Predicting fall rates for separated deployment

Hi Guys,

In my never-ending effort to over-think things, I hit upon a design problem; how to design the two halves of a separated rocket so that the resultant two parts (tied together by a shock cord) will fall from apogee in a flat "spin".

I've toyed with using standard parachute rate-of descent formulas, using the surface area of each part and a common Cd, and the results seemed to make sense. My Green Weenie was a test of this concept and it appeared to fall as predicted. But, a single occurance does not a statistically relevant event make... could have been luck. (the second flight was with a drogue, and I didn't like the drogue's effect. Too much twisting and end swaping for my comfort.)

So, I was curious if anyone out there had refined the process of designing the two separated halves to flat spin, or even experimented in that direction.

Ashley

OOkkaayy...

So no one has any input...

Well - this weekend I did some testing. On two flights with the same rocket (my rebuilt Elipse), same configuration, but slightly different motors (I211W and I245G), I had an interesting result.

On the first flight, the rocket went to 1900+ feet, separated at apogee, and fell in a nice flat 'spin'. (As a side note, the chute fouled on deployment (the nomex pad covered the shroud lines) and the rocket hit the ground in a nice sequence of parts with only minor dimpling of the drogue compartment tube lip. The Elipse design with dynawind tube is a tough cookie.)

The MAWD altimeter data curve showed the fall rate to be 50 fps.

On the second flight (I245G) the rocket went to 1414' and separated perfectly. Again, it settled into a flat fall. This time the chute opened properly (hey - I learned a lesson...) and the descent under chute was 25 fps. But the curve also showed that the fall rate was 70 fps!

So - same rocket, same weight, same configuration, same temp (within 3 degrees), had two significantly different fall rates.

Anyone have any guesses as to why?

Ashley

Crickets chirping....

uncle_vanya said:

Crickets chirping....

Click to expand...

I don't think enough information was given to draw a conclusion .....

It could be many things including differences in ...

• What sections of the curves were used to calculate the descent rate
• The attitude of the rocket when the ejection charge fired at apogee
• How long it took the drogue 'chute to open
• Air density because of humidity or temperature

-- Roger

jadebox said:

I don't think enough information was given to draw a conclusion .....

It could be many things including differences in ...

• What sections of the curves were used to calculate the descent rate
• The attitude of the rocket when the ejection charge fired at apogee
• How long it took the drogue 'chute to open
• Air density because of humidity or temperature

-- Roger

Click to expand...

I think both of these flights cited were drogueless. So time to open is not relevant.

Air density is unknown but both flights were from the same location within 3 degrees surface temp and flown to 2K and 1500 roughly. Not a lot of reason to expect 20fps difference in fall rates.

I suspect curve fits and orientation of the rocket could be involved - but hard to know.

Brad's correct -

Both flights were drogueless. Sorry - I meant to state that directly in the post.

Atmospheric data is roughly the same - not different enough to create a 40% difference in fall rate.

When examining the curves (and superimposing them on each other), The difference in fall rates is exceptionally obvious. (I'll post the two graphs when I get home this evening.)

The only reason I can think of for the disparity relates to a possible aft shift of the fin can's CG because of the smaller motor casing of the I245G vs. the I211W. It's possible that the resultant CG shift of the fin can caused it to become slightly more aerodynamic, i.e. point downward slightly and thus fall more quickly. (Thus the "orientation" point in your response, Brad) But the CG shift would be small - probably not more than an inch on a 21" fin can. Maybe I should do a quick CG-CP comparison on the fin can by itself. Of course, I'm clueless on how to model the fact that it's connected to the upper portion of the rocket by a 17' nylon cord.

I took video of both flights, but I didn't get much footage of the second flight's fall before main ejection. I should probably get a better cameraman.

Ashley

Ashley

The drag for each section of the rocket is proportional to the side-on cross-sectional area of the halves.

What may be useful is to locate the CG and the CP of each half. If the CG and the CP are located in the same place, the rocket should fall flat. If the CG is offset from the CP then the rocket will assume a non-horizontal attitude and will have a lower aerodynamic cross-section and therefore less drag and will fall faster. The shock cord should be fully extended to keep the halves separated if the CGs are >= or outward of the CPs.

For the same rocket weight, it requires a 40% drag reduction (in cross-section) to increase the fall rate from 50 fps to 70 fps.

Bob

Thanks, Bob. I thought it might work along those lines. I did a quick simulation for each of the halves and found that the motor casing size only shifted the CG of the fin can by .2", and the CG and CP are separated by two inches - a change that shouldn't effect the fall rate via stability by 40%. So I'm still scratching my head.

I'm attaching the two altimeter graphs for comparison.

Ashley

The second (steeper) curve appears to have some wave-like variation in the slope which also appears to be damped to some extent. Your CG may have moved only .2" but the moment of inertias of the fin-can may have changed enough so that it was more prone to porpoising. If that's the case then your "reference area" is changing with time and that could produce a quicker decent. I know that a .2" shift in the cg of a free-flight model airplane could drastically alter rate of decent.

...just a theory