be swamped with prep. But we need more stuff like this

Click to expand...

Our stress testing consisted of pointing a heat gun at samples of the tubing. I did not personally witness most of the testing, but the high temp epoxy we are using only softens (in a really thin layup that starts out flexible) whereas the commercial tubing distorts easily and distorts badly.

Also, while we wait for the nosecone to initial cure we have plenty of time for documentation.

The NC being the same OD as the motor case is very similar to one I sent Bryce (Bandman44) a few months ago. Unfortunately, I never heard back.
It differs slightly but the concept is the same.
Your design:


A quick sketch of mine.


Very interesting project. I cant wait to see how this turns out. Make SURE to have a video camera running!

Alex

This whole MD5800 contest is getting better and better -- reminds me of Senior group projects to achieve my undergrad in Engineering -- subscribed!

BTW, the fin can sounds exactly like the design I put out for bids last week -- I'm way behind

I NEED PICTURES!!!! Extremely interested in this thread

MasonH said:

I NEED PICTURES!!!! Extremely interested in this thread

Click to expand...

Click on the link in the first post. There's a few pictures in there, as well as a link to more pictures.

Very nice approach... so very pleased to see the engineering minds working this out, thanks for sharing...

So I don't know if CCotner is about to post the same thing as me but with more photos, but we demolded the nosecone to find that it is structurally very sound, but the forward half of the layup had the outer layers not attached to the basalt hybrid cloth underneath. We have cut out about half of it, in preparation for building up the low spot with more fabric and filling the cracks with microballoons.

What do you people think of that? The nosecone is super, super strong, just it needs a refresh of the outer layer in patches.

Found a pics link:

https://www.facebook.com/media/set/?set=a.4056301220188.155137.1661827772&type=3&l=87d00fbb26

Greg

This project makes me think of one word to describe it- WOOF. Great work...

The only thing I haven't seen mentioned is if there is concern necessary about the g-loading on the electronics, connections, devices, etc. It would be a terrible shame to have this survive the up only to bury itself several feet underground afterwards because something moved/broke during the burn. I imagine the g-load will be pretty impressive? The raven has that huge cap just sort of laying there.....

N

CarVac said:

What do you people think of that? The nosecone is super, super strong, just it needs a refresh of the outer layer in patches.

Click to expand...

My first thought was "If it's just superficial and non-structural, then it's fine", but as I thought about it...

You're putting a lot of work into this, and this entire project is one of those things where a minor thing going wrong can ruin all the effort put into everything else. Do you want to risk the self-doubt of not making a new nosecone, and worrying about whether or not the current one has issues?

Considering the attention to detail on everything else, if it were me, I'd make a new nosecone.

-Kevin

I agree wholeheartedly with Kevin on this one. Attention to detail is everything. However, if we re-make a new nosecone, we'll have to either make it from carbon fiber (sacrificing everything possibly because of GPS problems), or order more of the e-glass/basalt hybrid from Soller composites-they took ages to ship to us in the first place. Which means we can't fly in the 2012 flying season at Black Rock (and remember there's a minimum 60 day if we file our own waiver under the university rules). We would be inelegible for the N5800CS record prize, which would in all likelyhood be claimed by Bryce or CJL anyway. We fly with this nosecone.

IF you were to try to fly this nosecone at Mach 4, how would you prep it?

We're going to cut approximations of the original shapes (very slightly smaller) of fabric that we removed, and use more of the same epoxy (both straight in the fabric and with microballoons everywhere else) to re-form the surface. We'll take very lightweight (.7oz/yd^2) fiberglass and lay it over the entire effected areas, overlapping the original, and then heat-shrink-tape cure the whole thing (room temperature cure after heating the tape). We'll then sand down the overlapped glass layers(but the currently exposed edges of laminate will still be covered by the thin glass), apply more microballoon layer to the entire surface and fill in any remaining imperfections, allow to cure, post-cure the assembly, and then finish sand.

Any and all thoughts are appreciated.

The max acceleration is frighteningly close to max-Q, right at burnout. It's under 70g, so it must be fine. I've had ravens survive bad crashes before (certainly more g's than that); the cap's get yanked around quite a bit but I haven't seen one come off or malfunction as a result of lakestaking from almost a mile. We don't expect the GPS to lock during boost anyways due to ITAR, so we aren't concerned about the acceleration affecting them. We'll make sure the mountings are g-resistant, shock-resistant, vibration-resistant, etc..

We did that, since no one replied quick enough. It went surprisingly well; we used a syringe to inject epoxy into pinholes we drilled in the outer layers where there were voids, and visually confirmed that the epoxy completely filled the voids. In sections where we cut off the outer layer (complete delamination), we filled in the difference with e-glass weave. We coated the entire outer surface of the nosecone with microballoon-filled epoxy, except for the repaired areas (which were too wet for the microballoon paste to stick to). Tomorrow we will sand the microballoon paste down, as well as any protruding bits of fiberglass repair-and tomorrow night we will re-coat the skin (all of it) with microballoons. We will sand it to the final finish and contour before post-curing on friday night.

Which GPS are you planning to use? Be aware that most MTK and Sirf based GPS units will not function at all above 60,000ft. A Venus chipset and some uBlox are the only ones i know of which will, through having actually putting them up that high (with balloons). A Venus638 will most likely get a lock within a few seconds of you leaving ITAR restrictions, I don't know of any other GPS that wont take several minutes to come out of ITAR.