Author Topic: Ares I-X Flight Data Evaluation (Read 14512 times)

rdale · « **on:** 02/23/2011 10:19 pm »

This overview provides summary information from assessment of the Ares I-X flight test data and represents a small subset of the detailed technical results. The Ares Projects Office published a 1,600-plus-page detailed technical report that documents the full set of results.

http://www.nasa.gov/pdf/515784main_APO-1096%20_%20Ares%20I-X%20Flt%20Eval%20Exec%20Overview.pdf

robertross · « **Reply #1 on:** 03/06/2011 02:34 am »

Thanks for the link.
I guess we'll not see the full report, even on L2?

Being slightly suspicious, I have to wonder about some of the 'recomendations' proposed that are to be taken away from all this (like Loads, First Stage TO, Instrumentation, and Data Analysis).

rdale · « **Reply #2 on:** 03/06/2011 02:47 am »

Never... Tried every angle I could and it's not coming out.

madscientist197 · « **Reply #3 on:** 03/06/2011 10:15 am »

What a load of [censored]. Statements like: 'based on the results [which we're not going to tell you]... the Ares I acoustic environment prediction models and processes will be updated.'

There's absolutely nothing there -- twelve pages of fluff. I know that there are limitations on what they can publish, but IMHO they ought to be ashamed of publishing this particular document

robertross · « **Reply #4 on:** 04/18/2011 05:20 pm »

At the request of Chris's post, I'll refrain from further comments on the CCDev2 thread and simply state (while bumping this thread), that it's the LACK of detailed information in key areas as detailed on this thread that cast doubt, especially on Thrust Oscillation - something you wouldn't think is ITAR.

I can appreciate an internal source indicating all went according to plan, but I can also look back on the history of the CxP development and not be impressed with the level of disclosure.

Chris Bergin · « **Reply #5 on:** 04/18/2011 07:02 pm »

Thanks Robert!

So let's have the lines quoted in question and we'll go from there.

JohnFornaro · « **Reply #6 on:** 04/19/2011 12:57 pm »

So TO is still a problem, eh?

They need to study the effects of varying the grain geometry of the finocyl. Based on those waterfall charts, it should be possible to analyze the timing, location and amplitude of the vibrations, and create a geometry to ameliorate this. Existing combustion models should be able to be modified to accomodate the new geometry.

If this effort shows that the geometry variance does indeed smooth out the vibrations, a physical model, appropriately sized, could be made to measure the results empirically, so as to improve the math model. There should be several iterations of this, because the other thing that needs to be studied is the manufacturing.

Somehow, the varying grain must be cast, and the molds must be removed from the model. The new techniques must be scalable as well.

Hey. That's what I recommend.

robertross · « **Reply #7 on:** 04/19/2011 01:12 pm »

Quote from: JohnFornaro on 04/19/2011 12:57 pm

So TO is still a problem, eh?

They need to study the effects of varying the grain geometry of the finocyl. Based on those waterfall charts, it should be possible to analyze the timing, location and amplitude of the vibrations, and create a geometry to ameliorate this. Existing combustion models should be able to be modified to accomodate the new geometry.

If this effort shows that the geometry variance does indeed smooth out the vibrations, a physical model, appropriately sized, could be made to measure the results empirically, so as to improve the math model. There should be several iterations of this, because the other thing that needs to be studied is the manufacturing.

Somehow, the varying grain must be cast, and the molds must be removed from the model. The new techniques must be scalable as well.

Hey. That's what I recommend.

I'm trying to place together the issues in a concise format (taking longer than I hoped for).

The grain pattern 'helps', but that seems to have been already addressed through testing. The main issue comes near burnout, where there is little propellant left, where the predicted oscillations would be highest, and there is a huge uncertainty factor in WHEN it will rear its ugly head. Also, not two castings are alike, and if you play the statistical odds, you need to ensure the mitigation is in place to counter the effects. But if you don't know WHAT they effects will be because they happen so infrequently, and it's difficult to model so you can't really effectively counter it.

JohnFornaro · « **Reply #8 on:** 04/19/2011 01:36 pm »

Hey! Thanks for the response!

Now that I've just read the report, with an eye to TO, I find that they are at least considering a physical, scaled model:

Quote

Continue development and use of empirical scaling and proceed with use
of hybrid statistical energy analysis finite element modeling for future flight tests.

Maybe a physical model ["empirical scaling"] will be the result of this effort. I understand that no two castings are alike, but they are enough alike so as to reliably make them and count on the results. Although the unpredictability of the vibration is, well, unpredictable, my understanding is that it is predictable enough to warrant the introduction of a massive (1.5mT?) "shock absorber" for mitigation.

The suggestion is simply that if the mitigation can actually happen within the mass of the propellant, there is that savings as well as the reduction in the vibratory environment.

rocketscintist · « **Reply #9 on:** 04/20/2011 04:08 am »

Quote from: JohnFornaro on 04/19/2011 12:57 pm

So TO is still a problem, eh?

They need to study the effects of varying the grain geometry of the finocyl. Based on those waterfall charts, it should be possible to analyze the timing, location and amplitude of the vibrations, and create a geometry to ameliorate this. Existing combustion models should be able to be modified to accomodate the new geometry.

The SRB thrust oscillation phenomenon can't be fixed with grain reshaping. By the burn-time when thrust oscillation is a problem, most of the propellant is gone. The motor is down to being a collection of short (in length) CP grains that have only a 1/6th of their web left. Shaping the finocyls can't do anything to change that.

JohnFornaro · « **Reply #10 on:** 04/21/2011 11:50 am »

Quote

The motor is down to being a collection of short (in length) CP grains that have only a 1/6th of their web left...

Thanks. A bit more detail than Robert offered, about the geometry of the CP grains at that time. Could somebody re-post those waterfall charts? I can't find the ones I've seen, and don't recall the specifics very well.

At the end of the burn, clearly, from an intuitive understanding, the "tube" resembles a column of vibrating "air" more and more. Surely the case resonance becomes more a factor, as the damping effect of the grain lessens. If the locations of the oscillations along the axis of the tube are known, and I think they are, perhaps stiffening at these points can ameliorate. Understood that this is an addition of mass, as opposed to a strictly geometric rearrangement of the "same" mass of propellant. It would be less mass, and less complicated than the current shock absorber.

Somehow, there seems to be a way to get the shock waves to cancel one another. The "perfection" of the current geometry must be altered carefully for this to happen.