"Contents Under Pressure" - Static Test Cato

Usually with an o-ring blow by the failure method isn't the case coming apart and exiting the area at high speed and in many pieces. Usually you see a leak, some gas, then you develop a blowtorch from that area and the burn slows down as you have a much lower Kn than you planned on. Things melt. Things don't go bang. My best guess is exponential Kn increase or the case wasn't rated to that pressure and gave up the ghost trying to contain it.

Edward

Your case definitely failed from over-pressuring. The end looks exactly like the two dozen or so bolted aluminum 38mm cases I've destroyed.

I would suggest at least 10, maybe 12 bolts for a much improved safety margin next time.

Even with the additional bolts I think you would have been in trouble. Since tapping the mandrel on the floor caused such extensive visible damage to part of the grain, I'm guessing there were lots of other internal cracks you just weren't aware of.

Aksrockets said:

That's a fantastic idea. I'll look into that, maybe do a few tests.

Thoughts?

Alex

Click to expand...

That is what Prophecy and I have done with a small composite case bonded test motor using stock PR 54mm tubing and coupler. Pins run through the motor case and the "coupler tubing," allowing for a better attachment. Also the last two composite case flights used the same approach, except the tubes were not stock by any means.

As far as the failure, I do not think it was blow by. I think AlphaHybrids is on to the right idea. I do have a question about the numbers you have for tiger tail, where did you get them? I have two different sources for the a and the n of tiger tail and they are different than your numbers. Just curious as I have never characterized tiger tail.

Would it be reasonable to assume the case failed because your fuel debonded from the phenolic liner when you broke it originally? Maybe your case design was just fine.

Eric

What was your Kn curve on the motor. I generally run tiger tail around 280-300. I've gone as high as 338 initially.

Edward

Aksrockets said:

Theres evidence to believe the failure was caused by an O-ring blow by. There's the fact that the forward closure failed (with 1.25 in of tubing in front of it) and not the nozzle carrier (with .5in of tubing in front of it).

Click to expand...

The pressure is higher at the forward end and the action of the nozzle itself reduces the loads at the nozzle carrier. Unless the nozzle carrier (or its attachment) is weaker, the forward closure is more likely to fail.

Reinhard

That is certainly a failure of the case. You need more retaining pin surface area to spread the load and help the relatively weak FWFG. You may also want to investigate the use of dowel pins in lieu of screws.

I think the best recommendation is the idea of gluing in a coupler retaining ring. If you're a belt and suspenders kind of guy, you could glue and pin the ring.

3stoogesrocketry said:

Would it be reasonable to assume the case failed because your fuel debonded from the phenolic liner when you broke it originally? Maybe your case design was just fine.

Eric

Click to expand...

We were able to recover most all propellant afterwards and showed no signs of debonding from the liner.

Clay

AlphaHybrids said:

What was your Kn curve on the motor. I generally run tiger tail around 280-300. I've gone as high as 338 initially.

Edward

Click to expand...

Kn ranged from 220 (initial) 301 (final) We were running pretty safe margins for this particular motor. However, we were not using tiger tail for this particular test.

Clay

OK guys. I don't want to rain on your parade, but if you want to make a composite casing that works, don't reinvent the wheel. Learn how the pros do it. DoD, NASA and the aerospace industry have spent billions learning how to do it right, so do some research of your own and find out how it's done.

0.) Motor casing just don't happen, they are designed and engineered. You need to do some mechanical and thermal calculations to see what properties and materials you need for your casing.

1.) You only use a composite case to save weight. Otherwise there is no advantage to them and they cost up to 10x as much as a metal casing.

2.) Professional composite casings are single use. They are not designed to be reused no matter what you have read here. The heat soak after the burn is likely to permanently change the mechanical properties and render the casing useless. You could make a stronger composite casing, but then you could have made it from metal and saved a bundle.

3.) Standard composite casings are no more resistant to heat than aluminum casings. Composite casings are insulated from the propellant grain by rubber or case bonding with unpolymerized rubber. You could make them thicker however both steel and titanium maintain their mechanical properties at much higher temperatures which is why they are used instead of composited in many SRM systems.

