How much stability for short stubby rockets?

[bjphoenix]

I'm thinking of a scratch build that starts with a Big Daddy and shortens it quite a bit. In a quick OR model I see 0.6 caliber stability (1.8" for a 3" diameter airframe). This doesn't seem like it would be enough but it would be very difficult to get 1.5 diameter of stability so how much is really needed?

 

[rharshberger]

In this case .6 should be plenty, short fat rockets tend to be that way due to a stronger base drag effect, of course transonic velocities change that number.

 

[BigMacDaddy]

Simulate in OpenRocket and use the base drag trick to simulate the base drag for a short-stubby rocket. Basically put a 0g mass cone the same diameter as body tube and 3.14 x diameter long sticking off the back of the model.

https://www.apogeerockets.com/education/downloads/Newsletter154.pdf

Here is my Little Joe II rocket with base drag simulated showing it should be stable with 7x 18mm C-6 engines in the current configuration.

 

[hobie1dog]

I wanted to ask this same question

 

[RocketScientistAustralia]

Simulate in OpenRocket and use the base drag trick to simulate the base drag for a short-stubby rocket. Basically put a 0g mass cone the same diameter as body tube and 3.14 x diameter long sticking off the back of the model.

https://www.apogeerockets.com/education/downloads/Newsletter154.pdf

Here is my Little Joe II rocket with base drag simulated showing it should be stable with 7x 18mm C-6 engines in the current configuration.

View attachment 521729

If base drag is there all the time, why not just include it automatically in open rocket?

 

[neil_w]

If base drag is there all the time, why not just include it automatically in open rocket?

The base drag *is* already there, which is why the cone hack should be removed for flight simulation. What is *not* there is the effect on CP of the base drag. I am hoping that it can be incorporated into OR in some fashion in the future.

 

[lakeroadster]

The base drag *is* already there, which is why the cone hack should be removed for flight simulation. What is *not* there is the effect on CP of the base drag. I am hoping that it can be incorporated into OR in some fashion in the future.

But, depending on the design, if the base drag hack is removed the rocket simulation may be so unstable you don't get a good sim.

Which is weird... if it is already there, why isn't it changing the CP?

 

[neil_w]

But, depending on the design, if the base drag hack is removed the rocket simulation may be so unstable you don't get a good sim.

As we have previously discussed, you sometimes need to override the CG location using an override to get a working flight sim.

Which is weird... if it is already there, why isn't it changing the CP?

The effect of base drag on CP is a separate effect that must be explicitly modeled in (and currently it is not). I am not sure if there is a lot of "official" literature on it. The hack that was (I believe) originally described in an Apogee Peak-of-Flight seems to work pretty well, and I would personally be in favor of building it into OR, to be enabled at the user's discretion (not everyone believes in it).

 

[RocketScientistAustralia]

Would it be easy for OR to do a quick calculation of the L/D ratio and flag to the user about the base drag hack, like the jagged fins warning.
Not problem solved, but highlighted to the user at least. Easy quick win.

 

[neil_w]

At that point there’s no reason not to just build the hack into the program. Which, for the record, I am in favor of.

 

[RocketScientistAustralia]

At that point there’s no reason not to just build the hack into the program. Which, for the record, I am in favor of.

My suggestion can be implemented now. Seems like you guys are still discussing the detail and appropriatness of it..

 

[bjphoenix]

I had the idea to build a much shortened Big Daddy some years back because it could be made to look like something from old scifi. I did an OR simulation back then and it looked like stability would be a problem. I had the idea to try again so I found an .ork file for a Big Daddy and did some modifications to it. This one looked somewhat promising. If it needs a bit of nose weight that would be possible. I'm going to pursue this a little bit more to see what I can get out of it, one of the keys will be fin shape to add some stability and still fit in with my concept of the design. The difference between the first trial and now is that I'm adding more fin area.

 

[RocketScientistAustralia]

Doing a swing test always gives you a difinitive answer if it flies stable. It can easily be done on any rocket up to a few Kg. Depends how much space you have.
If it swing tests unstable it MAY still be stable to launch, but RSO decision needed anyway.... With short rockets you also really need to watch out for acceleration at take off packing things backward and moving the CG into danger level.
So high g motors need to be considered carefully.

 

[lakeroadster]

As we have previously discussed, you sometimes need to override the CG location using an override to get a working flight sim.

A working flight sim? What does that mean?

Overriding the CG location results in a simulation that doesn't reflect the actual flight condition of the rocket. That flies in the face of why a simulation is performed in the 1st place, accuracy matters. By moving the CG, much of the other flight data generated by OR is no longer correct. I fail to understand the logic in doing so?

