Casing Overheating and Rupture diagnosis 3

[sugarshot]

Some folks I'm working with on a hobby project could use some insights into addressing an experimental motor casing overheating problem.

It's a low energy solid fuel core-burning motor;

Say you have a long-ish motor casing with the usual bulkhead and nozzle. This casing has a midjoint, which is also a convergence-divergence zone. The fuel is located in the top half. The bottom half functions as a chamber.

The problem is midway up the lower half, there is a rapid heating effect (heat appears to accumulate more rapidly here, by measurement). On the test stand, the lower half splits open during the burn.

We don't want to approach the problem just by making the tube thicker or by using titanium.

So then, at considerable pressures and heating, how do we
reduce the heating effect and eliminate bursting? Between us, we have offered the following guesses;

1) This heating effect is mostly due to conductive thermal gains from the hot gases to the tube, and could be
addressed by adding a thin liner, and maybe use more resistive metal in the casing. The liner would be ceramic fiber with generous epoxy binder.

2) This heating effect is mostly due to thermal gains by thermal radiation from the gases to the tube, and could be addressed by changing the emissivity of the tube surface, and diffusivity of the tube metal. (my guess)

3) This heating effect is mostly due to the effect of
the convergence-divergence zone at the mid-joint. The shape of this 'shaped restrictor', results in gases interacting with the tube at steep angles, creating frictional overheating of the tube.

Again, the question is ultimately how to avoid casing rupture without increasing thickness. The next test casing will be with 4130 steel. The first two casings were standard electrical tubing. Someone suggested a fiberglass and thermoset polymer casing, which sounds unusual to me.

Ideas ?

 

[Sparweb]

Perhaps a series of cooling fins around the nozzle would help a bit. I assume your other tube joints are welds, so I don't expect fitting a series of radial fins would be difficult.

I don't know how easy it would be to get a flow of cooling air over them in your situation, or how sensitive your rocket is to drag. The cooling airflow would need a simulated source on the test-bed, too. Just my 2 cents worth. Though your project doesn't quite fit with the "professional" intent of this website, this sounds like too much fun to resist.

Have you measured the temperature of the nozzle? If you know the rate of temperature rise, you can take a guess at the number of BTU's absorbed by the nozzle per second. That would be the amount of heat your cooling system must reject. This risks turning your hobby into work, however.

Steven Fahey, CET

 

[Timelord]

Sugarshot,

Can I assume from your handle that you are a member of the "Sugar Shot to Space" team? I am a member of a team with similar goals called Project Morning Star. We are an amateur group mostly from Albuquerque with the stated goal of getting video from space.
Based on the position of the over heating my first pick for a explaination would be your guess 3. The diverging gases from the intermediate nozzle are probably impinging on the wall and drastically reducing the boundry layer, leading to localized heating. A liner that will delay the thermal transfer is probably the quickest solution. There is a material called "graphoil" (a registered trademark of Union Carbide) that may be the ideal solution. I have used this material for radiant heat insulation in a vacuum chamber and as a high power laser target. In thinner thicknesses, it can be cut with a sissors and rolled. It has anisotropic thermal properties, i.e. good heat conduction in the long dimensions and low conduction thru the thickness. Since it is pure carbon, it ablates very slowly (that's why I used it as a laser target). Google for graphoil.
That said, what is the purpose of the lower chamber? If you were using EMT as a casing, it must be a low pressure long impulse burn, making the thermal problem worse.

Timelord

 

[sugarshot]

Thanks, I'll follow your morning star project with interest. Yes, Sugar Shot to Space.

I'm not a motor development person for ssts, but participate in some of the issues.

When I describe the lower chamber problem, it's quite like that. I'd tell you the whole picture off-list. Even though ssts publishes a lot of results, I'm reluctant to go online myself and describe how we hope to set up the motor. Not everyone likes rocketry for the hobby and craft, so I don't want to upset anyone who might take exception to a good description. I'm erring on the side of caution there, even though the fundamentals may be common knowledge.

