Simulation of Ring Fin Designs in Open Rocket

[lakeroadster]

In Apogee's "Peak of Flight" Newsletter" Issue 27 Bruce Levison discusses "Simulation of Ring Fin Designs Using RockSim v4.0"

I followed Bruce's instructions and tried this in Open Rocket on an Estes Alpha (A model created by @K'Tesh ).

It seems to yield good results but without some actual real world testing, it's hard to know just how accurate the simulation is with respect to the flight data.

 

[Dotini]

In Apogee's "Peak of Flight" Newsletter" Issue 27 Bruce Levison discusses "Simulation of Ring Fin Designs Using RockSim v4.0"

I followed Bruce's instructions and tried this in Open Rocket on an Estes Alpha (A model created by @K'Tesh ).

It seems to yield good results but without some actual real world testing, it's hard to know just how accurate the simulation is with respect to the flight data.

Levison's discussion does not seem to rise anywhere near to the level of peer-reviewed or established science.

I, it seems uniquely, have tried real world testing to verify or falsify his claims. My real world tests appear to show his claims to be seriously wanting. But I need and intend to do more testing. So far, it seems no other rocketeer beside myself has committed the time, money and effort into investigating his claims. If anybody out there wants to build a model to test Levison's claims, I would be happy to help with my data, and send you the plastic ring material if you can't afford it.

 

[Jeff Lassahn]

Could you tell us a bit about your experiment design and data so far? What strategy do you recommend for testing claims about CP?

 

[Dotini]

Levison claims his studies have made RockSim a useful and valuable tool relative to ring tail design. However, he admits he has done no verification with actual field tests and data. He says he's leaving that up to folks like me. Last December when I turned 72, I decided to start building model rockets as an anodyne to boredom. I wound up building some ring tails and flew them to what seemed to be anomalous altitudes, altitudes that were ~4 times higher than what was predicted on RockSim in the case of the 6th Observable.

These are some of my ring tails:

I've documented the build and flights of these and other models here in the Scratch Built thread, including some video. I've recently acquired an altimeter and will continue my tests using that tool. That's the only strategy I have.

 

[Jeff Lassahn]

OK, so you are not concerned so much about the stability estimates as you are about the altitude estimates.
Things that can contribute to the altitude estimate being bad include:
The rocket mass estimate being wrong
The actual surface finish (paint quality, rounding of fin edges, etc) being different from the estimate.
The estimate of form drag caused by the rocket shape being wrong.

It might be useful to do some analysis to find out where the problem is coming from. Some possible tests that might provide enlightenment:
Check that the rocket mass reported by the program matches the real measured mass, and check that the mass estimates for the fake fins intended to model the ring fin match the mass of the actual ring.
Check how close the estimated altitudes are for a non-ring-fin design that otherwise uses the same construction techniques and materials, to make sure the problems are all about the ring and not some other unrelated factor.
See how much the program's estimated altitude changes for the same design with just a ring fin added or removed. If just adding a ring to a design makes the predicted altitude drop by close to a factor of 4 that seems like a pretty good sign that something is wrong with the program.

 

[Dotini]

OK, so you are not concerned so much about the stability estimates as you are about the altitude estimates.
Things that can contribute to the altitude estimate being bad include:
The rocket mass estimate being wrong
The actual surface finish (paint quality, rounding of fin edges, etc) being different from the estimate.
The estimate of form drag caused by the rocket shape being wrong.

It might be useful to do some analysis to find out where the problem is coming from. Some possible tests that might provide enlightenment:
Check that the rocket mass reported by the program matches the real measured mass, and check that the mass estimates for the fake fins intended to model the ring fin match the mass of the actual ring.
Check how close the estimated altitudes are for a non-ring-fin design that otherwise uses the same construction techniques and materials, to make sure the problems are all about the ring and not some other unrelated factor.
See how much the program's estimated altitude changes for the same design with just a ring fin added or removed. If just adding a ring to a design makes the predicted altitude drop by close to a factor of 4 that seems like a pretty good sign that something is wrong with the program.

