Cause of RUD at Max-Q

[Theory]

That there was a shred, NOT an explosion of any kind, and do not believe it had anything to do with the motor

Believe Steve is on the right track with his analysis

 

[DAllen]

Lol rocket most definitely did not explode. I was there and the 54mm case was most definitely in good shape. This was most definitely a shred of some sort.

 

[rfjustin]

Agree with @rharshberger, one team rep please, avoid duplication of efforts.

Collegiate teams are entertaining enough as is.

 

[Q38]

I'm a member of a rocketry team at Purdue University, and our team's rocket broke apart right before burnout, probably supersonic. I thought we broke a fin, but the fincan was recovered with all 4 fins still solidly attached. Any ideas on the cause of failure? Video footage, wreckage, and .ork are attached.

I should say that the rocket was supposed to be spin stabilized (bad idea from the start, I know), and the failure probably had something to do with that, as OpenRocket predicted a 2200 RPM spin at Max-Q. I'd still like to understand the exact cause of failure though, as our team is looking towards a space shot in the (far) future, and spin stabilization is very attractive for that type of launch.

Rocket design:
54mm min diameter, flying on a K540M with the Aerotech adapter in a 54-2800 case.
Full fiberglass
Fincan was recovered fully intact, other than a zippered body tube

Avionics:
AltusMetrum Telemega and EasyMini, on fully redundant circuits
Featherweight GPS
Head-end traditional Dual Deploy

Roll pitch coupling, CP shift with increasing M number, for high spin rates the air frame needs to be balanced....Cg must be coincident with the axial center-line. Remember the airframe WANTS to spin about the principle axis (it want to be in flat spin not an axial spin)

 

[George L]

Roll pitch yaw coupling should definitely be the thing here. Our school had a similar thing happen to our rocket at SA cup a few years back. High LD ratio coupled with aggressive motor resulted in a bent coupler (the load causing the issue was not axial loading). Rocket did an identical sharp turn...and came back in pieces.

 

[SolarYellow]

Shaft whirl is a well known dynamic behavior, although apparently not in Rocketry. You should ask for advice from your mechanical vibrations professor. That bent coupler is a large red flag in my opinion.

Did you ever try and test the different components for stability at 2200 RPM? Did you look at the DYNAMIC rotational balance?

TWENTY-TWO HUNDRED RPM.

Spinning up.... rocket = bullet... mass not accounted for laterally from bore centerline. At some given RPM and velocity the lateral mass/spin moment(s) caused the progressive material failure at the avbay/switch band interface. The airframe came apart and the recovery system zippered BOTH ends at approx. same time, and re-kitted properly. Recovery mostly stayed with avbay remains. Booster remnants came down a la Newton's apple.

But freaked out electronics due to the the rapid spinning, or insufficient or OVER-provision of venting at the high spin rates are also good explanations. One or more of those failure domains was GOING to present first...

It was just a glance and I can't see anymore.... but it's easy to see, physics killed that bird. Or, rather an inattention to physics killed it.
Who told you to spin that rocket like that?

Spinning a rocket causes a bending moment which is repeated at the rate of spin. That bending moment may eventually cause the coupler to fail, just the way it looks in your one photo. Once that happens the rocket comes apart, the chute comes out at high speed, and bad things follow. Because rockets fly fast, nearly everyone who tries to spin stabilize greatly overestimates the fin cant needed.

Being a car guy, I'm with all these guys. For a one-piece driveshaft, there's a "critical speed" at which it will fail mechanically and come apart. It's impossible to build it perfectly balanced (and even if you could, there would inevitably be some dynamic input from the end couplings that would generate a bending moment, so it doesn't really matter), so the rotational speed causes it to bend a little in the middle, making it further off balance, increasing the bending, etc. Basically, something consistent with what it appears your rocket did. Since you basically can't make the friction slip joints between the two airframe sections and the AV bay rigid enough to resist much force and keep the whole thing straight, it's going to flex in the middle due to whatever perturbation happens in flight, and you're guaranteed to get that result if you spin it fast enough.

Even with bullets, that are solid metal, if the harder copper alloy jacket isn't tough enough and the bullet is spun too fast, it's possible for the thing to just blow up due to the centrifugal force tearing it apart, shortly after leaving the barrel.

 

[David Schwantz]

I've had a 300 PRC fragment on the way down range.

 

[cjl]

Roll pitch coupling, CP shift with increasing M number, for high spin rates the air frame needs to be balanced....Cg must be coincident with the axial center-line. Remember the airframe WANTS to spin about the principle axis (it want to be in flat spin not an axial spin)

Eh, spin about either the lowest inertia axis or the highest inertia should be stable, in theory. It's only the intermediate axis that's problematic.

