Supersonic Tailcones in SImulators

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

CCotner

Well-Known Member
Joined
Sep 4, 2012
Messages
888
Reaction score
3
I'm starting this based on discussions with Casey Barker and Kurt Von Delius, and also brief interactions with Ken Biba. I want to know about people's flights with extreme tailcones and boat-tails (more than 10 degrees of total included angle and more than a 25% reduction in base area or somewhere around those numbers) at supersonic (>M1.2 for >1sec) speeds. I want quantitative comparisons between simulator data and recorded flight data.

From looking at both data comparisons and underlying physics assumptions, it seems apparent to me that RASAero is basically the only hobby-accessible rocket simulator that has any hope of accurately predicting supersonic flight information. OpenRocket seems to do slightly better than Rocksim. I do not consider Rocksim Pro to be hobby-accessible for it's $1000 pricetag, although I have heard mixed things about it's accuracy.

The allegation is that while RASAero accurately predicts supersonic CD and CP-shift in most cases, it exaggerates the contributions of extreme tailcones (as defined above), leading to inflated altitude performance (such as designes I have made for a J510 flying case rocket reaching 32,000', or more topically, the predictions of my and cjl's N5800CS MD rockets reaching 120,000' plus compared to the ~60,000' expectations of Bandman444 and others).

So, as the internet was once misquoted,

SHUT UP AND GIVE ME YOUR DATA!

:)
 
I can't contribute flight data, but an illustrative example about RASAero and extreme tailcones. A minimum diameter rocket launched on M2020 simulated to an altitude of 29751ft. Changing the parameters "Boattail Length" and "Boattail Base Diameter" from 0.0000 to 0.0001 increases the altitude to 38194ft - an increase by 28% for a practically negligibly change in airframe geometry.

RASAero apparently changes between mathematical models for different configurations. I guess it is safe to assume that the model without boattail is reasonable accurate. Most rockets don't have a boattail and RASAero wouldn't have its reputation for accuracy otherwise. Therefore, one might conclude that RASAero's model for boattails starts to diverge from reality as the boattail becomes more extreme.

My example above is not a critique of RASAero or the models behind it. Nearly every, if not every, mathematical model of physical processes can be broken by feeding it absurd input data. It would be interesting to hear Chuck's opinion about how far one can stretch the input parameters without compromising accuracy too much.

Reinhard
 
Hi,

No data to contribute, but as I've written the aerodynamic methods for OpenRocket, I have to say that I wouldn't consider them reliable in your case. :) The methods are mainly based on subsonic methods, and hackishly extended to supersonic speeds (in order to keep my Master's in a reasonable scope).

However, I have noticed that even if the aerodynamic simulation is way off during part of the flight, it doesn't necessarily affect the primary figures of flight (e.g max altitude of velocity) very much, as it's integrated with the more-valid data of the rest of the flight. Based on the literacy comparisons I did I assumed the accuracy of OR for supersonic flights to be much worse than it apparently is.

Cheers,
Sampo N.
 
Yeah, It's usually decent if the time spent above transonic is small. Out N5800CS spend ~25 seconds above supersonic, though. =p

I will probably PM you sometime about OR; I'm interested in properly extending it to supersonic flow and low hypersonics, I just don't know the programing end of it.
 
I'm starting this based on discussions with Casey Barker and Kurt Von Delius, and also brief interactions with Ken Biba. I want to know about people's flights with extreme tailcones and boat-tails (more than 10 degrees of total included angle and more than a 25% reduction in base area or somewhere around those numbers) at supersonic (>M1.2 for >1sec) speeds. I want quantitative comparisons between simulator data and recorded flight data.

From looking at both data comparisons and underlying physics assumptions, it seems apparent to me that RASAero is basically the only hobby-accessible rocket simulator that has any hope of accurately predicting supersonic flight information. OpenRocket seems to do slightly better than Rocksim. I do not consider Rocksim Pro to be hobby-accessible for it's $1000 pricetag, although I have heard mixed things about it's accuracy.

The allegation is that while RASAero accurately predicts supersonic CD and CP-shift in most cases, it exaggerates the contributions of extreme tailcones (as defined above), leading to inflated altitude performance (such as designes I have made for a J510 flying case rocket reaching 32,000', or more topically, the predictions of my and cjl's N5800CS MD rockets reaching 120,000' plus compared to the ~60,000' expectations of Bandman444 and others).

So, as the internet was once misquoted,

SHUT UP AND GIVE ME YOUR DATA!

:)


<< SHUT UP AND GIVE ME YOUR DATA! >>

Well, here it is! :)

On the RASAero web site (www.rasaero.com) go to the Comparisons with Wind Tunnel Data tab, and you can check out the RASAero CD comparisons with NASA TR R-100 free-flight model data with boattails of varying length and base area.

Note though that as noted on the plot, the RASAero supersonic CD predictions are not valid for rb/R < 0.5. I'd further add that the RASAero supersonic CD predictions are not valid for boattails more extreme than the boattails for which comparisons are shown on the plot (not just boattails with rb/R < 0.5, but also extremely short, steep boattails).

Stay within the range of boattails on the plot, and the CD predictions should be accurate, although the comparisons only go up to Mach 1.2 (the general limit of the data), but transonic/low supersonic Mach numbers are where boattails provide the greatest benefit in drag reduction.

