Bogus sim files posted online - why?

[Dane Ronnow]

[Deleted]

 

[ThirstyBarbarian]

ALL builders should manually check the CG . . .

Dave F.

I always recommend to manually weigh the finished rocket, and then balance it to find the Cg. You can’t rely on calculated weight or Cg, and that’s true for manual calculations or the sim files. Once you have measured the real-world, as-built mass and Cg, you enter those into the sim programs as global overrides for the flight sims.

 

[jqavins]

On this point are are in full agreement. Pre-build CG estimates are good for planning purposes… recommended procedure is to always measure and override mass and CG before making final flight calculations.

On this point I'm in 90% agreement. After years of wondering and asking "Why one caliber (or better 10% of total length)?" I finally found my answer in the Galejs paper. But why 1.25 to 1.5 calibers (or better 12.5% to 15% of total length)? Because the CG and CP are both still approximations, and I like to add some computational margin. While it's undeniably a good idea to manually measure final mass and CG, the small movements of CG that will be detected are darn near always not enough to take a rocket with 12.5% stability margin and make in unstable.

I think. And that's why I usually don't manually check CG, and this has thus far never caused me a problem. (I did have a CG problem once, but that was because I made a design change to a kit and failed to check CG at all, in reality or in silico.)

I always recommend to manually weigh the finished rocket, and then balance it to find the Cg. You can’t rely on calculated weight or Cg, and that’s true for manual calculations or the sim files. Once you have measured the real-world, as-built mass and Cg, you enter those into the sim programs as global overrides for the flight sims.

And if you feel like getting deep into the weeds, that's still not good enough. Global overrides and phantom mass objects don't lead to accurate determination of the longitudinal moment of inertia, which is part of what determines the rate at which a stable rocket recovers from a perturbance. If that's not optimized, it can make a small but noticeable difference in the altitude at apogee, potentially enough to be important in competition. And if it's really large, it can contribute to a rocket becoming unstable despite an ample static margin.

 

[neil_w]

On this point I'm in 90% agreement. After years of wondering and asking "Why one caliber (or better 10% of total length)?" I finally found my answer in the Galejs paper. But why 1.25 to 1.5 calibers (or better 12.5% to 15% of total length)? Because the CG and CP are both still approximations, and I like to add some computational margin. While it's undeniably a good idea to manually measure final mass and CG, the small movements of CG that will be detected are darn near always not enough to take a rocket with 12.5% stability margin and make in unstable.

"Small movements of CG" is doing a lot of work in that sentence. It can move a lot, depending on the design and how you build. And I don't think it's fair to characterize measured CG at liftoff to be "an approximation". It may not be perfect but if should be pretty darn close, more than close enough for the needs here.

In addition to the issue of unaccounted-for glue and paint and whatnot, it is also quite possible to discover quite late in the game that you have a wrong material entered into a key component which is throwing off CG estimates. A measurement of the whole rocket will essentially correct *all* such errors in one quick shot.

For some builds there may be lots of designed-in margin, and maybe you really are unlikely to see any real-world effects of correcting CG with measuring. But it is best practice to always do so.
Adding additional computational margin etc. all good... but the only way to really understand your margins is to start with the most accurate center point and then expand your margin from there. That means measure.

I think. And that's why I usually don't manually check CG, and this has thus far never caused me a problem. (I did have a CG problem once, but that was because I made a design change to a kit and failed to check CG at all, in reality or in silico.)

In rocketry as in life, there are endless occasions where bypassing best practice does not lead to mishap, most of the time.

 

[SolarYellow]

I'll just note again that, by building with a gram scale, weighing each component and assembly before and after gluing more stuff together or painting, and accounting for glue and paint with appropriately-placed mass objects, I get really, really close to where the c.g. measures out on the finished rocket. I can usually nail it in the sim by shifting some paint weight around a little, rather than relying on an override.

 

[Banzai88]

I'll just note again that, by building with a gram scale, weighing each component and assembly before and after gluing more stuff together or painting, and accounting for glue and paint with appropriately-placed mass objects, I get really, really close to where the c.g. measures out on the finished rocket. I can usually nail it in the sim by shifting some paint weight around a little, rather than relying on an override.