4.) Virtually all composite motor casings are wound on a 4(+) axis winding machine and many smaller composite casings are filament wound around the propellant grain and possibly around a nozzle assembly, or a flange to allow attachment of a nozzle assembly. They are then heat cured in an oven to a known temperature to provide characterized mechanical and thermal properties. Checkout https://www.xwinder.com/ for a $3000 4-axis winding machine.

5.) Flanges are not pinned or bolted to a composite tube. As you observed, you will get a shear failure at the pinning points due to stress concentration unless you reinforce the area which would make the casing heavier which defeats the reason why you used a composite casing. Composite casings and components are always bonded by resin which is heat cured to maintain mechanical properties of the composites or overwound with a filament winding machine.

[YOUTUBE]Gd7ZrMxyJ6U[/YOUTUBE]

[YOUTUBE]MxvUU7Ig1yI[/YOUTUBE]

I'm not saying you can't hand make a composite motor casing. I've done it, but you have to understand what you're doing and you're not there yet.

Bob

bobkrech said:

I'm not saying you can't hand make a composite motor casing. I've done it, but you have to understand what you're doing and you're not there yet.

Bob

Click to expand...

This has been summarized to me before as "Sometimes, you don't know what you don't know."

Edward

bobkrech said:

2.) Professional composite casings are single use. They are not designed to be reused no matter what you have read here. The heat soak after the burn is likely to permanently change the mechanical properties and render the casing useless. You could make a stronger composite casing, but then you could have made it from metal and saved a bundle.

Click to expand...

We are not reusing the casings! I think that's the third time I've said that.

You only use a composite case to save weight. Otherwise there is no advantage to them and they cost up to 10x as much as a metal casing.

Click to expand...

Thus our motivation for the project. I thought it was implied.

4.) Virtually all composite motor casings are wound on a 4(+) axis winding machine and many smaller composite casings are filament wound around the propellant grain and possibly around a nozzle assembly, or a flange to allow attachment of a nozzle assembly. They are then heat cured in an oven to a known temperature to provide characterized mechanical and thermal properties. Checkout https://www.xwinder.com/ for a $3000 4-axis winding machine.

Click to expand...

I'm all for it if you're buying.
Half the point of the project is to use commercial components, making it easier to get more altitude and a higher mass fraction with a smaller wallet.

5.) Flanges are not pinned or bolted to a composite tube. As you observed, you will get a shear failure at the pinning points due to stress concentration unless you reinforce the area which would make the casing heavier which defeats the reason why you used a composite casing. Composite casings and components are always bonded by resin which is heat cured to maintain mechanical properties of the composites or overwound with a filament winding machine.

Click to expand...

The reason we did it is because it's been done successfully before (Steve H's stuff).
If you've always been against pinning composite cases and knew it would fail, why didn't you bring it up before?

Care to give us any details of your composite cased projects?

Well, now we know what not to do. Back to the drawing board.

Alex

About a decade ago I had a great conversation with Nick of (now defunct) Aerosleeves.com. He was doing some composite cased motors and came upon a solution that was probably the best of both worlds and had a lot of documentation in the aerospace industry. He used a very thin aluminum case and then overwrapped that with composites. The aluminum acted as a gas tight vessel and gave a very nice sealing surface. The composite gave the hoop strength. He was running 3" OD cases with 0.049" thick walls and then an additional overwrap. They were single use and worked out nicely.

Like Bob said, don't re-invent the wheel. I've mentored a couple of university teams for different competitions and it is a similar theme. They are given previous work, test data and even a lot of industry research. They go out and re-invent the wheel and then they end up going down the same path and re-discovering all the same things that did and did not work. If they would have slowed down, took the previous experience into account then the outcomes would have been vastly different.

Edward

Aksrockets said:

We are not reusing the casings! I think that's the third time I've said that.

Click to expand...

I did not say you were. I reviewed the whole thread and several have advised that a composite casing could be reused and some amateurs have done this. They were lucky and had no skin in the game. It didn't matter if their motor blew up.

Thus our motivation for the project. I thought it was implied.

Click to expand...

Motivation is fine, but reality is lacking.

I'm all for it if you're buying.
Half the point of the project is to use commercial components, making it easier to get more altitude and a higher mass fraction with a smaller wallet.

Click to expand...

You premise is flawed. I'm not buying, and composite cases are far more expensive than a larger convention casing and more propellant...

