3" Tailcone Aerodynamics Research Rocket

[Chuck Rogers]

One question, though: why is the CD so much higher for hexagonal beveled fins with LE .5" and TE .25" than for rounded? At Mach 0.3, the CD for rounded is 0.388, while for the hexagonal fins as described above it's 0.484, a dramatic difference. The peak CD is 0.619 for hexagonal, but only 0.525 for rounded.

Going to really get into some details here on fin airfoil drag.


In RASAero the subsonic rounded fin airfoil drag is based on Estes TIR-11 Figure 38. The subsonic rounded fin airfoil drag is not a function of Reynolds number or friction coefficient, it's only a function of the thickness to chord ratio. Whether the flow is laminar-transition to turbulent-turbulent or all turbulent (full turbulent) the subsonic rounded fin drag is the same.

On your runs you had the All Turbulent Flow box checked.

So you were comparing a rounded fin airfoil (same drag whether laminar or all turbulent) with an all turbulent hexagonal airfoil.

The better comparison is a rounded fin airfoil compared to the hexagonal airfoil with the All Turbulent Flow box Not Checked (laminar-transition to turbulent-turbulent flow).
Then at Mach 0.30 with a rounded fin airfoil (0.082 in fin thickness) the rocket power-off CD is 0.348.
At Mach 0.30 with the 0.50 in/0.25 in hexagonal airfoil (0.082 in fin thickness) the rocket power-off CD is 0.363. Much closer.

The hexagonal and NACA airfoil subsonic fin drag equations in RASAero are much more accurate than the rounded airfoil model. They are more theoretical and less semi-empirical. The rounded airfoil model is probably off by more than 0.02 (the difference in CD).



Comparing the hexagonal airfoil with a 0.50 in leading edge and a 0.25 in trailing edge diamond airfoil length to the rounded (again with the All Turbulent Flow box Not Checked) at the peak (transonic) CD:

Rounded fin airfoil (0.082 fin thickness) - Mach 1.05 - Rocket CD = 0.491
0.50 in/0.25 in hexagonal airfoil (0.082 fin thickness) - Mach 1.05 - Rocket CD = 0.478


RASAero has the rounded and square fin airfoils because some people use them. But the subsonic and supersonic fin CD models are more accurate for hexagonal, NACA, biconvex, and diamond airfoils. There's just better theoretical approaches available, and less semi-empiricism in the models.


Actually, it would probably be even better to use a hexagonal airfoil on the rocket rather than rounded. It will be a more accurate CD prediction.


Chuck Rogers
Rogers Aeroscience

 

[Chuck Rogers]

Made a correction using the Edit feature. Original correction was here, so ignore this post.

Chuck Rogers

 

[CarVac]

Here's what I did yesterday (yesterday as in the 4th).

Today (as in the 5th) I sanded.

Tomorrow (as in today, the 6th) I'll glaze coat it and take an after-photo.

 

[CarVac]

Glaze coated. The pits are still there; I'll have to sand down the surface and recoat it to really fill in those indentations. Or maybe I won't; the surface finish is already better than the first rocket's fins, and that one beat the simulations.

Here's a view of the fincan from farther away:

Man, the lab is a mess right now...

As you can see, we reinforced the fins diagonally to resist twisting flutter. Though it really might not need it; the Apogee formula is clearly wrong, at least for this fin geometry. This layup is considerably stiffer than the first rocket's, which didn't flutter at all, and the Apogee formula says that the new fins will flutter at Mach 0.56.

 

[accpool]

that looks really nice. are you going to be able to fly that Saturday?

Have your minions clean up the mess,

 

[CarVac]

that looks really nice. are you going to be able to fly that Saturday?

Have your minions clean up the mess,

Yes, I intend to fly on Saturday. All we have to do with the booster is glue in the tailcone attachment ring.

Then we have to re-glue avionics mounting points onto the sled.

 

[BLKKROW]

Friendly Bump?

Crossing Fingers.

 

[nikitynick]

Anything happen with this? I'd be interested in the findings if there is a significant difference with various tail cones on this scale.