I can't find any place in the Apogee Newsletters, 154 - 158 - 162, where it specifies to remove the “massless” conical-transition when the simulation is run, yet you continue to tell folks to do so?

I did find however that Bruce Levinson states the following on Page #2 of Newsletter 162:

:Refresher: This and the previous two articles in this series are based upon the assumption that the dynamic center of pressure (CP) of fl at plate lying perpendicular to a fl ow, lies about 2.2 diameters behind the plate behind the plate along its central axis, due to a base vortex that forms when the air begins fl owing over its surface. This is a conservative estimate for the CP value since other mathematical extensions of the wind tunnel data seem to indicate the CP may even be further aft! Again, any inaccuracy in this CP value will drastically affect the simulation results using this approximation."​

I interpret this to mean the simulation data generated by Open Rocket, with the a “massless” conical-transition left in place, provides a more accurate overall simulation of the rocket flight. The CG is correct and the CP is correct.

 

[BigMacDaddy]

I think, although I could definitely be wrong, that Neil means base drag is already taken into account in altitude calculations and air speed but not CP / stability calculations. That would mean the cone would reduce flight speed and altitude unrealistically when it is on there...

 

[lakeroadster]

That would mean the cone would reduce flight speed and altitude unrealistically when it is on there...

Minor variations of speeds and altitude, based on the sim's I've run. But the point is, the simulation reflects the actual as built rocket.

Here's an example:

 

[QFactor]

I'm thinking of a scratch build that starts with a Big Daddy and shortens it quite a bit. In a quick OR model I see 0.6 caliber stability (1.8" for a 3" diameter airframe). This doesn't seem like it would be enough but it would be very difficult to get 1.5 diameter of stability so how much is really needed?

Is the 0.6 caliber before you add the motor or after you add the motor - but just before you run the simulation.

If the 0.6 is after you add the motor - then what is the stability caliber before you add the motor?

 

[pathtouch]

For what it’s worth I used the massless cone transition on my 7.5” LOC doorknob. With the cone it simmed 750 feet on a CTI J381, and it flew 768 feet according to my Eggtimer Quantum.
Stability was 1.33 as modeled and it was definitely a stable flight. Without the cone I think I had 0.83 calibers of stability.

 

[lakeroadster]

.... then what is the stability caliber before you add the motor?

I've seen other folks ask this? I am curious why you would care what a rockets stability is without a motor? It doesn't fly without a motor, well unless something goes horribly wrong?

And the simulation program will output stability margin of the rocket, see below, from ignition to burnout if your wondering what the stability is during the flight.

Note this is not the rocket being discussed above.

 

[neil_w]

I've seen other folks ask this? I am curious why you would care what a rockets stability is without a motor? It doesn't fly without a motor, well unless something goes horribly wrong?

I agree with this. Unloaded stability is usually irrelevant.

When creating a model in OR, I try to get add the motor mount and configure a motor as early as possible so while I'm designing I'll be seeing numbers that matter.

 

[QFactor]

I agree with this. Unloaded stability is usually irrelevant.

When creating a model in OR, I try to get add the motor mount and configure a motor as early as possible so while I'm designing I'll be seeing numbers that matter.

The stability margin is not static during the launch. You are loosing mass as the motor burns, and so your
stability margin increases the second you light the motor. With reloadable motors you basically left with
just the weight of the case and liner. Also, you have mininal Base Drag as the motors burns because of
Base Bleed, also known as a Power On & Power Off condition. The full affect of base drag kicks in once
there is no longer an exhaust plume, but you no longer have the full weight of the motor.

I ask what the stability is with and without the motor to see if there is a minimal or extreme difference.
Reloadable motors can be challenging with stubby rockets. I have a few of my own stubby rockets.

In RockSim I plot the stabililty margin, and other parameters, over time to see where it is about the
time it is leaving the rail. Sometimes the margin is acceptable just as it leaves the rail. Some stubby
rockets will have that split-second wobble as they exit the rail and then suddenly they straighten up
and fly dead straight. So is it beacuse they reached a stabilizing velocity? Or they lost enough motor
mass to improve the stability margin? Or is it both?

 

[lakeroadster]

...So is it beacuse they reached a stabilizing velocity? Or they lost enough motor
mass to improve the stability margin? Or is it both?

It could be a lot of things, depending on the rocket design. Typically the issue is insufficient speed, or the transition from having external mechanical stability, to actual stability and / or windcocking.

The motor propellant mass lost from ignition, to the end of the rail, is miniscule in regard to effecting the stability of the rocket.

But none of this explains why you would want to know stability with no motor at all? The motor case will still be there, and that weight is significant.

