Source: Development Problems section of Wiki's Ares I article.

Since NASA considers the problem to be severe (they ranked it 4/5 on their risk scale)...

Gas vortices inside first stage SRBs aren't fun. Lots of vibration.

Still, I'm very surprised at this, as Ares I's first stage is about 90% identitical to the Shuttle's SRBs:

- Wikipedia

Sounds to me like Ares I's first stage is a minor retooling job for the boys at Michoud. As it turns out, they were initially designed for the Space Shuttle, in order to provide it with an additional 20,000 lb payload capacity and the ability to perform a mid-Atlantic dog-leg maneuver so as to enter polar orbits.

So where did the vortex problem come in? Or has it always been a problem with the Shuttle's SRBs?

Regardless, here's how to fix vortex problems in an SRB:

Prime Numbers.

Why primes? Simple - when inducing artificial turbulence into a system in order to prevent the establishment of uncontrolled oscillations, you use parameters which, by their nature, avoid being multiples of one another.

The Tacoma Narrows bridge incident was a prime (pardon the pun) example of how multiples used in building structures can lead to catastrophic reinforcement of oscillations. By using primes, you minimize reinforcement, and generally induce damping effects, rather than reinforcing effects.

The question is - how does one use primes to reduce/eliminate harmonic reinforcement, particular when it comes to turbulent vortices?

Well, we know that the propellant has various shapes throughout it's length due to mission requirements. For example:

"The propellant is an 11-point star-shaped perforation in the forward motor segment and a double-truncated-cone perforation in each of the aft segments and aft closure. This configuration provides high thrust at ignition and then reduces the thrust by approximately a third 50 seconds after lift-off to avoid overstressing the vehicle during maximum dynamic pressure." - Wikipedia

Thus, we implement primes by simply reshaping the surface topology of the interior of the SRB's APCP (solid fuel). The eleven-point star is a good start, possibly followed by stars of other primes, including 7, 13, etc., of varying depths in order to achieve the desired effect and minimize harmonic reinforcement. To help alleviate vortex flow, the stars could be installed at angles slightly off of straight-line, or with gentle back and forth waves, which would tend to create turbulent counter-vortices thus preventing harmonic vortex flow from occurring in the first place.

This approach is similar to the small (about 2" x 2") eddie generators found atop most wings on airliners to generate many small turbulent flows which in turn facilitates gradual stalls by preventing massive laminar flow detachments which cause rapid and very significant stalls.

Forgive my bluntness, but what do you think the chances that a solution to a major engineering problem faced by NASA is going to be first proposed on a random web forum?

If the solution were as simple as you make out (and your solution does not sound simple at all, it sounds like a complete redesign of the fuel grain based on numerological principles...) then I am sure one of the many people working on this problem would've figured it out by now.

__________________
"I worry that, especially as the Millennium edges nearer, pseudo-science and superstition will seem year by year more tempting, the siren song of unreason more sonorous and attractive." - Carl Sagan, 1995

So?

Doesn't mean we can't have the fun of tossing ideas around

I sometimes think that too, especially if it starts with "NASA iz t3h dumb cuz they din't thunk it up" or when such a sentiment is obviously implied. Still, as Neverfly said, it can be fun, even educational.

__________________
"Ignorance more frequently begets confidence than does knowledge" -- Charles Darwin
"Your right to hold an opinion is not being contested. Your expectation that it be taken seriously is." -- Jason Thompson
Meet the OOONG TOE.

OK, maybe I was a bit snappy, but my first impression remains that there isn't much merit to the idea in itself.

It seems to me that you are saying the cross section of the fuel should change over the length of the SRB. As you are changing the number of points in the star shaped hollow in the middle of the fuel, this means introducing new edges and my (quite limited) knowledge of fluid dynamics suggests this will increase turbulence not decrease it.

__________________
"I worry that, especially as the Millennium edges nearer, pseudo-science and superstition will seem year by year more tempting, the siren song of unreason more sonorous and attractive." - Carl Sagan, 1995

The cross section of the fuel already changes over the length of the SRB. It does that to reduce the SRB thrust throughout the Max Q transition.

I'm merely mixing up the shapes in the same way that the re-designed Tacoma Narrows bridge mixed up the bridge's numerous and varied harmonic oscillations so that none reinforced one another, as was the case in the original (and disasterous) design.

First, BAUT is hardly a "random web forum." Look around you. I see a handful of people who truly are rocket scientists, along with a very healthy dose of smart people who easily could have been had they chosen it as their profession.