Thanks very much for your insights and recommendations, which I will forward. My role is payload team lead for the project, and we're doing a video too. If it could be useful, I'd be happy to discuss strategies and systems, possibly inviting some collaboration across projects.

Thanks very much for the tips!

Does Project Morning Star have a web presence?

Regards,
Geoff

 

[sugarshot]

The nozzle heating (the casing at the nozzle) seems to be quite limited. I will try to resist bringing the lighter issues here though, as clutter. This one seemed technical enough.

You're cooling airflow insight sounds like something we could have overlooked, and I'll pass it on to someone.

Cooling fins sound like a good idea.. it depends on other choices though, as the casing might have a relatively high R value.

Thanks very much.

Geoff

 

[Kwan]

What material are you using now? You said that you are using standard electrical tubing? What material is this made out of?

 

[sugarshot]

Yes, I said it was 'standard electrical'. I assume it was a galvanized steel, seam welded.

That was the first two tests. The next test will use
4130 steel.

We test using a 1/4 scale version.
The point of the tests initially were take temperature and pressure readings and see if it works as expected, so I think the choice of this was just a reasonable low-cost way to start testing. The tubes were also turned on a lathe, taking some wall away. Pinholes needed patching.

So the answer is 4130 steel.

 

[Timelord]

Geoff,

Project Morning Star is a low budget & minimal effort project that is not as well organized as your Sugar Shot, so we do not have a web presence, only a google.group where we discuss and communicate. Since I have a machine shop my position on the Morning Star team has defaulted to machinist. I just spent yesterday making test chambers for the booster guys to do fuel mix testing. Our plan is to launch a dart on top of a big solid fuel booster. The dart will carry a TV camera and broadcast the video & flight data in real time to the ground. We have tentively scheduled our shot for late this year and we hope to launch from Black Rock. Local preliminary test shots are scheduled for this summer.

Best of luck with your shot,
Kim

 

[btrueblood]

Okay, I've looked over your sugarshot.org website, and have a couple of comments.

Why are you trying to keep the 1st stage motor casing attached, as opposed to jettisoning it and thus improving the mass ratio for the rocket? I know that a jettison mechanism and igntion for 2nd stage is pretty complex, but you are asking for trouble trying to hold onto and fire through the long 1st stage casing.

Second comment, is to avoid a grafoil thermal barrier. Grafoil is intended to be used as a gasket material, i.e. squeezed between two hard surfaces. It is formed by calendaring (pressing) multiple plies/sheets of graphite. It has lovely in-plane strength, but the cross-plane (thru the thickness) strength is pretty poor. If you rolled a sheet of it into a tube, and glued it to the inner wall of the chamber, I would worry that large sheets of material would ablate/erode away, and potentially plug, or partially plug, the nozzle. I think you'd have better luck using a filled epoxy, or epoxy/fiber laminate construction. Even plywood would potentially work better.

P.S. the Russians used plywood as heat shields for their early ICBM RV's.

 

[sugarshot]

I proposed that we drop not the tube, but the 'lower' nozzle letting a re-designed mid-joint 'nozzle' do the job. A release mechanism with a 50,000 pound load seemed tricky, so we put that aside.

My understanding is regulatory limitations somewhat prohibit two-stage designs for amateur rockets or our kind. The idea is to go up in one piece, and return in two pieces by drogues and parachutes, dropping nothing along the way (my guess was dropping the nozzle might pass muster with some means of slowing descent), and it would be not far from the launch site. I agree entirely that were 'asking for trouble', esp. that prior heating of the lower tube doesn't help the situation, though rather than think of it as trouble, we think of it as something to overcome.. a design problem.

Early on someone made passing mention of grafoil, and we forgot about it, but one of our directors is presently interested in this option. Thanks very much for your opinion, I would think the ablation/erosion situation is 'dire', and the plan had recently been to use ceramic fiber impregnated with epoxy. The obvious question is what glue can hang onto graphite-anything? Chunks of that could be more problematic? On thing they are going to try alumina trihydrate coating on the exterior, for cooling. I recommended a forced-air tunnel contraption for better results, an idea that SparWeb presented above.