That having been said, at this kind of spin rate, you'd still need very good balance around the spin axis, and any coupler flex is going to be a serious problem. If you can keep the rocket straight and the balance around the spin axis is good, it substantially decreases trajectory dispersion compared to pure fin stabilization (which is why sounding rockets tend to spin), but you do need to make sure the spin isn't going to cause other problems.

 

[SolarYellow]

If you want to try pursuing spin stabilization, one approach might be to build the rocket and then put it on some kind of rotisserie/lathe arrangement and spin it. See what happens at what speed. Might be wise to include some kind of containment structure. Definitely safety glasses.

When doing destructive testing of c.f. structures, I've seen broken-off needles of carbon fiber composite embedded in materials a surprising distance away.

 

[Rschub]

I was reading reading Richard Nakka's site and came upon this interesting quote :

"For composite motors, a spin induced acceleration of at least 10g's is required before appreciable burn rate augmentation results. Is this a concern for spin-stabilized amateur rockets, then? A simple calculation shows that for a motor with a diameter of 4 inches (10 cm), a spin of 420 RPM is required to develop a 10g acceleration normal to the motor axis. Such a high spin rate is well beyond that required for stabilizing, so for amateur rockets, acceleration augmented burn rate is not a concern."

This is way out of my experience level, but I thought it would add to the discussion, or at least fuel to the fire.

 

[wclaybaugh2]

I was reading reading Richard Nakka's site and came upon this interesting quote :

"For composite motors, a spin induced acceleration of at least 10g's is required before appreciable burn rate augmentation results. Is this a concern for spin-stabilized amateur rockets, then? A simple calculation shows that for a motor with a diameter of 4 inches (10 cm), a spin of 420 RPM is required to develop a 10g acceleration normal to the motor axis. Such a high spin rate is well beyond that required for stabilizing, so for amateur rockets, acceleration augmented burn rate is not a concern."

This is way out of my experience level, but I thought it would add to the discussion, or at least fuel to the fire.

Not pouring gas on any burning bridges here...

If Richard is correct about 10g's then it follows that larger diameter rockets are more likely to show (slightly) higher burn rates over the burn duration compared to smaller diameter grains...which would suggest a consistent and measurable difference between the (published) static test data and flight data. But we also have to recognize that any such effect would be expected to "kick in" only toward the end of burn since the felt acceleration at the burn surface will be less than 10g's until the very end of the burn even in spinning larger diameter hobby motors.

If there is some data source on measured flight burn time for four to twelve inch diameter motors, and if the spin rate of the subject motor is known, then it might be possible to document what if any increased burn rate occurs from felt internal acceleration (ne: spin rate) in such motors.

As to the subject of the OP, this was almost certainly roll, pitch, yaw coupling. If one doesn't do the relevant analysis, then keeping the spin rate between seven and ten hertz will likely avoid coupling (at lower spin rates) and achieve the goal of averaging the effect of thrust and Cg asymmetries.

Bill

 

[MattM]

+1 What Tony said above at failure point. "How does rapid spinning effect the baro sensor"? Anyone have any idea...

BUT then my other question/comment: it looks like you are using a "sparky" motor. Is that correct? I wonder if the spinning caused metal to build up on the ID of the motor burn, and then that cleared the nozzle as chucks, a few times in the flight. The 3rd of which caused enough off axis thrust to bend / fail your coupler.

1st. Expelling of "material" see the "puff" a few feet behind rocket.
View attachment 568201

2nd expelling of "material"
View attachment 568202


3rd expelling of "material". It sticks and causes off axis thrust.
View attachment 568200

Last question. What is the core geometry? If not a centered circular core burner. Then the off center propellant mass would be changing dynamically during the burn, at some RPM versus, weight, versus stability shift; they coupled together and made a large oscillation.

Interesting theory about the 'chuffs' of propellant. The motor was my (and my club's) first flight on a sparky, didn't realize that chuffing wasn't normal. Motor was a K540M, which has a centered circular core.

 

[MattM]

What did the motor case and nozzle look like after you recovered it? Was there any burn-through of the case or around the top of the nozzle? One frame of the video looks like you have two exhaust streams, like a burn-through, but it could also be the result of a pyro event.

View attachment 568226

No burn-through on the case, no leaks, nozzle looked normal. The whole RMS was spotless, which saves some money but isn't super helpful.

 

[MattM]

Answers to common questions:
Static margin was 2.1 on the pad, only increased as rocket flew. OR and Rasaero predict CP only moves aft after the sound barrier, so the rocket was *statically* stable through the flight.
We did not balance the rocket or do any kind of analysis regarding roll-pitch coupling, although we should've, and we will in the future.
The avbay was vented according to Altus Metrum guidelines, but the guidelines obviously did not account for centrifugal forces of the air or anything like that.