Second, just because someone has their doctorate and is the president and CEO of a $130 million annual income company who's primary business is, say, enhance GPS solutions to meet real-world requirements doesn't mean that a "casual" passerby at their booth during a Tech Expo can't offer valuable advice, which is precisely what happened on Tue, where I spent 15 minutes explaining what the Schuler cycle was and how it was important that they factor it into the design of the inertial navigation portion of their equipment.

They'd never even heard of the Schuler cycle. I guess even those with a PhD learn something new every day.

Actually, you're wrong on both counts.

First, it's not a "complete redesign." It's simply a modification of the current internal contouring of the fuel in the SRB.

Second, your ending conlusion is a logical fallacy. While one would think that might be true, it's often not the case when applied to large organizations. Excellent solutions to various issues are often generated from within, but due to a number of psychological and sociological reasons, those solutions die or dead-end somewhere between the originator and implementation.

Case in point: The numerous internal memos from various NASA engineers which detailed the problems inherent with the o-ring design. It took a major catastrophe (Challenger) before NASA, as a corporation, modified the design.

Easier solution: Dump the SRBs.

Would you want to step into a car or airplane that you knew you could not throttle or turn OFF????

They're much simpler and have far fewer parts to fail than a liquid motor. I'd actually prefer to be on an SRB over a liquid motor.

Why do you think JATOs are solid? It's because of the simplicity.

__________________
WANTED:

Schroedinger's Cat

Dead And Alive

That and the fact that they have a tremendous specific impulse, especially when staged as are the Shuttle's SRBs.

That and so they don't go BOOM! when a nervous ground troop drops one on the tarmac during a firefight. It helps when you're getting shot at, too, as they were designed to take standard NATO and AK-47 rounds without going BOOM!

Their utility with respect to Herks, though, is pretty much limited to the evacuation of bunches of people, as the Herk can almost always take off in less distance than it takes to land.

Actually, their specific impulse is rather poor. The shuttle SRB's can only manage an Isp of 270 seconds or so, compared with the 350+ seconds of most liquid motors.

__________________
WANTED:

Schroedinger's Cat

Dead And Alive

If you turn off a rocket, liquid, solid or otherwise, you crash.

For those who might be interested here is a pitch from the NASA working group that is looking at the problem.

http://www.nasa.gov/pdf/221186main_t...int_report.pdf.

A few points might be of interest:
1. The problem is hypothetical and is based on analysis of the new design and not on any experimental data for the 5-segment design. Pressure-time curves from the 4 segment design are being analytically extrapolated to the 5-segment design.
2. The pressure oscillations seen in the 4-segment design are on the order of 1 psi.
3. Acoustic analysis for solid rockets is notoriously difficult and inaccurate.
4. Upcoming tests will be revealing.

I have no problems with SRBs. We are always going to have solid augmentation.

Are you including the weight of the propellant...?

Qualification: Prior to accomplishing orbital entry burn.

Reviewing...

Question: What's the orientation of this differential? Is it merely a change in thrust? Is it symmetrical about the longitudinal axis of the SRB? Is it assymetrical?

A 5 Hz 1 psi oscillation spread out over the interior of a cylinder wall ten feet long and 6 feet in radius exerts 54,286 lbs, five times a second.

That's like getting hit with 5 tractor-trailer rigs each second.

A constant pressure across such an area is one thing. An oscillating pressure across such an area is another.

If that pressure is assymetric, ie, exerting force on one interior side of the booster, and not the other, it can exert tremendous torsional loads along the booster, which then imparts those oscillating loads to the main tank.

Looking at your link, the Roadmap, it appears the issue is at the "Validate Analysis Approach & Eliminate Over-Conservatism."

Hmm... I don't like that last one, as it imparts a fudge factor (factor of safety) for those "I don't knows." Thorough analysis, design, manufacturing, and testing can reduce or eliminate those "I don't knows," particularly if the manufactured assembly is tested before use.

It appears as if they're focusing on design modifications external to the propellant itself, including propellant damping and the use of a tuned mass absorber.

From the Grid x Acceleration T115, it's clear there's some harmonic oscillations going on, at around 9.8 Hz and 11.7 Hz. The Motor Test Pressure doesn't look very good, either, as the spikes at 50s, 78s, 86s, 95s, 100s, and 113s require signficantly stronger (heavier) structural container, yet contribute very little to the overall thrust of the motor.

Given the nature of the spikes, it appears as if they're due to the sharp demarcations between laminar and turbulent flow, either of which would be preferable to transitional back and forth between them. Is there a way to induce toroidal flow within the motor? If so, you'd need to redesign both the interior geometry of the solid fuel as well as the nozzle.

