An experiment with ring tail simulation in OpenRocket

[neil_w]

OpenRocket famously (?) doesn't support ring tails, or tail rings, or ring fins, or whatever you want to call them. It's easy enough to use an internal tube to produce a proper visual representation and mass/CG calculation but what to do about CP and drag has been a bit of a mystery. I just tried a quick experiment and the results are interesting.

Here's what I've been doing until now. Given:

od = diameter of ring​

id = diameter of tube inside ring​

len = length of ring​

th = thickness of ring​

I have been creating what I think of as a "ring equivalent" set of trapezoidal fins with the following dimensions:

root cord = len​

tip chord = len​

sweep length = 0​

thickness = th​

height = od - id​

number of fins = ((od-th)*PI) / height​

​

Let's talk about those last two. What I am aiming for here is a set of fins that has the exact same total area as the ring, with the same average distance from the inner tube. This is entirely based on intuition and I have no idea if how accurate it might be. I have to think it's at least "in the ballpark" for typical ring configurations. It has worked for the ring designs I have done so far, where "has worked" means the rockets have been stable; I have no easy way to know how accurate my calculations are. Note that I only apply this when the difference between ID and OD is large enough that it seems like there will actually be airflow (this eliminates, for example, the two rings at the back of the Sirius Eradicator).

In my OR files I usually make those fins *almost* transparent, so I can just barely see that they're there but they don't ruin the 3D renders.

It occurred to me recently to try a different approach: use a single tube fin to represent the ring. In order to position it properly I need to attach it to a single pod with a phantom body tube.

I created a trivial design and compared the two approaches.

First, with the ring-equivalent fin set. Key metrics are CP = 10.685", and Cd = 1.454:


Now the tube fin, which yields CP = 10.651 and Cd = 1.152:


Mass and CG come out equal, which is no surprise. More interesting is that the CP comes out almost exactly the same (within a fraction of a percent), but drag is quite a bit lower with the tube fin. The difference in drag is large enough that I am now very interested in trying to figure out which drag calculation is likely to be closer to correct. Each approximation has its own obvious errors. Maybe the answer is somewhere in between? I suppose a comparison to Rocksim would be useful, anyone want to try it out?

Please note that all this is based on intuition and seat-of-the-pants calculations. I'm just trying to figure out a straightforward way to model my oft-used tail rings until OR supports them natively.

 

[KenECoyote]

I was just talking to a friend the other day about this when he told me that his ring finned rocket was highly unstable in OR. I suggested that he use the dimensions of the ring and convert them to fins like you did. I don't know if he's done so in Rocksim yet (he just got it).

 

[neil_w]

I was just talking to a friend the other day about this when he told me that his ring finned rocket was highly unstable in OR. I suggested that he use the dimensions of the ring and convert them to fins like you did. I don't know if he's done so in Rocksim yet (he just got it).

Ring fins do contribute enormously to stability, so you have to account for them *somehow* in OR to get a reasonable estimate of CP. Anecdotally, my various ring fin rockets fly most consistently straight, usually with no rotation. And I like the way they look.

 

[KenECoyote]

Ring fins do contribute enormously to stability, so you have to account for them *somehow* in OR to get a reasonable estimate of CP. Anecdotally, my various ring fin rockets fly most consistently straight, usually with no rotation. And I like the way they look.

Yes, that was what I was noting to him, that just based on looking at it, it should be much more stable than he was coming up with and then I recalled reading here that OR doesn't sim ring fins very well and suggested he try re-simming the ring as fins. I said that just looking at the rocket the Cp should be much further back than what he had.

Edit: It's great that you're taking the time to investigate this!

 

[neil_w]

Edit: It's great that you're taking the time to investigate this!

Unfortunately my "investigation" is completely intuitive and empirical and full of hand-waving.

 

[rharshberger]

Unfortunately my "investigation" is completely intuitive and empirical and full of hand-waving.

But it should work based on the fact the fin area of the ring fin and the fins that are standing in for the ring fin have the same rough area, and should have similar total leading edge length.

 

[mjennings]

Your approximation is very similar to what I remember people doing way back before RockSim could run tube/ring fins. There maybe an old Apogee news letter with the method. Or a Sport Rocketry article but that's gonna be about 15 years ago. Also may have one been for tube fins.I may have time to play in RockSim in the next few days.

 

[bad_idea]

I have a pair of rockets that have six fins and a ring fin. I simmed the design as having six fins, each with a fin-tip pod that has two fins, each 1/12 of the circumference of the ring. Simulation is a few seconds optimistic with respect to delay. Never did get a good altimeter reading before I killed my JL Altimiter 2 in a different rocket, so I've not gone back to correct cd in the sim.

These rockets fly straight as an arrow. If I had to guess I'd say the sim was optimistic with respect to drag and pessimistic with respect to stability.

 

[lakeroadster]

There maybe an old Apogee news letter with the method.

Simulation of Ring Fin Designs​

 

[neil_w]

I have a pair of rockets that have six fins and a ring fin. I simmed the design as having six fins, each with a fin-tip pod that has two fins, each 1/12 of the circumference of the ring. Simulation is a few seconds optimistic with respect to delay. Never did get a good altimeter reading before I killed my JL Altimiter 2 in a different rocket, so I've not gone back to correct cd in the sim.