The reason we did it is because it's been done successfully before (Steve H's stuff).
If you've always been against pinning composite cases and knew it would fail, why didn't you bring it up before?

Click to expand...

I don't care if an amateur has done something before. Steve was lucky, and did not make an optimized casing.

I am not your parent or mentor. (BTW - Where is your mentor, and are you listening to him?) You would not have listened to what I had to say so I let threads like this proceed as you will learn a lot more by failing than arguing.

Care to give us any details of your composite cased projects?

Click to expand...

No. I am restricted by ITAR regulations for giving the details. I can say the casing were 8 layer tubes made from 12" wide uni-cloth with 2 +/- ~25 degree angle layers overwrapped by 2 hoop roll layers and the sequence repeated once. The casings tested to 3000 psi hydrostatic.

They were slow cook-off test cases so the end flanges were pipe flanges held in place by threaded rods (very large), not a flight motor. Custom CTI Pro75 propellant grains without casting tubes were case bonded to the casing using liquid rubber prepolymer and oven cured for several days.

Well, now we know what not to do. Back to the drawing board.

Click to expand...

Without violating the TRF policy on discussing motor building on the open forum, I would suggest using Garolite xx Phenolic casting tubes with the propellant grain cast with removable polyethylene or ptfe endplugs and bore rod that recess the grain several inches from the end of the casting tube at each end. (If you are using bate grain segments, they should be bonded to the phenolic liner) The xx provides the necessary thermal insulation to prevent damage to the composite casing. I would then make a xx forward closure with O-rings and epoxy it in place. Similarly I would make a cup type xx nozzle holder with O-rings and an O-rings graphite nozzle and epoxy that in place inside the xx motor casing. I would then overwrap the xx casing tube with sufficient uni-cloth to withstand the pressure. I would then make a CF ring to retain the nozzle holder and a CF disk to back the forward closure. You also need to use heat cured resins and must be careful as you are curing the epoxy after the propellant has been loaded so you need tight thermal control and a remote site to do it safely. (Not a garage or basement.)

Again. This approaches professional motor making and it will not be less expensive than simply using a larger casing and more propellant.
(Aluminum tube and plate is <$5 per pound, APCP is $10 a pound if you make your own. Good CF is $60 per pound if you make your own.) Composite casings cost far more if done properly with potential speed/payload/altitude gain of 10%-20%. Is it worth it?

Bob

Okay since I feel I have been called into this I will reply; something I really didn't want to get involved in. Bob, I know you bring a lot of experience and knowledge in certain areas to the forum and you generally try to help but your comments appear to be read as a little rude.

bobkrech said:

I did not say you were. I reviewed the whole thread and several have advised that a composite casing could be reused and some amateurs have done this. They were lucky and had no skin in the game. It didn't matter if their motor blew up.

Click to expand...

Composite cases made by both professionals and amateurs are capable of being reusable, period. There have been composite cases specifically designed to be reusable by both professionals and amateurs, that were successful and not because of luck! I think the most impressive example is by the student amateurs at uscrpl, which have now reused a composite case 9 times! If you think that is luck then we clearly have a different definition on the word luck.

https://uscrpl.blogspot.com/2014/04/semester-update.html

You premise is flawed. I'm not buying, and composite cases are far more expensive than a larger convention casing and more propellant...

Click to expand...

So if I understand this correctly, you are stating that you can save far more money by using a larger metal case capable of containing more propellant to account for the increase in mass? If so I also disagree with this on a couple of premises. Maybe in your experience they have been far more costly but not in my limited experience. Professional vs amateurs, I know but still I think this can vary.

I don't care if an amateur has done something before. Steve was lucky, and did not make an optimized casing.

Click to expand...

If you do not care what an amateur is doing then don't come to an amateur forum. Also by that logic maybe people don't care what a professional has been able to do or what they have not been able to do. Again with the luck statement; do you know how Steve went about designing/making a composite case and who he consulted? I do not think so, so how can you say he was lucky? Maybe you think any amateur who successfully uses a composite case is lucky because they are not a professional? Pinned composite cases have been used many times now without failure... Steve (since he was referenced) has done it 3 times now without failure. Maybe the case was not optimized but it was never designed to be optimized. Progression in steps.