I like the plywood idea just because it sounds too weird.. that it could be the innovation we need. Our propellant will be cool enough upon casting to use wood.

One thing that worries me is the idea of using thermoset polymer / fiberglass tubing for the motor casing. I trust that that direction is well-considered, I just find it a little unnerving. Ultimately, testing will determine what is possible, but everyone's ideas here might well save us.

Among my more bizarre passing thoughts is injecting water during coast.

Intuitively, it seems that a good thick liner is the way to go, as often the reason that some materials are useful at all is they can be kept from getting past 300 degrees F. or so with a liner.

Considering we already have to swing a waiver or two, I think the directors and regulatory team were hoping to avoid a two stage design as that waiver would be harder to get approved.

Now that I've talked your ears off, I have a question.
It seems to me that none of us have done the research or gained knowledge of the problems of deploying a drogues at apogee (100km) with both halves spinning at 4 rps. A recent decision to not include a compressed gas de-spin system seemed to pass muster because no one could argue for the need to do it. Ultimately, the 'payload electronics' guys are getting our domain extended to recovery systems (co-located with most of our gadgets), and are responsible for answering questions like how well a swivel might work. The recovery team has rigging wizards, but they are just as green as us on these issues.

Everyone's feedback has been greatly appreciated.

Geoff

"mach 5 or bust"

 

[btrueblood]

"I proposed that we drop not the tube, but the 'lower' nozzle letting a re-designed mid-joint 'nozzle' do the job. A release mechanism with a 50,000 pound load seemed tricky, so we put that aside. "

50k lbs.? Wouldn't a simple slip-fit between stages work okay, provided you have enough strength to pass bending moments across the joint? I.e. you just need a way to transfer thrust, plus some %age of the thrust in bending moment to allow for gust loads & misalignments, from lower to upper stage, no need to restrain the joint (much) in the anti-thrust direction, you just let the "g" forces keep the parts together. As far as regulatory issues, as long as you have a recovery device for the 1st stage, what's the worry? If I were a regulatory person, I'd like the idea of multiple smaller pieces coming back down in controlled manner rather than one big heavy chunk. Especially if I knew that the one-piece design might seperate unintentionally into two pieces, one of which would be on a ballistic re-entry.

Okay, just read thru the "Preliminary Review" document on the website, and read the section regarding an "unclean staging" event being the regulator's worry. Will you not have a kill/destruct ability for the flight vehicle? Put a blow-out charge on the 2nd stage nozzle, and blowdown the 2nd stage chamber if an "event" occurs, then wait for a bit of coast-down and fire the recovery device. The issue is no worse than a poor launch sequence, other than that your impact dispersion is a larger area given the greater altitude and velocity at 2nd stage ignition. Should still be within the realm of the do-able provided you can clear a large enough range, or at least show a good probability of controlled descent for all the pieces.

Glue can hang onto graphite fibers, and monolithic graphite, just fine. The problem with grafoil is the inter-laminar strength of it just isn't there. Gr/epoxy would probably work okay, but I like the ceramic fiber idea too.

Not sure about the high-altitude aero- despin. The risk of twisting/winding up your shrouds is pretty high, even with a swivel. Is there a good reason for putting the spin on in the first place, or is it just a residual (e.g. calculated from thrust vector misalignment)?

 

[sugarshot]

I'm not an ME, but you draw attention to something that is worrisome, gust loads and misalignment. I might be able to get an answer on the single-stage choice to bring back here, which might be as you put it; "If I were a regulatory person..."

There will be a destruct system (no work done on this yet), and software/systems sophistication to accomodate 'bizarre' flight modes. I'm a big advocate of the 'what if's'.