 

[bjphoenix]

Is the 0.6 caliber before you add the motor or after you add the motor - but just before you run the simulation.

If the 0.6 is after you add the motor - then what is the stability caliber before you add the motor?

Well it is easy enough to answer the question for discussion sake.
WITH a D12-3 the stability is 0.611 cal. Without a motor the stability is 0.89 cal.

This sim is a pretty stubby rocket, almost all nose cone. Length of the body plus nose is about 11" total, with fins swept back another 3.5". This is a crude sim at this point, just trying to look at some ideas I've had for awhile. I picked a Big Daddy sim off the internet and started modifying it, I would have to buy a kit and weigh components to see how close it would get to the numbers in the sim. Mass without motors is 4 oz.

Many, many years ago I built the stubby thing below to fly on G64 reloadable. I'm not going to dig it out to take measurements, it is based on LOC 4" components that I had leftover from another build. I put in nose weight and did a swing test on it before flying it, it flies pretty well on a G64, probably hasn't been flown in 20 years. But overall it's not even as stubby as the stock Big Daddy.

 

[QFactor]

Well it is easy enough to answer the question for discussion sake.
WITH a D12-3 the stability is 0.611 cal. Without a motor the stability is 0.89 cal.

This sim is a pretty stubby rocket, almost all nose cone. Length of the body plus nose is about 11" total, with fins swept back another 3.5". This is a crude sim at this point, just trying to look at some ideas I've had for awhile. I picked a Big Daddy sim off the internet and started modifying it, I would have to buy a kit and weigh components to see how close it would get to the numbers in the sim. Mass without motors is 4 oz.

Many, many years ago I built the stubby thing below to fly on G64 reloadable. I'm not going to dig it out to take measurements, it is based on LOC 4" components that I had leftover from another build. I put in nose weight and did a swing test on it before flying it, it flies pretty well on a G64, probably hasn't been flown in 20 years. But overall it's not even as stubby as the stock Big Daddy.

May want to do a swing test on your new stubby design.

If you move ahead on the new stubby be sure to post your findings. It would be interesting to see how it turned out.

 

[jqavins]

If the 0.6 is after you add the motor - then what is the stability caliber before you add the motor?

I've seen other folks ask this? I am curious why you would care what a rockets stability is without a motor? It doesn't fly without a motor...

Yes, that. (See Lake, we agree on something. )

As for which matches the actual flight condition, with the cone or without, the answer is obviously that neither one does. One adds drag to the sim that's not present in flight, and the other leaves out stabilizing force that is present in flight. And that's why, if I an correctly reading his mind, Neil advocates doing two versions of the OR (or RS) model. One with the cone gives a better representation of the stability than the model without. Note the distance between the reported CP and CG. Then remove the cone, giving more accurate drag. Note the new CP position and move the CG forward so that the distance between CG and CP is the same as it was with the cone.

Does that perfectly reflect reality? Of course not, because the CP is in the wrong place and the moment of inertia is now wrong. But that will have much subtler and less important effects on the simulation results than the other options.

Remember, base drag doesn't really change the CP. The CP is the imaginary point through which the sum of aerodynamic lift forces can act to produce the same net result as the combined lift forces in their actual positions. Let me emphasize a key word: lift. Base drag isn't a lift force, so it doesn't, strictly speaking, affect the CP. What it does is provide stability, but by a different means. How much does it improve stability? Well, by about as much as the stability would improve if the CP were yay much behind where it actually is. So we approximate the stabilizing effect of base drag by forcing the model's CP aft by yay much, but the factual CP isn't really any different with base drag than it would be in a magical world where base drag doesn't exist. What we've got with the cone in place is an effective pseudo-CP. And it's an approximation.

And here's another thing.

Base drag isn't the most important thing here. 0.6 cal is probably fine, and would be in that magical, base dragless world too.

Why 1 cal? If the CG is forward of the CP then the rocket will right itself and be stable; what does "marginally stable" mean? (As applied to rockets that is; applied to our average mental state it's an entirely different thing.)

The CP calculations performed by OR and RS, and by people using the Barrowman equations by hand, do not take body lift into account. Body lift will cause the (actual) CP to migrate forward with increasing angle of attack (AoA). The amount that it moves depends a lot on the tube's length. For relatively "normal looking" rockets it will generally not move more than one diameter. So if you've got better than one caliber static stability margin you can be confident that the CP will not catch up to the CG during flight. The thing is, when a rocket is stubby, the CP migration will be slight, and your 0.6 caliber margin is probably fine. A stock Mean Machine has a stability margin of - well, I don't remember, but it's a lot. It needs to be a lot because there will be a lot of CP migration with AoA.