These rockets fly straight as an arrow. If I had to guess I'd say the sim was optimistic with respect to drag and pessimistic with respect to stability.

I'd been meaning to try something like that as well. My test file doesn't have any fins, so I'll just do the pods with the fins:


Drag is about the same as the previous "fin" version, but CP is a *lot* worse. That's actually a very surprising result, and it makes me confused.

 

[David Schwantz]

Don't have OR, but I used Rocksim for the ring tail on the grain bin and it worked well.

 

[bad_idea]

Drag is about the same as the previous "fin" version, but CP is a *lot* worse. That's actually a very surprising result, and it makes me confused.

Yes, that's sort of what I was thinking in saying that I think stability is pessimistic. Then again, with six fins and a ring fin, my rocket was going to end up pretty stable even despite being short and squat,

 

[Jeff Lassahn]

Drag is about the same as the previous "fin" version, but CP is a *lot* worse. That's actually a very surprising result, and it makes me confused.

Is each of the six faces a single fin, or two fins on either side of the pod (for a total of 12 fin elements)?
It makes a huge difference.

 

[BABAR]

I have also found a surprising absence of rotation with my box fin rockets (maybe an oxymoron, but they are square rings)

I previously assumed was due to the square configuration, but your post suggests it may be something different. Maybe that misalignment of ring of a ring (if any) is less likely to cause rotation than a standard fin ( orthogonal to the rocket long axis. For rings, with the exception of the pylons the ring itself is effectively all “tip”.)

The effect of the pylon/attaching fins could be a plus or minus. At 3 or less, should act like standard fins. As the number increases and the cross-sectional area of the spaces between ring, fin, and tube get smaller would expect these to act more like tube fins, and if effective diameter to length ratio gets high enough, like closed cylinders.

Have you seen less weathercocking with rings? I’d predict less for two reasons.

1. The ring itself has a lot of effective surface area from a Barrowman standpoint, and right where you want it as far from axis as you can get (for standard fins the tip has much more “moment” force than the base).

2. Any pylon fins are effectively completely shielded from wind being internal. So up to implications above regarding overly narrowed “inlets”, more pylons give you more CP effect with zero increase in weathercocking. Not a complete free lunch, as you do pay a price for increased drag.)

 

[neil_w]

Have you seen less weathercocking with rings? I’d predict less for two reasons.

Well, among 3 rockets and about 6 flights there has been exactly zero weathercocking. A very small sample size. Also, I tend to tune my rockets to about 1.25 caliber stability, so I don't really expect a lot of weathercocking anyway. Hard to establish any sort of causal relationship here, but anecdotally the ring rockets fly really well.

 

[JackC]

OpenRocket famously (?) doesn't support ring tails, or tail rings, or ring fins, or whatever you want to call them. It's easy enough to use an internal tube to produce a proper visual representation and mass/CG calculation but what to do about CP and drag has been a bit of a mystery. I just tried a quick experiment and the results are interesting.

Here's what I've been doing until now. Given:

od = diameter of ring​

id = diameter of tube inside ring​

len = length of ring​

th = thickness of ring​

I have been creating what I think of as a "ring equivalent" set of trapezoidal fins with the following dimensions:

root cord = len​

tip chord = len​

sweep length = 0​

thickness = th​

height = od - id​

number of fins = ((od-th)*PI) / height​

​

Let's talk about those last two. What I am aiming for here is a set of fins that has the exact same total area as the ring, with the same average distance from the inner tube. This is entirely based on intuition and I have no idea if how accurate it might be. I have to think it's at least "in the ballpark" for typical ring configurations. It has worked for the ring designs I have done so far, where "has worked" means the rockets have been stable; I have no easy way to know how accurate my calculations are. Note that I only apply this when the difference between ID and OD is large enough that it seems like there will actually be airflow (this eliminates, for example, the two rings at the back of the Sirius Eradicator).

In my OR files I usually make those fins *almost* transparent, so I can just barely see that they're there but they don't ruin the 3D renders.

It occurred to me recently to try a different approach: use a single tube fin to represent the ring. In order to position it properly I need to attach it to a single pod with a phantom body tube.

I created a trivial design and compared the two approaches.

First, with the ring-equivalent fin set. Key metrics are CP = 10.685", and Cd = 1.454:
View attachment 653192

Now the tube fin, which yields CP = 10.651 and Cd = 1.152:
View attachment 653194

Mass and CG come out equal, which is no surprise. More interesting is that the CP comes out almost exactly the same (within a fraction of a percent), but drag is quite a bit lower with the tube fin. The difference in drag is large enough that I am now very interested in trying to figure out which drag calculation is likely to be closer to correct. Each approximation has its own obvious errors. Maybe the answer is somewhere in between? I suppose a comparison to Rocksim would be useful, anyone want to try it out?

Please note that all this is based on intuition and seat-of-the-pants calculations. I'm just trying to figure out a straightforward way to model my oft-used tail rings until OR supports them natively.

After reading the Simulation of Ring Fins paper from Apogee, I found that if you use 6 square fins of your ring fin diameter (say 4 inches wide and 4 inches tall) you will perfectly simulate your ring fin.

 

[JackC]

I have a review for my ring fin Galadriel on EMRR. That review explains my simulation of the ring fins I used for my L3 build.

https://www.rocketreviews.com/scratch-galadriel-by-jack-caynon.html