With that level of granularity, most reasonable folks would expect no less of a result than what you report.

For the rest of us, and for a vast majority of rocketeers who were in rocketry well before sims were a 'thing' that we argued about over the internet, thankfully rocketry is much, much more forgiving, and other than academic exercise or personal need for such granularity or certain competition or fringe ultra high performance projects, global overrides are more than sufficient.

If it wasn't, then them good 'ole boys sure were lucky all them decades of (successfully) throwing big rockets into the sky (without computer sims)!

 

[Ez2cDave]

"Small movements of CG" is doing a lot of work in that sentence. It can move a lot, depending on the design and how you build. And I don't think it's fair to characterize measured CG at liftoff to be "an approximation". It may not be perfect but if should be pretty darn close, more than close enough for the needs here.

In addition to the issue of unaccounted-for glue and paint and whatnot, it is also quite possible to discover quite late in the game that you have a wrong material entered into a key component which is throwing off CG estimates. A measurement of the whole rocket will essentially correct *all* such errors in one quick shot.

For some builds there may be lots of designed-in margin, and maybe you really are unlikely to see any real-world effects of correcting CG with measuring. But it is best practice to always do so.
Adding additional computational margin etc. all good... but the only way to really understand your margins is to start with the most accurate center point and then expand your margin from there. That means measure.

In rocketry as in life, there are endless occasions where bypassing best practice does not lead to mishap, most of the time.

Static CG, once determined, moves forward ( in most designs ), as propellant is consumed, during powered flight and the Coasting Phase.

Dave F.

 

[neil_w]

Static CG, once determined, moves forward ( in most designs ), as propellant is consumed, during powered flight and the Coasting Phase.

Dave F.

Of course, and the sim accounts for that. The key is to establish the most accurate possible starting point.

 

[cls]

Static CG, once determined, moves forward ( in most designs ), as propellant is consumed, during powered flight and the Coasting Phase.

Dave F.

not always. you assume a motor short enough to be aft of the rocket + empty motor CG. long motors with mass fwd of the unloaded CG have different behavior, especially considering the grain geometry, end burning vs core vs stepped etc.

 

[Seatechnerd83]

not always. you assume a motor short enough to be aft of the rocket + empty motor CG. long motors with mass fwd of the unloaded CG have different behavior, especially considering the grain geometry, end burning vs core vs stepped etc.

that assumes the same amount of propellant mass is on both sides of the CG which is VERY rare, possible (maybe) but improbable.
Edit: also assumes that the dry mass of the motor with casing and nozzle is equal on both sides of the CG

 

[jqavins]

"Small movements of CG" is doing a lot of work in that sentence. It can move a lot, depending on the design and how you build. And I don't think it's fair to characterize measured CG at liftoff to be "an approximation". It may not be perfect but if should be pretty darn close, more than close enough for the needs here.

In addition to the issue of unaccounted-for glue and paint and whatnot, it is also quite possible to discover quite late in the game that you have a wrong material entered into a key component which is throwing off CG estimates. A measurement of the whole rocket will essentially correct *all* such errors in one quick shot.

For some builds there may be lots of designed-in margin, and maybe you really are unlikely to see any real-world effects of correcting CG with measuring. But it is best practice to always do so.
Adding additional computational margin etc. all good... but the only way to really understand your margins is to start with the most accurate center point and then expand your margin from there. That means measure.

In rocketry as in life, there are endless occasions where bypassing best practice does not lead to mishap, most of the time.

I don't disagree. I know that measuring is best practice, and when I don't it's only after due consideration. And that ends up being most of the time, not all the time.

Also, if I stated that measured CG is "only" an approximation then I misstated what I meant. There's (basically) never a measured CP, i.e. that is always an approximation, albeit a good one. And the CG measurement is not altogether perfect either, but yes, it's very very close.