 

[CCotner]

Both vehicles were destroyed in recovery failures and no data recovered.

 

[TopRamen]

Tail Cones are awesome!

 

[CarVac]

Both vehicles were destroyed in recovery failures and no data recovered.

These were what I consider to be my greatest failures in rocketry, simply because I have absolutely no clue how the parachutes didn't open during the 14k ft descent without cable ties.

 

[Onebadhawk]

Carlo,
Did you find evidence of the e charge still intact....
Did you see it descending,,, was it separated......
This is what happened to my 3" Intimidator,,,,,,,,
The altimeter browned out on launch........

Teddy

 

[CarVac]

Carlo,
Did you find evidence of the e charge still intact....
Did you see it descending,,, was it separated......
This is what happened to my 3" Intimidator,,,,,,,,
The altimeter browned out on launch........

Teddy

It most definitely separated at apogee. The cable cutter's cable was ripped by the deployment shock.

I found the rocket on the ground, with a rolled-up parachute next to it, shroud lines extended.


IMG_5008-output by CarVac, on Flickr

Only a month afterwards did I realize that since I was flying at Black Rock, single-deploy from 15,000 feet would have been no issue...

 

[lkal32]

The cable cutter's cable was ripped by the deployment shock.

Carlo,

If you never found the cable remains, it is probably impossible to know, but do you think the cable tie actually severed at deployment or do you think that the parachute bundle somehow pulled out from the cable's grasp? I understand you have your hesitations about the cutter on short, fat bundles, but unless you did not include enough slack in the e-match lines, there should be only negligible force trying to separate the two.

Needless to say, the chute not opening and being completely unhindered is baffling.

 

[Onebadhawk]

That really is a good one,,,,,
aside from everything else,,,
the chute coming down that far and never inflating,,,,,,
you got me,,,,,

Teddy

 

[CarVac]

Carlo,

If you never found the cable remains, it is probably impossible to know, but do you think the cable tie actually severed at deployment or do you think that the parachute bundle somehow pulled out from the cable's grasp? I understand you have your hesitations about the cutter on short, fat bundles, but unless you did not include enough slack in the e-match lines, there should be only negligible force trying to separate the two.

Needless to say, the chute not opening and being completely unhindered is baffling.

The cable tie was ripped, but the remains were still through the cable cutter. If you go to the Flickr page, you can right-click on the image, and view the original size. The Cable Cutter is up and just to the left of the parachute, and you can kinda see the cable tie.

 

[lkal32]

The cable tie was ripped, but the remains were still through the cable cutter. If you go to the Flickr page, you can right-click on the image, and view the original size. The Cable Cutter is up and just to the left of the parachute, and you can kinda see the cable tie.

Interesting. Was it still the full length? I am curious if the tag end may have been cut too short (if you cut it at all) and that allowed it to be pulled through.
Was the cutter attached to the rocket at all? I dont see any cord connecting them, and if not, it is interesting that it and the rocket landed so close together if it separated at apogee.

 

[CarVac]

Interesting. Was it still the full length? I am curious if the tag end may have been cut too short (if you cut it at all) and that allowed it to be pulled through.
Was the cutter attached to the rocket at all? I dont see any cord connecting them, and if not, it is interesting that it and the rocket landed so close together if it separated at apogee.

The whole thing was there; it was just ripped.

The cutter was no longer attached to the rocket.

 

[Buckeye]

Given your flight vehicle difficulties, wouldn't wind tunnel or CFD testing be a better way to get the drag data you seek? Are these tools available at your university?

 

[CarVac]

Given your flight vehicle difficulties, wouldn't wind tunnel or CFD testing be a better way to get the drag data you seek? Are these tools available at your university?

CFD could, certainly, but you'd really want to calibrate the parameters with real-world data, hence the need for empirical measurement anyway.

I'd need a supersonic wind tunnel for the data I want, and supporting it would be difficult since what I wanted to measure was at the back (where most of these things are typically supported in windtunnels).

If I'd started the project with the experience I had at the end, I would have made a recovery system that worked, and I would have flown many times and collected good data. However, I still wouldn't know exactly why the recovery failed as it did, and that unsettles me.