Again. This approaches professional motor making and it will not be less expensive than simply using a larger casing and more propellant.
(Aluminum tube and plate is <$5 per pound, APCP is $10 a pound if you make your own. Good CF is $60 per pound if you make your own.) Composite casings cost far more if done properly with potential speed/payload/altitude gain of 10%-20%. Is it worth it?

Click to expand...

To some it may be worth it and to others it may not. Not everyone makes paper toy rockets and not everyone makes high performance composite and/or metal rockets, so some people care about the costs and others care about the optimization and performance. What is proper is up to the individual. I think that a proper composite case design is one that has demonstrated to be successful on multiple occasions. No one that I am aware of on this forum is trying to sell their composite cases for use on a professional level, so therefore different standards. With that being said the cost differentials are negligible in my opinion. 3" OD tube with a 0.125" wall and 36" Lg made from FWCF can be made for about ~70USD where as you can buy the same tube made from 6061 for about ~23USD. Not enough price difference for me to care about. However the aluminum tube will need some machining and a lathe to accommodate that size of tube will cost more than the filament winder. The bigger the lathe the far more the cost of the lathe. Not the case with the X-winder. If you are talking buying a case from a manufacturer then again the price differences are up in the air. Things like forward bulkhead, nozzle, pins etc are needed in both motors so it is not fair to account for the cost in one and not the other.

Bob,

I was cued in to this via multiple text messages and have little interest in engaging you in battle of intellect, link sharing, or historical data. Don't mistake my disinterest in having this conversation with an inability to do so.

For years, I've found the dichotomy between your online stature and demeanor and your in-person stature and demeanor to be stark. You're not the only engineer to contribute to this or any other hobby rocketry venue, so the comical haughtiness in your posts has never really made sense to me, but it's not my place to comment in others' business until you patronize me directly, which you've now done.

I write ads for a living - I could have gone the engineering path but elected to keep rocketry as a hobby. In rocketry as in life, however, I've found that there's a difference between saying something and actually doing it. If the amateur stuff is such a sleepwalk for you, why is it that you only fly commercial motors at CMass that are the size of my bowl scrapings? You can belittle my projects as amateur and poorly done, and the successful outcome of my projects as blind luck, but don't try and justify it with links to corporate projects you once touched, justify your stance on an amateur rocketry forum about amateur rocketry matters with your achievements in amateur rocketry. And guess what - in that arena, you can't hide behind your resume for a change, especially with me.

By the way, nowhere did I claim that my projects were optimized or even close to it. That wasn't the intent, and each of my composite cased projects (the largest with about 70 pounds of propellant) have been built in about a week's time. It's crazy what you can accomplish when you don't spend time on an online forum as the haughty godfather of conceptual knowledge and greatness.

In the spare cracks of time in my life, Mat and I are working on a more optimized composite case now that we've proven the concept - or "gotten lucky" as you've so eloquently coined it. I'm not going to sit here and throw links and theoretical performance data at you because I fancy myself not a hypocrite and that means nothing until you've gone out and done it. I'll do it at Aeronaut in August. See you there?

I know my place. Do you?

Steve Heller
Houston, TX
Formerly of Groton, MA
(203) 273-8718

Do what makes you happy. Truth of the matter is that essentially none of the projects listed on this forum fit in the "fully optimized" bucket. If you're able to make a composite cased motor that is finitely more efficient than an aluminum cased motor and results in some marginal improvement toward achieving your goal, go for it!

Also remember that there are multiple ways to achieve your goals. Certain people in this thread always steer toward a singular method. They suggest things such as "I would do it this way." That's great, but it may be more beneficial for those individuals to say "have you considered doing it this way? It may be more efficient/beneficial/stronger/safer than the way you're doing it."

Add some more pins, confirm your casting process, and try again. It's research rocketry.

bobkrech said:

The shuttle booster casings were steel, not FG. BTW it cost more money to recover and recondition them than they were worth.......

Bob

Click to expand...

They did make some composite casings, IIRC. In fact, I think they are the ones displayed on the outdoor shuttle stack at the Rocket Center in Huntsville.

