lakeroadster
When in doubt... build hell-for-stout!
I modify the component weights as I build the rocket. Therefore the CG is very close to the built model.
Last edited:
Let's put an end to the "Base Drag Hack"
- Thread starter Buckeye
- Start date
Help Support The Rocketry Forum:
The BDH is obviously a way to correct an issue with the simulation software not handling reality correctly and get it back closer to real world.
If you want to eliminate the BDH, you need to correct the simulation software, or at least define which programs don't need the hack and which ones do.
You can't have it both ways. If neither the sim or the hack is right, tell us what is.
If you want to eliminate the BDH, you need to correct the simulation software, or at least define which programs don't need the hack and which ones do.
You can't have it both ways. If neither the sim or the hack is right, tell us what is.
boatgeek
Well-Known Member
- Joined
- Dec 27, 2014
- Messages
- 7,413
- Reaction score
- 7,960
Having re-read the PoF articles just now, I'm kinda changing my mind on the BDH. If you look at #1 in the series, the Fat Boy the author has modeled has a stability margin 13% of its length without the BDH. That's less than a caliber, but well within the rules of thumb for short or long rockets. So for rocket-shaped objects, the BDH may not even be necessary. I can see its usefulness for oddrocs (spools, cones, etc.). The question is where to find the line between the two. I suspect that the difference is whether the airflow at the aft end of the rocket is flowing along the tube or is still being pushed outward. Looking at two extremes, the air at the aft end of a Mean Machine (and probably a Fat Boy too) is going to be more or less parallel to the body tube. On a spool, the airflow is going to be going outboard pretty significantly.
Taking @lakeroadster's excellent designs as a reference, I would tend to put R2 and Red Columbine in the oddroc category, since it's pretty plausible that the hemisphere-to-slightly-flat nose cones induce airflow outboard. The L
ratio is also approaching 1. The Warhawk is right on the line, with an L around 3 and a pointy enough nose cone that you'd get airflow down the side. The fact that it swing tests well clearly means that reality needs to adjust itself to my theory. So I guess I'd come down on the line that it's not time to put the BDH out to pasture, but really to define what kinds of projects it is applicable to, and also to discourage its use when just looking at percentage of length resolves any perceived stability issues.
Indeed, physics tells us a rocket will be technically stable if the (current) CG is forward of the (current) CP at any moment in the flight (assuming there's enough air pressure to stabilize). Rules of thumb are introduced to add margins for dynamic shifting of either centre points and whatnot.
I guess that if (for a short fat rocket not expected to go seriously supersonic) if the modelling is expecting the CG to move forward as propellant is consumed, it might be safe to assume the stability margin won't tend to reduce from take-off conditions, so tighter margins could be adopted [insert disclaimers, caveats, flame suits here].
TP
- Joined
- Aug 6, 2022
- Messages
- 3,272
- Reaction score
- 2,969
Short-fats are also going to be relatively less affected by AoA factors, vs. the effect they have on long-skinny rockets.
boatgeek
Well-Known Member
- Joined
- Dec 27, 2014
- Messages
- 7,413
- Reaction score
- 7,960
I think what we really need (short of major wind tunnel time) is to get a sense of how much the stubbies’ CP moves in crosswinds. If it’s reliably a similar percentage of length as “normal” rockets, then we can maybe move to percentage rather than calibers for many rockets.
- Joined
- Aug 6, 2022
- Messages
- 3,272
- Reaction score
- 2,969
The challenge is selling your RSO on switching.
If any violation of the Baroman assumptions is detected, the simulation should just fall over and play dead.
- Joined
- Aug 6, 2022
- Messages
- 3,272
- Reaction score
- 2,969
If any violation of the Baroman assumptions is detected, the simulation should just fall over and play dead.
Does that include the "very small angle of attack" assumptions? Because horizontal wind at rail clearance can get there.
I don't have to know what is wrong to or how to fix it to think something is fishy. Fortunately most people are cautious. I have attempted to point out some issues and theories. The Levison paper was just a first pass at this after all, so surely the BDH could be tweaked.
The R2-D2 is an extreme example. OR shows the CP completely in front of the rocket, which is bizarre, there is some form of fin on the back of the thing after all. With BDH that moves to a behind the airframe, but that also seems unbelievable. Fortunately someplace in between works, and my eyeball sim even slightly favors the BDH. Actually trying a similar shape in OR, I can see a possible issue with how it models the nosecone on such a short tube and the effect of the "tailcone" in what is almost immediately turbulent air was ridiculous, try it. I think the base drag has some effect and should be considered a component, but really?