No. You are not getting it. All my ring tails are designed or at least intended to be unstable with only the support fins in use. Maybe half the size they'd need to be according to Van Milligan's chapter on stability.

When I swing test the model I'm working on now, I will test it with and without the ring. If it's stable without the ring, it's a failure right at the start, and I'll wad it up and throw it away. Or at least trim the fins to past the point of instability.

IMHO, to deliberately build a rocket with a ring fin that is stable without it is a purposeless exercise in superfluity.

 

[Jeff Lassahn]

I guess I'm unclear about the goal. The kind of rocket you would build for sport flying or for altitude performance is different than the kind of rocket you would build as a test of a specific scientific hypothesis.

If your goal is to test how accurate a simulation model of ring fins is, building (or at least simulating) models that have the ability to add or remove the ring fin while leaving all other design features identical is a way to get more information about how the ring fin simulation in particular is performing.

 

[Dotini]

I guess I'm unclear about the goal. The kind of rocket you would build for sport flying or for altitude performance is different than the kind of rocket you would build as a test of a specific scientific hypothesis.

If your goal is to test how accurate a simulation model of ring fins is, building (or at least simulating) models that have the ability to add or remove the ring fin while leaving all other design features identical is a way to get more information about how the ring fin simulation in particular is performing.

What is sport flying? I don't build for altitude, as my field is only 500 x 500. I build to test for and demonstrate horizontal recovery with Magnus Effect. I also build rockets that would be unstable if it weren't for a ring fin. I try to go places where there is no beaten path. I'm retired from using computers for a living. If I live to be 100 I will not use a RockSim. I am not a rocket scientist. But I'm building rockets which seemingly do not conform to any known science. I think that is hilarious good fun. (But I have signed the pledge to follow the safety code!) If you would like to know the science, then get to work building experiments. It's lonely out here on the cutting edge of rocket science.

 

[JoePfeiffer]

The lack of validation in both RockSim and OpenRocket is concerning (I'll note that while I know the situation in OR, my impression is that RS is the same).

The only real way to validate the CP results is with a wind tunnel. A comparison with CFD results would give more confidence, but of course it pushes the validation question back a step. Pursuing either of these would make several good MS projects.

It would be possible to run a series of experiments to test altitude predictions, though. The sort of huge discrepancies you're describing seem really big.

 

[neil_w]

The only real way to validate the CP results is with a wind tunnel. A comparison with CFD results would give more confidence, but of course it pushes the validation question back a step. Pursuing either of these would make several good MS projects.

Well, there are two separate issues here.... CP and drag. CP doesn't seem to be the big issue here, it's all about drag. Although drag would also benefit from a wind tunnel analysis, it could also presumably be done with CFD (let's put @Nytrunner to work) or by actually analyzing flight data. To really do it thoroughly would require a bunch of different rocket configurations and a lot of flights, but I would guess that an assortment of well-instrumented flights on a single rocket configuration as @Dotini has built would take us a long way. Since OR doesn't have ring tail support built in, we'd have to compare with "simulated" versions, using some fin configuration we calculate to be equivalent.

Although I've never fully bought into the Levison method (and I don't use it in my own OR sims), I'm really surprised that in this case we're seeing such a huge divergence between sim and actual. I wonder if there's actually a bug in the Rocksim code for ringtails (even if Levison's method isn't totally correct, it shouldn't be *that* far off).

 

[jrap330]

Levison claims his studies have made RockSim a useful and valuable tool relative to ring tail design. However, he admits he has done no verification with actual field tests and data. He says he's leaving that up to folks like me. Last December when I turned 72, I decided to start building model rockets as an anodyne to boredom. I wound up building some ring tails and flew them to what seemed to be anomalous altitudes, altitudes that were ~4 times higher than what was predicted on RockSim in the case of the 6th Observable.

These are some of my ring tails:

I've documented the build and flights of these and other models here in the Scratch Built thread, including some video. I've recently acquired an altimeter and will continue my tests using that tool. That's the only strategy I have.

Not getting it to anyone's response to your thread....I like to state...Your models looks great, beautiful work.