Bob,

Your statements are only half true (no bolted joints on large composite SRMs is totally false), but even if they were 100% accurate I don't see the rational behind your stance on a project that is producing net positive rocketry related experience. There are plenty of fully functioning engineering systems that are not 100% optimized for mass; I would consider using carbon cases simply for the tolerance control to minimize liner gapping over metal. Looking at simply the material strength/temp curve is a net failure of a system engineering approach to the problem. What about cost? Availability/Schedule? Manufacturability? In today's environment composites are a lot easier if purely from an accessibility standpoint. I bet you have some kick ass composite resources or papers you could toss out there in ref to this failure? Point is, I don't think their stated goal of making a composite case is wrong.

Alex,

To add something productive, to this thread, your failure type seems to be a shearout/tensile failure (google your heart out). The problem won't necessarily be solved by adding more bolts or more rows of bolts (in the case of multiple rows the load is not shared equally anyway). I doesn't appear from the video that anything was happening off nominal, so its likely you just over pressurized the case by virtue of your MEOP- do you have any pressure data available?

You can decompose the pressure loads on the case into two parts: radial pressure loads and axial bearing loads (via the bulkhead -> pin -> case load path). This bearing load on your pin holes is quite large and in composite applications material is often added in this area to thicken it up and reduce the loads. Sorry if you already knew that!

Keep in mind also that the case (bodytube) you were using has been ground on the surface to make painting easier. Structurally this ruins the tube (chops fibers), so if you are taking a wag at strength via thickness, make sure you assume a significant knockdown.

Don't get discouraged - rocketry requires a massive knowledge base that has no clear starting point; like the design learning is an iterative process. It is always helpful to get some hands on experience along the way even if you screw up, just dont make the same mistake twice. Don't let Bob's advice go unheard either, literature reviews are helpful and often time you can save massive amounts of time and effort just reading a few papers.

There are many commercial composite SRM strap-on booster casings but AFAIK none are reusable. For certain proposed Shuttle polar missions out of Vandenberg, NASA was planning to use composite motor casings to reduce weight, but as AFAIK, they were never brought to flight readiness since the Shuttle missions from Vandenberg were cancelled. Since weight was an issue, AFAIK there was no plans to recover these boosters as a recovery system would have added weigh and reduced the payload that could be placed into a polar orbit.

Bob

That escalated quickly. Much like the static test.

We're going to redesign almost everything about the motor. We will no longer be using pins for obvious reasons. As of right now the plan for closure/nozzle retention will be to bond short sections of coupler to the inside of the tube in front of each closure. Here we are limited by the shear strength of our epoxy. A quick back of the napkin calculation says that with 1000psi of epoxy shear strength and 1.5in of tubing, it would take ~10,000lbf to break the coupler lose. This is about 2.7x the pressure applied to the closures at 1000psi internal pressure. Of course this is skipping over a few very important things, the biggest being heat damage. Im considering implementing a phenolic spacer between the nozzle and the bonded coupler to slow the heat transfer to the coupler. This 10,000lbf calculation is also optimistic because it assumes we get a perfect bond between the coupler and the tube. Couplers don't always fit perfectly, so the bond may not be ideal. We'll need a wide safety margin and will adjust accordingly when we find the right epoxy.
Other improvements will be the nozzle and FC design, We'll be using 3 O rings because I'm now paranoid about leakage.
To prevent another mandrel release disaster we may be changing our grain design again to ease the casting and release. It's a work in progress.

Thanks to all those who contributed useful suggestions. I really appreciate your input and I haven't been able to put my popcorn down.

Don't get discouraged

Click to expand...

I wasn't until I saw this thread

Alex

I have epoxy! good stuff too, first taste is free, :eyepop:

When you choose your epoxy make sure you know the proper bond line thickness. This is the epoxy thickness in the joint where it is rated at. Thicker and thinner do you no good. If you have to, cut your coupler so that you can expand it into the casing to create the proper bond line. Three o-rings is overkill. Two is more than enough. I've ran a single one on the forward closure and two on the nozzle for hundreds of motor tests and never had a problem. Get a Parker O-ring design book and use that to design your glands.

I would seriously consider looking to the styrofoam mandrels. They make casting much more predictable and enjoyable. I have cast 36" long 5" diameter motors with them. I even use them to for plain circular bores in 6" Bates geometry motors. They melt out easily and produce very nice results. The first fin-o-cyl I tried was a 75mm and used a hex bar with some tapered fins attached with wax. I spent a long time tapping that middle hex out and then prying the tapered fins out. Being stuck with easy to fabricate shapes also took away design freedom. With styrofoam you can create almost any shape you want. The 5" fin-o-cyl I have has an initial KN of 280, dips to 260 for the first third of the burn, then for the last 2/3 of the burn gradually rises to 305.