We all (even I) measure the CG (when I measure it) with the heaviest motor installed that we expect to use. But if, some time down the road, you decide to (for example) shove an E12 where a D12 was intended and just let it overhang by the extra inch, do you recheck the CG there at the field, and have the CP number available to compare it to? If so, hats off to you (not sarcastic). I know that I'm not alone in taking a few moments to think about it and deciding "I've got plenty of margin; it'll be fine" or "I'm not so sure of the margin; better not risk it."

 

[neil_w]

We all (even I) measure the CG (when I measure it) with the heaviest motor installed that we expect to use. But if, some time down the road, you decide to (for example) shove an E12 where a D12 was intended and just let it overhang by the extra inch, do you recheck the CG there at the field, and have the CP number available to compare it to? If so, hats off to you (not sarcastic). I know that I'm not alone in taking a few moments to think about it and deciding "I've got plenty of margin; it'll be fine" or "I'm not so sure of the margin; better not risk it."

I override Cg of my model without the motor. Enter all possible motors into OR to verify stability, performance, everything. Choose from among those motors. The number of non-reloadable options is sufficiently constrained that I am not generally confronted with a "surprise" motor at the field.

Further, I know approximate relative weights of motors, and if I'm using a motor that is within weight ranges of other "OK" motors, I'd be OK go to with it. I'll use judgment at the field as needed.

But it all *starts* with a properly measured and overridden model, at least for me.

 

[bad_idea]

We all (even I) measure the CG (when I measure it) with the heaviest motor installed that we expect to use. But if, some time down the road, you decide to (for example) shove an E12 where a D12 was intended and just let it overhang by the extra inch, do you recheck the CG there at the field, and have the CP number available to compare it to? If so, hats off to you (not sarcastic). I know that I'm not alone in taking a few moments to think about it and deciding "I've got plenty of margin; it'll be fine" or "I'm not so sure of the margin; better not risk it."

I mark Cp on the rocket and check final Cg before flight. Knowing Cp and Cg is actually a requirement for Tripoli: "the flier shall document the location of the center of pressure and be able to demonstrate the center of gravity."

 

[JoePfeiffer]

That is true . . . Of course, with the "tolerances" for actual motor performance in play, the finest sim or equation is, at best, an estimate.

Looking at BOTH Barrowman & the Sims, none of them calculate the "CP shift" when velocity varies, above Mach 1, either.

Open OpenRocket. Load the "Simple Model Rocket" example. Open the "Component Analysis" tab. Move the velocity slider. Watch the CP move.

I expect RS does this too, and of course RASAero does.

Basically, a rocket is either stable, or it is not . . .

Or it's stable under some conditions and not others.

Personally, I think sims are fine for calculating the CP of a rocket . . . However, I would never trust the CG, as calculated by a Sim. I always manually check the CG, no matter what, and never say "ok, the computer says the CG is right here, so we're good to go" . . .

The only reason I don't also check CP and CD against the finished rocket is I don't own a wind tunnel.

In closing, my main objection to sims, in general, is that so many rocketeers "blindly" rely on them, without verifying the CG. My advice is to ALWAYS check the CG, manually . . . always !

Anyone who doesn't verify all they can before launch isn't following the developers' advice.

 

[Tractionengines]

if, some time down the road, you decide to (for example) shove an E12 where a D12 was intended and just let it overhang by the extra inch, do you recheck the CG there at the field, and have the CP number available to compare it to?

I always mark the OR approximation of the CP. Then right before going to RSO, balance it on "something" as a quick check. If CG (balance point) seems too close to CP... repack recovery, or smaller motor, or more careful checks.
No measurement tools needed!

 

[Ez2cDave]

not always. you assume a motor short enough to be aft of the rocket + empty motor CG. long motors with mass fwd of the unloaded CG have different behavior, especially considering the grain geometry, end burning vs core vs stepped etc.

Hence, the reason I said "most designs" . . .

Hybrid motors often had some of the most extreme rearward CG shifts, in my experience.

Dave F.

 

[NateB]

I mark Cp on the rocket and check final Cg before flight. Knowing Cp and Cg is actually a requirement for Tripoli: "the flier shall document the location of the center of pressure and be able to demonstrate the center of gravity."

I usually mark the CP and minimum CG when I paint my rocket and then demonstrate the CG with the prepped motor when asked.