Also, I should say that we caught a pyro event on video long after the breakup which appeared to be the main charge due to the altitude. Couldn't make out a deployment event at max-Q, but if there was, it was only one, as the rocket was also recovered with 2 charges intact.

Seems like likely cause was a roll-pitch issue combined with an unbalanced rocket causing dynamic instability.
More importantly, if we rebuilt/relaunched the rocket with straight fins (no spin), do you think there's any chance of a similar failure?

 

[MattM]

What did the motor case and nozzle look like after you recovered it? Was there any burn-through of the case or around the top of the nozzle?

Case and nozzle were spotless, no sign of burn-through. One main and one drogue charge fired based on the recovered avbay, and the main charge was visible, so if the separation was due to a charge it must've been the drogue, but doesn't altus metrum computers have Kalman filters to prevent things like this?

 

[MattM]

It looks like in the video an ejection charge goes off before the rocket takes a hard turn. Did you have a the motor ejection charge as well as your electronic ejection charges. I had a K2050 set off the motor eject right at burnout with an un-drilled delay, I believe hot gasses got past the delay due to mis-asembly on my part. In my case it was obviously the ejection charge as the airframe was coated in soot but both electronic ejection charges clearly happened after separation.

No motor ejection charge installed, and the forward closure we had installed on the case actually prevented any misfires such as that, since there were no holes in the 54-2800 forward closure.

 

[MattM]

Spinning a rocket causes a bending moment which is repeated at the rate of spin. That bending moment may eventually cause the coupler to fail, just the way it looks in your one photo. Once that happens the rocket comes apart, the chute comes out at high speed, and bad things follow. Because rockets fly fast, nearly everyone who tries to spin stabilize greatly overestimates the fin cant needed.

Guilty. 2200 RPM seemed fast in the sim, never mind in real life. Is there a "happy medium" RPM zone where the spin is able to keep the rocket pointy-end-up outside the atmosphere, but won't cause problems with dynamic stability and bending moments?

 

[user 30299]

Guilty. 2200 RPM seemed fast in the sim, never mind in real life. Is there a "happy medium" RPM zone where the spin is able to keep the rocket pointy-end-up outside the atmosphere, but won't cause problems with dynamic stability and bending moments?

What is the overall gain in the rocket's performance by having the spin condition? Does it add a percent or two
to the performance (altitude, straight tracking, etc.) or is it double digits? Is it a large amount of work for an
insignificant gain? Some one might say that the gain is the learning experience - which is true.

And the happy medium for the rpm zone is most likely controlled by the rocket's unbalanced mass condition that
varies along the rocket's length. How would you model that unbalanced spinning mass in a dynamic analysis?

How are shafts, cylinders, pulleys and sheaves balanced, in industry, for rpms over a large range of rpms?

 

[user 30299]

More importantly, if we rebuilt/relaunched the rocket with straight fins (no spin), do you think there's any chance of a similar failure?

What do you think?

Did you ask you faculty advisor and mentor this question?

 

[Worsaer]

Guilty. 2200 RPM seemed fast in the sim, never mind in real life. Is there a "happy medium" RPM zone where the spin is able to keep the rocket pointy-end-up outside the atmosphere, but won't cause problems with dynamic stability and bending moments?

There is abundant reference material on spin stabilization of sounding rockets. You may want to consider that imperfect spin may create more issues than benefits.

 

[boatgeek]

Case and nozzle were spotless, no sign of burn-through. One main and one drogue charge fired based on the recovered avbay, and the main charge was visible, so if the separation was due to a charge it must've been the drogue, but doesn't altus metrum computers have Kalman filters to prevent things like this?

You should be able to download the barometric altitude and maybe vertical acceleration from your altimeter. Compare to your OR sims to see what the altimeter was doing. You might also be able to determine when election events were commanded by the altimeter. I would be concerned that only one set of charges blew.

 

[Donnager]

Spinning a rocket causes a bending moment which is repeated at the rate of spin. That bending moment may eventually cause the coupler to fail, just the way it looks in your one photo. Once that happens the rocket comes apart, the chute comes out at high speed, and bad things follow. Because rockets fly fast, nearly everyone who tries to spin stabilize greatly overestimates the fin cant needed.

I've always correlated this to critical shaft speed (I'm a chemical engineer and have build more than a few agitators). Driven shafts (a rocket motor could be considered a power source) can spin something at an RPM it is simply unsuited for. Aerodynamic concerns will make this worse, in my estimation.