You treatment of what I loosely call 'horizontal mode' is very thoughtful, as blowing out the nozzle wouldn't work. I'm not sure what you mean by 'blowdown' the second stage chamber. Is splitting the 2nd casing reasonable ala shuttle SRMs?

The reason for spin is we can't have any other guidance, so we spin to stay vertical, using shape not tilt of the fins.

Thanks again.

 

[berkshire]

sugarshot (Electrical)
One thing you might look at is a composite shell using phenolic liquid resin hand lay up using a glass fibre or carbon fiber reinforcment. The Phenolic resins are inherently fire retardant. An example would be Tamp resins 27351. made by Hyper industries of Bonita Ca While these resins do not have the strength of epoxies they also do not lose their strength as the temperature climbs. An example of their fire resistance a ma pp gas torch ply ed on a layup for 30 Min's will result in some spalling and the formation of carbon. A similar epoxy or polyester resin will erode a hole in a few minutes.
B.E.

 

[btrueblood]

How to snuff a KN-based propellant? Induce a negative dP/dt transient. How strong a transient depends on the propellant mix, but back in my rocket engineer days, it sure didn't seem to take much, although we were using a pretty low-rate KN propellant (for gas generation not propulsion). Only testing of your propellant grains will tell you what the magnitude of the dP/dt transient needs to be. We would do testing of grains in a special "tee" fixture, where we could pop a burst disc on the tee branch, and vent the test case through various sized orifices. Blowing out the nozzle on the 2nd stage should be enough, or yes, popping a hole in the side of the casing would work to (but then you have issues with the hole in the casing, how to insulate it yet have it reliably seperate, etc.).

Is the payload bay intended to return on its own recovery device? De-spinning a smaller chunk will make your job easier. I can envision two small drogues, each attached to a "boom" (maybe a coiled up piece of epoxy/fiber that would unspool when the payload capsule is ejected).

 

[sugarshot]

The burn times are fairly short, so the flight termination modes which may involve splitting the casing are few, but we will include (automated) this as it extends range-safety (manual).

With how I picture it, 'blowing out' the '2nd stage' (midjoint) nozzle would mean letting the lower stage casing etc free fall. Though the motor is constructed in two pieces, we want to keep them together for a few reasons (unless strategy changes later). The plan is to have drogue / chute setups in both the payload capsule and rocket body, with duplicate flight computers in each, such that one takes over should the other quit.

The burst disc concept sounds very good.

The rough cut plan is to seperate at apogee (100 km),
wait a few moments, and let out drouges. It's to let some velocity build so there is enough air to give the drogues some shape and also to let them drift apart. However, still spinning at 4 rps, we're not sure how much that would hurt us.

Berkshire, I forwarded your suggestion, it's sounds very useful to us. I forwarded pieces from everyone, by the way.

Speaking for myself and the project, having a few pros in the shadows of our project like this could be very significant for us.

If spin makes a mess of a difficult deployment situation, than I'm sure we would rush to reinstate it, however backup drogues might weigh about the same as the de-spin unit (5 lbs).

It would be a real gem for us if you fellows could help us understand the need for de-spin.

It's interesting, I didn't think snuffing a burning propellant was possible. thanks again !
-g

 

[Timelord]

Geoff,

I cannot help but comment again. You said " It would be a real gem for us if you fellows could help us understand the need for de-spin". The usual need for despin is so the payload can be stabilized to do its job. The Morning Star team looked at not despinning, but we found it was extremely difficult (and expensive) to post process the video to despin it. Rather than trying to despin the payload and all the attendant complications, we decided not to use spin stabilization. Our dart will not be spun so we can get usable video which is our stated mission. As far as I know there is no need to despin unless it is required by the payload instrumentation or recovery method. We are anticipating some tumbling after apogee due to the lack of atmosphere and spin, but by then our camera should be looking back up and it won't matter.