Levison wrote: "Various aerodynamic texts confirm the fact that the dynamic Center-of-Pressure (CP) of flat plate lying perpendicular to a flow, lies behind the plate along its central axis, due to a base vortex that forms when the air begins flowing over its surface."
The CP at a slight angle of attack of a flat plate moves away from the center towards what has become the leading edge of a wing. The EFFECT of that can be described as the CP being behind the plate. However this plate has a front side and a back side and it seems to me they're both involved. Or at least the shape of the front side of anything similar is more significant. The back end of a tube is not the same as a plate rocket. But even if the drag is just centered on the base, it still has a stabilizing force, just not that strong and its point of effect is not behind the rocket.
@Buckeye --
That is a BEAUTIFUL FatBoy model you posted !
I am still working on getting FreeCAD working with FemGUI for the Rocket WorkBench on my Slackware Box so I haven't been able to follow along in your CDF with FreeCAD thread ( yet ).
Looking back at all the code I've compiled from source, I would have been way ahead by making the AppImage version work
Anyhow ...
I am curious: How do the stagnation regions change on the FatBoy at non-zero angles of attack ?
Especially the symmetric stagnation point on the base of your FatBoy Model:
I wonder if there could be a significant normal force on the rocket at non-zero AoA due to asymmetrical pressure gradients on the base of the rocket ?
But then what happens when the motor is disturbing that vacuum ?
Thanks for sharing your work !
-- kjh
That is a BEAUTIFUL FatBoy model you posted !
I am still working on getting FreeCAD working with FemGUI for the Rocket WorkBench on my Slackware Box so I haven't been able to follow along in your CDF with FreeCAD thread ( yet ).
Looking back at all the code I've compiled from source, I would have been way ahead by making the AppImage version work
Anyhow ...
I am curious: How do the stagnation regions change on the FatBoy at non-zero angles of attack ?
Especially the symmetric stagnation point on the base of your FatBoy Model:
I wonder if there could be a significant normal force on the rocket at non-zero AoA due to asymmetrical pressure gradients on the base of the rocket ?
But then what happens when the motor is disturbing that vacuum ?
Thanks for sharing your work !
-- kjh
- Joined
- Sep 5, 2009
- Messages
- 3,538
- Reaction score
- 1,675
Indeed. The back end of a rocket body tube is not the same as a flat plate in isolation. The front of the plate has drag as well and sets up the flow separation angle.
Here is a simple comparison on the FatBoy. The body tube base drag Cd from OpenRocket (0.132) vs. Hoerner's flat plate (1.17).
The Hack is too powerful and pulls the CP too far rearward.
Last edited:
- Joined
- Sep 5, 2009
- Messages
- 3,538
- Reaction score
- 1,675
This is 4 degrees AoA nose up in the z-direction. I had to shift the Cp scale a bit to show the non-symmetry developing on the top of the base. Also, the counter-rotating vortices in the wake are becoming of different size.
I will look at some more angles, say 0 to 20 degrees, but I want to change my mesh scheme a bit to accommodate.
From my observations of low L/D flights, the low pressure stagnation at the base overrides most non-zero AoA flight issues.
The high velocity exhaust gases from the motor creates a lower base pressure through the venturi effect than what is depicted. I'm placing sensors on the motor mount to record the pressure variations between the motor induced venturi and base stagnation.
@Spacedog49Krell --
I am REALLY looking forward to your data !
I don't believe it's ever been done on a model rocket before !!
-- kjh
WoW !
My eyeball is telling me that if I was to integrate pressure over the tail of that FatBoy there might be just a little bit of normal force, even at a 4-deg AoA !
Imagine a MiniMag !
Maybe that explains @Rschub's YEET ?
-- kjh
lakeroadster
When in doubt... build hell-for-stout!
So on all flying rockets, while the motor is thrusting, the thrust is helping to minimize the base drag, until coast phase, at which time the base drag increases?
Rschub
Well-Known Member
The exhaust fills the space behind the tube and acts like a fairing, as long as the motor is producing thrust.
For a large diameter tube, this fairing effect is smaller such as a Minni Magg with an "H" motor. So, you get appreciable base drag, and therefore some stability.
When you bump up to a "J" motor the fairing effect is stronger, which will reduce base drag and thus stability.