It wouldn't surprise me to see a failure, simply from rotational velocity. Without engineering data, I'd hate to speculate further.

 

[sr205347d]

Is there a "happy medium" RPM zone where the spin is able to keep the rocket pointy-end-up outside the atmosphere, but won't cause problems with dynamic stability and bending moments?

No.

Contrary to popular belief, spinning a rocket will not stabilize it. In fact, it will de-stabilize it. Go back and read my earlier post, and the ones that mention roll-pitch coupling.

The engineers who launched the first US satellite, Explorer I, thought that spin would stabilize it in space. However:



The satellite was shaped much like a rocket, long and slender.

 

[MattM]

You should be able to download the barometric altitude and maybe vertical acceleration from your altimeter. Compare to your OR sims to see what the altimeter was doing. You might also be able to determine when election events were commanded by the altimeter. I would be concerned that only one set of charges blew.

Altimeters were in the coupler tube, so no flight data was saved. The TeleMega died first (both charges unfired), with the EasyMini surviving longer, and in better condition, blowing both of its charges

 

[MattM]

No.

Contrary to popular belief, spinning a rocket will not stabilize it. In fact, it will de-stabilize it. Go back and read my earlier post, and the ones that mention roll-pitch coupling.

The engineers who launched the first US satellite, Explorer I, thought that spin would stabilize it in space. However:

View attachment 573008

The satellite was shaped much like a rocket, long and slender.

View attachment 573009

Heard about this. My understanding was satellite spin-stabilization didn't work in the long term due to energy loss and momentum conservation, causing the vehicle to tumble end-over-end after a few orbits. Our spaceshot theoretically wouldn't have this problem due to the much shorter flight time. I believe spin-stabilization is an industry solution to smaller sounding rockets, but not orbital-class boosters, for this reason

 

[MattM]

What is the overall gain in the rocket's performance by having the spin condition? Some one might say that the gain is the learning experience - which is true.

It would be more of a reliability/stability issue, rather than a performance one. Right now, the spin only hurts the rocket, as we're stable without it and we don't need it, so it's a waste of energy. In the future, we'll be flying outside the atmosphere, so fins won't work. I assume we would need a secondary form of stability to keep the rocket pointy end up for this phase of flight, and spin seemed like the easiest solution.

As far as the balance dynamics analysis, there was none, and likely won't be in the foreseeable future. That kind of analysis is above my team's pay grade at the moment.

 

[wclaybaugh2]

Seems like likely cause was a roll-pitch issue combined with an unbalanced rocket causing dynamic instability.
More importantly, if we rebuilt/relaunched the rocket with straight fins (no spin), do you think there's any chance of a similar failure?

Matt:

Depends. Without spin, an unbalanced rocket (Cg not coincident w/ longitudinal axis) will gravity turn. You can also induce turns from thrust asymmetry. Both of these effects are mitigated by spinning.

And getting fins *exactly* aligned for no spin is very difficult…even one thousandth of an inch offset will induce spin at sufficient dynamic pressure.

In any case, if it is spinning at all, the vehicle will transition to a flat spin once exoatmospheric; just physics….

Bill

 

[wclaybaugh2]

As far as the balance dynamics analysis, there was none, and likely won't be in the foreseeable future. That kind of analysis is above my team's pay grade at the moment.

Matt:

The tool that generates the force analysis you need is Missile Datcom. It is available to DOD contractors that are US persons. Since most (US) universities have DOD contracts most university rocket teams use this tool for dynamic analysis.

Bill

 

[boatgeek]

It would be more of a reliability/stability issue, rather than a performance one. Right now, the spin only hurts the rocket, as we're stable without it and we don't need it, so it's a waste of energy. In the future, we'll be flying outside the atmosphere, so fins won't work. I assume we would need a secondary form of stability to keep the rocket pointy end up for this phase of flight, and spin seemed like the easiest solution.

As far as the balance dynamics analysis, there was none, and likely won't be in the foreseeable future. That kind of analysis is above my team's pay grade at the moment.

Forgive me if I'm oversimplifying, but if there's so little atmosphere that fins are ineffective, is there any noticeable increase in drag from a tumbling rocket? Obviously, tumbling would be an issue if you were trying to maintain an orbit, but for coast to apogee above the Karmann Line, this seems like a relatively small effect.

Put another way, is the mass of a spin-despin system more of an altitude penalty than just accepting tumbling at very high altitude?

 

[sr205347d]

Take a look at this: https://www.nakka-rocketry.net/ep_lr7.html

Bottom line: "It presently appears that inertial roll coupling may well have led to dynamic instability and consequential structural overload (bending moment) and breakup of the rocket, occurring at the forward/mid fuselage joint."