Kim

 

[btrueblood]

Hmm. Timelord has a good point. Why must your vehicle remain vertically oriented throughout the flight? If the aero loads are so small that the fins don't keep the nose pointed forward, then the aero loads are also so small that who cares if the rocket is going sideways, there's no drag penalty. In other words, just use standard fin-stabilization for flight, no spin.

 

[sugarshot]

We've taken it to be the way to get to altitude, and the way to keep the rocket's parts to land within reasonable driving distance.

The first burn event involves lumbering skyward to a mere 1000 feet. Wouldn't taking on angles that early invite further problems? After the first burn is done, the rocket will be "top heavy" - which is my interpretation.

We were pondering solar-referenced 'videos' at 4 frames per second, though capturing a good live video is indeed among our priorities. The intent of the video is to entertain the general public.

thanks

 

[btrueblood]

But you also won't have much spin in the first 1,000 feet - the aero forces haven't had much time to get the rocket spinning? So, you are really relying on the fins to maintain stability in classic "weather vane" mode. The only reason for spin is to null out a thrust vector misalignment, i.e. for targeting, but you don't really care about where, in terms of heading, the rocket goes, do you? You just want it to go mainly "up".

Please be aware, I might know quite a bit about interior ballistics and propellant, but external ballistics is more of a hobby too, not my professional experience. What my professional experience in general tells me is it seems that y'all are getting a bit too complicated. The old FLA "KISS" (Keep It Simple Stupid), and the phrase "simplificate and add lightness" are things to keep in the front of your mind when working in aerospace. If 'twere me, I'd build a two-stage rocket, no spin, and go to launch, perhaps with a dummy payload, and launch it to see how it goes. I'd only add complications later, if needed.

 

[sugarshot]

I'd imagine there will be small scale launches, though were still in motor development.

I'm a little unclear to me what tendency a rocket would have to favor the up direction, and I doubt genuinely that any off-angle attitude more than "several degrees" would allow us to reach altitude. Corrections?

My motto is "do the simplest thing that can't possibly fail".

Though I have much say in some strategies, the fundamental decisions like spin an motor design are out of my department, and those guys really know their stuff. I do try to contribute to the rocket fundamentals, and have passed forward most of the tips and opinions here so far. Even the experienced guys haven't done this before.

Isn't vertical a "target" ? Hopefully you can see which part I have yet to grasp. We don't care about heading though, N, S, E, W, as long as it stays in the desert and we know where it is and where it's going at all times. If our design had any guidance, we would not be able to get permission to launch. It doesn't make perfect sense, but those are the rules. We're keeping our waiver applications to small ticket items, which in the current design means just a hair over certain limits.
It's part of a good plan; stay on track to get
approved, with good analysis and nothing extreme.

When someone here make's suggestions to me, it is like talking to someone's secretary. Sure, if it's good it goes forward and we couldn't thank you enough just for the chance it may help. Especially from anyone who has been in the deep end of the pool, so to speak. I oversee flight electronics, the whole kit and kaboodle, and by some extension, these days I'm also the only guy in the hot seat over the choice to de-spin or not (recovery strategist?). If I could understand -why- leaving spin out would be an advantage, I'm sure my opinion would get full consideration. During a quiet week I could ask them why spin is critical.

Re: complication, approaching it with a complicated view
may be a mistake, but it was also my choice as a strategy, to frame how a sophisticated system could look like, and then pare it down from there. I take ''keep it simple'' to heart, and will keep bringing that voice to it.

Pardon my heavy overtones, I just need a nap.

.. and one more thing. One reason we are interested in going straight up is to keep the recovery distance down,
how far we'd have to drive, - tracking. The other reason is to gain launch approval for whichever area we will launch from. A common thread with our kind of experimental rocketeer is we are very attuned to safety issues, and the safety of others is always the highest priority. Coming back by parachute has to have some plausible guarantee, among others.

For those following the conversation, I'm leaning towards putting de-spin back, to favor the drogues, and catch a good video. Would you suggest I try to create a computer simulation of apogee events (seperation, deployment, etc)?

Thanks much.
-g