Ideally, you'd want laminar flow, but that's a bit difficult to achieve in a solid fuel motor, hence my idea of intentionally inducing turbulent flow to avoid laminar-turbulent transitions which result in pressure spikes like these.

Although harmonics are clearly at work, these can determined via sensors throughout flight to determine oscillation modes. They can be compensated either actively, by changing the inducing frequency from the motors (difficult to analyze/design without repetitive testing), or passively, by modifying the mass distribution throughout the structure (usually imparts a weight penalty).

The second set of graphs shows a smoother test pressure vs time. The "Driving Mechanisms?" answer is probably standing waves, as the increasing clarity (resolution of the peak about a particular geographic point) over time is indicative of the behavior of standing waves. We see this elsewhere, from dune size being proportional to the mean velocity of the wind, to ripples in the bottom sand of a river being proportional to the velocity of the water. Same thing's true for surface waves of water and wind.

MACH standing waves exhibit similar characteristics, such as the diamond shock patterns seen during tests of various engines and motors where the exit velocity is supersonic. It's easy to think of these as "internal diamond shock patterns." The reason they intensify over time is probably indicative of the fact that the greater heat and pressure in their location results in a more rapid burn of the fuel in that location - a ripple. This ripple (trough, actually) at 1L is repeated at 2L and 3L.

It's a standing wave.

The pressure response in the second series of graphs in response to a forcing function shows a good damping of the earlier oscillations, but a new peak at 12.2 Hz.

Let's play with some numbers, shall we?

Original Peaks: 9.8 and 11.7 Hz.
New Peak: 12.2 Hz

Original Differential: 11.7 - 9.8 = 1.9

Peak addition: 9.8 + 1.9 = 11.7 (no surprise...)

Peak addition: 11.7 + 1.9 = 13.6 (hmm...)

Peak subtraction: 13.6 - 12.2 = 1.4 (hmm...)

Ratios (we're looking for integers, or numbers very close to an integer):

9.8/1.9 = 5.158 (5.1578947368421052631578947368421)

11.7/1.9 = 6.158 (6.1578947368421052631578947368421)

Hmm...

13.6/1.9 = 7.158 (7.1578947368421052631578947368421)

Is anyone else recognizing this pattern?

Let me make it simple: Two frequencies, 9.8 Hz and 11.7 Hz, when suppressed, give rise to another frequency, 13.6 Hz, all three of which, when divided by the difference between the first two, result in a number that's very close to an integer (within the error rate inherent in reading off the values from the graphs will allow).

If I'm not mistaken, the result of suppressing the previous two modes of vibration at 9.8 Hz and 11.7 Hz, result in an accentuated intermodulation distortion peak at 13.6 Hz. This was a problem during the earlier (70's, and 80's) days of receiver-amplifiers, all of which tried to out-do one another via massive negative feedback to reduce the harmonic distortion to ridiculously low levels (several orders of magnitude below which the human ear can distingish). The result, however, was a significant increase in intermodulation distortion at higher power levels. Since IMD can be detected as a harsh grating in the music, amps with great THD specs stunk at higher power levels.

The forcing function, oscillation dampening, whatever you want to call it is analogous in that it's negative feedback to the system. The problem is that the energy in those oscillations aren't just stopped. It's transformed by the new system into either new harmonic distortion (oscillations) or intermodulation distortion.

Since the original two modes of vibration were not harmonics and were less than an octave apart, their suppression resulted in an IMD frequency of 12.2 Hz. It's not classic IMD, mind you. However, it's the expected result of trying to dampen (negative feedback) two signals of similar frequencies (<< 1 octave apart).

So... What does the 1.4 mean? It's pretty close to the square root of 2, but at this point, I'm guessing.

That's what the specific impulse is, by definition: total impulse divided by propellant weight. Solid motors have excellent thrust for their size and weight, and have very good propellant density, but they tend not to burn for very long, and their overall specific impulse suffers as a result.

__________________
WANTED:

Schroedinger's Cat

Dead And Alive

This is fun; but I am afraid this is a case of lighting a fuze and seeing what happens. Propellant burning is very complex: Think in terms of a room with corrigated rubber walls. There are globs of aluminum, and the chamber is clouded with dampening aluminum oxide. There are also self-stabilizing burning modes (at certain elevated pressures, the burn rates actually drop!).

__________________
jwj

It's a big universe out there...is it really unwinding, really burning out?