 

[FredA]

Yep - should have built a "conventional" DD rocket and ditch those cable-cutters.
They are a high failure item...

Why try to be so clever and ruin your experiment?
Would have been a great data set to have....

 

[CarVac]

Yep - should have built a "conventional" DD rocket and ditch those cable-cutters.
They are a high failure item...

Why try to be so clever and ruin your experiment?
Would have been a great data set to have....

I was young and naive... I had had good results with them right up until this rocket, so I had no reason to doubt them.

The reason not to do a "conventional" DD rocket is performance: I would have had to have more tubes, likely from FWFG instead of the ultra-lightweight CF tubes we found, and it would have been longer, thus reducing the top speed.

EDIT: Also, I have never once flown a "conventional" DD rocket.

 

[bandman444]

Yep - should have built a "conventional" DD rocket and ditch those cable-cutters. They are a high failure item... Why try to be so clever and ruin your experiment? Would have been a great data set to have....

Woah woah woah. Totally an opinion.

They are not a "high Failure item"

 

[GDJ]

Yep - should have built a "conventional" DD rocket and ditch those cable-cutters.
They are a high failure item...

Why try to be so clever and ruin your experiment?
Would have been a great data set to have....

Woah woah woah. Totally an opinion.

They are not a "high Failure item"

Agreed.

 

[watheyak]

Yep - should have built a "conventional" DD rocket and ditch those cable-cutters.
They are a high failure item...

I'm around 20 for 20 with the Archetype cable cutters...

Scott

 

[FredA]

I see them fail all the time....more often than not...
Not particularly Archetype, but other cutters.
I wouldn't trust my experiment to them when the trade off is about a foot of rocket length...especially when you just care about delta's....
YMMV...but the OP was 0 for 3...which is a shame.

 

[CarVac]

I see them fail all the time....more often than not...
Not particularly Archetype, but other cutters.
I wouldn't trust my experiment to them when the trade off is about a foot of rocket length...especially when you just care about delta's....
YMMV...but the OP was 0 for 3...which is a shame.

Actually, I'm 2 for 5 (and was 2 for 2 prior to these rockets).

Secondly, I believe I was configuring my deployment setup incorrectly for both. My 2 successful flights were on a much smaller rocket, where the effects of momentum from the deployment charge would be smaller.

 

[FredA]

0 for 3 or 2 for 5 ....both suck....
KISS

 

[CarVac]

0 for 3 or 2 for 5 ....both suck....
KISS

KISS? Nah.

It's a matter of keeping every part of the system within my capabilities to understand. Without high framerate video of what happened at deployment, there's no way for me to understand exactly why and how the cable cutter failed.

And then there's my experience from Bare Necessities to throw in the mix now: the deployment bag + ARRD method is what I plan on designing for now. It worked even when it shouldn't have (somehow it deployed the main even without a drogue parachute...), the opposite of my experience with the cable cutter. While it's more complicated parts-wise than a cable cutter (which is actually very simple), I can say that absolute trust in every single individual component is a sufficient condition for trust in the whole system. The descent is moderated by the drogue, with the parts hanging down vertically. The release of the drogue letting to act as the pilot is done by the ARRD, which is an incredibly robust and essentially perfectly reliable device. Then, the deployment sequence is moderated by the deployment bag, which requires almost no force to remove the parachute from.

A simple system? No.

Simple components that interconnect in a simple manner? Yes.



I understand what you're saying: I myself don't blindly trust cable cutters anymore. On that point, you are preaching to the choir.

However, I plan to never once use conventional dual deploy on one of my own rockets. In my opinion, it's boring and suboptimal for performance, and there are more elegant solutions with equivalent reliability.

 

[FredA]

However, I plan to never once use conventional dual deploy on one of my own rockets. In my opinion, it's boring and suboptimal for performance, and there are more elegant solutions with equivalent reliability.

Wow -- If I was your prof and I read this, I would fail you for sure....you botched the experiment because you didn't want to use a tried and true, but "boring," configuration....

Your experiment wasn't about optimal performance, but performance variations...
Epic fail, IMHO