The ultimate source of the "yeet" are what aerodynamicists call "non-linear aerodynamic effects" which is a fancy way of saying a behavior that is inherently chaotic in nature and does not follow a predictable, equation driven model.
John --
Apparently so and it makes sense if the motor exhaust is filling the void behind the rocket with exhaust gas.
RASAero produces pairs of CD -vs- Mach -- Power Off CD -vs- Mach and Power On CD -vs- Mach.
Check out some of @Chuck Rogers data here: RASAero II Comparisons with Flight Data
-- kjh
lakeroadster
When in doubt... build hell-for-stout!
Those are very long rockets... The BDH is for rockets that are less than a 10:1 length to diameter ratio, the majority of which are not flying at or even near Mach 1.
I respectfully fail to see the value of this data.
How does induction stability factor into this? If the engine thrust pulls air in around the base of the rocket how does this factor into things?
Assuming that base drag and induction can both have stabilizing effects there should be cases (between a flying disk and a Big Daddy) where CG is behind CP (via traditional calculations) but the rocket flies stable (or more or less stable).
Also it seems that previous statements that "if CG is in front of CP, the rocket is stable" might be an overstatement. Both percent-based and 1-caliber-based assessments of stability are on a continuum (no hard and fast limit) and it still seems that the base drag hack provides some correction that is valuable to understand "how stable" a rocket is absent more complex modeling in the software readily available to everyone.
Ultimately, all of this exploration is really interested but we need to make Open Rocket and Rocksim useable solutions for laypeople to calculate stability in a wider range of rockets.
Assuming that base drag and induction can both have stabilizing effects there should be cases (between a flying disk and a Big Daddy) where CG is behind CP (via traditional calculations) but the rocket flies stable (or more or less stable).
Also it seems that previous statements that "if CG is in front of CP, the rocket is stable" might be an overstatement. Both percent-based and 1-caliber-based assessments of stability are on a continuum (no hard and fast limit) and it still seems that the base drag hack provides some correction that is valuable to understand "how stable" a rocket is absent more complex modeling in the software readily available to everyone.
Ultimately, all of this exploration is really interested but we need to make Open Rocket and Rocksim useable solutions for laypeople to calculate stability in a wider range of rockets.
There are two events occurring during powered flight, a low pressure turbulent flow area around the downstream edge of the rocket with an under expanded nozzle exhaust attempting to fill this low pressure void. Where the two join is a very chaotic but interesting region. To study it, will require development of new sensor probes.
What started my interest in measuring base pressure issues, were finless cone rockets of the 1960's. If the cone rocket was too close to a flat flame deflector they appeared to freeze on the pad when launched. The exhaust gases deflecting in a circular pattern created a low-pressure region under the rocket.
lakeroadster
When in doubt... build hell-for-stout!
Yes, but this thread is about "Lets Put An End To The Base Drag Hack". So talking about things that aren't related clouds the issue.
Maybe so, John ...
But I wonder ...
IF @Buckeye's CFD models -and-or- @Spacedog49Krell's instrumentation studies were to identify and then quantify a normal force acting at the base of a low aspect ratio rocket at a non-zero angle of attack, THEN MAYBE the BDH could be 'baked into' the stability calcs for that class of rocket which otherwise violates the assumptions of the classical Barrowman Equations.
Then we would all be one step closer to "Let's Put an end to the Base Drag Hack" for that class of rockets ...
OTOH, I believe there will always be some rockets where the BDH needs to be applied and in that case, the BDH will continue to work as it now exists because it is after all only a cone appended to the tail of the rocket and the the Normal forces were well defined by Barrowman for that shape.
-- kjh
But I wonder ...
IF @Buckeye's CFD models -and-or- @Spacedog49Krell's instrumentation studies were to identify and then quantify a normal force acting at the base of a low aspect ratio rocket at a non-zero angle of attack, THEN MAYBE the BDH could be 'baked into' the stability calcs for that class of rocket which otherwise violates the assumptions of the classical Barrowman Equations.
Then we would all be one step closer to "Let's Put an end to the Base Drag Hack" for that class of rockets ...
OTOH, I believe there will always be some rockets where the BDH needs to be applied and in that case, the BDH will continue to work as it now exists because it is after all only a cone appended to the tail of the rocket and the the Normal forces were well defined by Barrowman for that shape.
-- kjh
Then "we" need to stop implying that base drag is a stabilizing force.