Edward

AlphaHybrids said:

I've ran a single one on the forward closure and two on the nozzle for hundreds of motor tests and never had a problem.

Click to expand...

Why two on the back and only one on the front? Does the temperature difference matter more than the pressure difference?

The forward closure seals fine with one, and I want to keep that part thinner. The nozzle I use two o-rings to help keep the nozzle more centered in the casing. It really doesn't matter as much because the groove and thrust washer are controlling how parallel the nozzle is, but there is room and it makes me feel better.

Edward

I'll second considering foam mandrels. We just recently did a test of what is essentially a 7600 case full M c-slot using liner bonded propellant in a custom liner, with a foam mandrel which we dissolved out with acetone after propellant cure. (get to full M by decent density ISP and pressure, high volume loading, no casting tubes so more volume, and if you're still making sponge propellant you won't get there) The burn was just barely successful at 8 seconds, the target burn time. I don't think it would have survived another half second. Since this was a sub-scale test, more work is required... The test likely cost me a case as it is now probably cooked. A cooked case is likely to fail on the next burn. Sometimes single use is just part of the assumptions, regardless of material choices.

As for larger aluminum tube based motors being expensive and requiring large tooling, not necessarily... Subject for another thread, another time. When I get some time I plan to make a small P motor on the cheap and will document it a bit. Expecting about 15 seconds burn so the biggest issues are thermal. I don't plan to use a cool burning propellant.

Too bad we are not allowed to fly steel.

Good luck with the composite case. I mean it. I like composites.

If someone tries the overwrapped aluminum tube method, and uses carbon fiber, watch out. You'll need to put a non-conductive primer layer over the aluminum to prevent galvanic corrosion leading to rapid premature bond failure particularly in the presense of moisture. Plus there are some other details that would need to be worked out for good bonding. It's a well known issue for carbon + aluminum so research it.

Gerald

Not that I am anywhere qualified to address this but if you glue the coupler to retain the bulkhead, what is the designed failure mode? Is it the case itself? If so at what pressure does it fail?

grouch said:

Not that I am anywhere qualified to address this but if you glue the coupler to retain the bulkhead, what is the designed failure mode? Is it the case itself? If so at what pressure does it fail?

Click to expand...

It depends on a lot of things. I'd prefer to have a 3x safety margin on the retention couplers, which I'm sure is much higher then our case. We've tested the FWFG tubing up to 1000psi and I'd imagine it can go a bit higher. We'll be running a max of about 800 to say below it. The real problem will be keeping the heat under control. Although we're using a high solids motor (well, moderately high...) (~84%), we're relying on a fast burn time (<2 seconds) and regressive or neutral thrust curve to help with limiting heat damage.

Edward (alphahybrids) also graciously volunteered to help with styrofoam mandrels, so it looks like we'll be heading down that route.

Thanks for the input Gerald, I'm curious to see how that P motor works out! I'm all for cheap performance.

Alex

G_T said:

....If someone tries the overwrapped aluminum tube method, and uses carbon fiber, watch out. You'll need to put a non-conductive primer layer over the aluminum to prevent galvanic corrosion leading to rapid premature bond failure particularly in the presense of moisture. Plus there are some other details that would need to be worked out for good bonding. It's a well known issue for carbon + aluminum so research it.

Gerald

Click to expand...

Overwrapping CF over aluminum tubing is a seriously bad idea on several other fronts. Aluminum is a thermal conductor and has a very high coefficient of thermal expansion. It's going to get hot quickly and expand rapidly adding a lot of tension and hoop stress to the CF overwrap, and as it gets hot it will weaken the resin and destroy the strength advantage of the CF composite.

Overwrapping Garolite XX or a similar phenolic resin liner with CF is a good way to go. The phenolic resin XX composite is an ablating insulator and has relatively low thermal conductivity so it won't get excessively hot during the burn. It also has a low coefficient of thermal expansion so it won't excessively stress the CF overwrap. It also provides a good surface for epoxy to bond the forward closure and a phenolic nozzle carrier to for a strong, gas-tight seal.

Bob