Also note from the TR R-100 CD data that extremely steep boattails have higher drag than less steep (normal) boattails. This is because the increase in wave drag from the boattail being extremely steep is larger than the reduction in base drag from the additional reduction in base area for an extremely steep boattail (assuming it's steepness allows you to reduce the base area for a given length).


Chuck Rogers
Rogers Aeroscience
 
Look up 'Design of Aerodynamically Stabilized Free Rockets' and then reference 5-47 - "Boattail Sections" with some data graphs shown on 5-185 through 5-187.

The 5-47 Sections reads as follows:

Design of Aerodynamically Stabilized Free Rockets said:
The purpose of the boattail is to decrease the drag of a body that has a squared off base. This squared off
base has local pressure on its surface much less than the free-stream pressure and, therefore, produces a
large drag force. When the exit diameter of the rocket nozzle is smaller than the body-cylinder diameter, the
afterbody maybe tapered to form a boattail to reduce the base area and thus reduce the base drag. However,
for a given boattail length, decreasing the base diameter increases the boat tail wave drag. Therefore, an
optimum configuration can be strongly influenced by the boattail design. This is clearly illustrated in Fig.
5-120 (Ref. 80) which shows the variation in the total zero-lift drag of a 10.5-cal body having a 3.0-cal
tangent-ogive nose, a 6.0-cal cylindrical afterbody, and a 1.5-cal conical boattail with various base
diameters. In subsonic flow, decreasing the base diameter continues to decrease the total drag. In transonic
and supersonic flows, however, decreasing the base diameter to a ratio of dbt/d = 0.4 or lower results in an
increase in the total drag. In supersonic flow, as the Mach number increases, the influence of the base drag
decreases rapidly and thus allows larger base diameters to achieve a minimum drag boattail design.
Fig. 5-121 presents theoretical subsonic and transonic boattail wave drag for various boattail angles and
diameter ratios (Ref. 81). Figs. 5-120 and 5-121 show that subsonic drag is relatively unimportant when
compared to transonic and low supersonic drag.
The supersonic wave drag of conical and parabolic boattails is presented in Figs. 5-122 and 5-123,
respectively. No analytical method or suitable parametric experimental data exist for the accurate
prediction of boattail wave drag at subsonic and transonic speeds. If experimental data for a particular
configuration cannot be found, it is suggested that supersonic data be extrapolated to peak value at a Mach
with a sharp reduction to a lower value at subsonic speeds.
The lift of a boattail is in the negative direction, and although the usual boattail is relatively very short
and thus decreases the lift very little, it does have a strong destabilizing effect—i.e., the boattail tends to
move the overall body center of pressure forward quite markedly. This means that a larger stabilizing
device is required to stabilize the rocket. Thus the increase in drag of the larger stabilizing device may
negate the saving in drag realized by the boattail.
Accurate determination of boattail drag is rather difficult because of the dependence on the geometry of
the fore and aft portions of the body and the real fluid effects. However, for preliminary design analysis,
results from the correlation of both theoretical and experimental data should be used when available.
 
<< SHUT UP AND GIVE ME YOUR DATA! >>

Well, here it is! :)


Chuck Rogers
Rogers Aeroscience

OH! *facepalm* I hadn't read the graph carefully enough clearly. Sorry for asking a dumb question. I'll be reading over that carefully tonight or this weekend if I can find the time.
 
I can't contribute flight data, but an illustrative example about RASAero and extreme tailcones. A minimum diameter rocket launched on M2020 simulated to an altitude of 29751ft. Changing the parameters "Boattail Length" and "Boattail Base Diameter" from 0.0000 to 0.0001 increases the altitude to 38194ft - an increase by 28% for a practically negligibly change in airframe geometry.

RASAero apparently changes between mathematical models for different configurations. I guess it is safe to assume that the model without boattail is reasonable accurate. Most rockets don't have a boattail and RASAero wouldn't have its reputation for accuracy otherwise. Therefore, one might conclude that RASAero's model for boattails starts to diverge from reality as the boattail becomes more extreme.

Reinhard

In RASAero when a rocket component is not entered, the component dimensions stay at the default dimensions of 0.0000 (as an example, for a boattail). But by entering 0.0001 for boattail length and boattail base diameter an extremely steep, short boattail was entered, this is not a neglible change in geometry. You'd see this on the RASAero View Rocket drawing, although you can't really see the boattail as it's so short and steep. This short and steep boattail would have increased the rocket altitude by a large amount (like the 28% quoted), except that the RASAero supersonic boattail predictions should not be used for rb/R < 0.5.

What you'd really want to do for a, say, 4 in diameter rocket, would be to add a small boattail (boattail length 0.01 in, boattail base diameter 3.99 in), and then progressively make the boattail longer and make the base diameter smaller. As a small boattail is first present, and then gets longer and has a smaller base diameter, the RASAero altitude predictions and aero data predictions blend in smoothly with the no-boattail data.

As noted in the CD comparisons with the NASA TR R-100 free-flight model data, the limitation in the RASAero supersonic boattail CD predictions is that the model is not accurate for rb/R < 0.5, and is not accurate for boattails which are significantly steeper than the boattails in the TR R-100 plot. If the boattail is a reasonable boattail and rb/R is not less than 0.5, then RASAero will give accurate CD predictions and altitude predictions for the rocket.

Several of the rockets in the RASAero Comparisons with Altitude Data Table on the web site have boattails, and the RASAero altitude prediction accuracy for these rockets was good, though these were short boattails, with shallow (not steep) boattail angles.


Chuck Rogers
Rogers Aeroscience
 
Back
Top