ANNOUNCEMENT: The OpenRocket 22.02 Beta Period is now finished

[Charles_McG]

View attachment 546666

I keep forgetting that Cd is there now and fully functional.

I went back to my sim and looked more closely. The Cp stays put when I override the Cd and the drag parameters are zero'd out. I guess I wasn't expecting the Pressure Cd to also drop to zero - but it does seem to work as advertised.

Maybe it was an earlier beta I was thinking that the Cd override wasn't done.

I like this implementation. Thank you.

 

[RGClark]

No. We don't model off-axis thrust at all.

It might actually go against ITAR restrictions since it could be used to locate landing distance/location when angled over.

Bob Clark

 

[Alan15578]

I went back to my sim and looked more closely. The Cp stays put when I override the Cd and the drag parameters are zero'd out. I guess I wasn't expecting the Pressure Cd to also drop to zero - but it does seem to work as advertised.

Maybe it was an earlier beta I was thinking that the Cd override wasn't done.

I like this implementation. Thank you.

There should also be a CP over ride for when you have actual wind tunnel data.

Ideally, the simulation should be able to use reams of wind tunnel tunnel test data for all the aero coefficients, vs, M. RN, alpha...

 

[Alan15578]

No. We don't model off-axis thrust at all.

Well, that certainly needs fixing. In particular, you need to make it easy to simulate a cluster flight when one motor fails to ignite. This is not just a useful cool feature, but an important safety consideration in a design tool.

 

[bad_idea]

Well, it's open source developed and maintained by volunteers, so if there's a new feature you want that's not been prioritized by the team, feel free to dive in and write it. That's often easier said that done, but the same is true for the developers: features are always easier to propose than to code.

Re. clusters, I won't fly one at a public launch unless I've split the cluster in OpenRocket, set at least one motor not to ignite (preferably two or more), and seen a safe launch speed in the simulation. Unless the motors of a cluster are spaced very far apart, I believe low launch speed due to failed ignition to be a more hazardous consequence than off-axis thrust, though of course the latter adds to the problems of the former.

Where space permits, I cant motors inward so that the thrust points closer to the center of gravity. I create new motor files to account for the reduction of forward thrust according to the angle they are canted and use those in the simulations.

 

[thzero]

Well, that certainly needs fixing. In particular, you need to make it easy to simulate a cluster flight when one motor fails to ignite. This is not just a useful cool feature, but an important safety consideration in a design tool.

Dive in and help out.

 

[rocketgeek101]

I'm thinking this must be a bug? When I override the mass of the stage it adds that value to the estimated mass from all the components instead of replacing it. This is with Beta 5.

 

[Jesse_Nyffenegger]

I've been having an issue where parachutes I've added to one stage are being simulated in another stage. I think it may be related to where the data is stored/cached because restarting openrocket will change the packed length of the parachutes and change which stage the parachuted is simulated in.

 

[4regt4]

I'm thinking this must be a bug? When I override the mass of the stage it adds that value to the estimated mass from all the components instead of replacing it. This is with Beta 5.

View attachment 547759
View attachment 547760

I hope this is how it's intended. That's how I compensate for paint and glue.

Hans.

 

[SiboVG]

I'm thinking this must be a bug? When I override the mass of the stage it adds that value to the estimated mass from all the components instead of replacing it. This is with Beta 5.

View attachment 547759
View attachment 547760

This is not really a bug, you just have to check the bottom checkbox "Override mass, CG and CD of all subcomponents". The stage itself doesn't have a mass, so any mass that you override it with will just be added to the total mass.

 

[SiboVG]

I hope this is how it's intended. That's how I compensate for paint and glue.

Hans.

This is indeed how I imagine that this functionality is useful.

 

[rocketgeek101]

This is not really a bug, you just have to check the bottom checkbox "Override mass, CG and CD of all subcomponents". The stage itself doesn't have a mass, so any mass that you override it with will just be added to the total mass.

Ok that makes sense. I'm having trouble figuring out what the Override CG and Override CD are doing though when the "Override mass, CG and CD of all subcomponents" is not checked. For example, adding an override of 1" to the CG does not move the CG forward from its current (i.e. calculated) position by 1" on the vehicle which is what *I think* should happen.

Something that I don't think is intended behaviour is that it seems the order in which you check the checkboxes matters. E.g. if I only check "Override mass" and then check "Override mass, CG and CD of all subcomponents" it seems to override the CD anyway. I then need to go back and check/uncheck Override CD to get it to use the calculated value by OR again.

I've demonstrated this in this video. Look at the simulated apogee at the bottom to see what I mean.
View attachment Screen Recording 2022-11-26 at 6.59.09 PM.mov

 

[H. Craig Miller]

In addition to what SiboVG has said, this feature is being replaced with the functionality shown below (the current unstable .jar). This should give more clarity regarding what is being overridden, as well as providing a wider range of options.



Until the next release, more detailed instructions on the current use of overrides may be found on the OpenRocket Wiki in the Tips and Tricks "how to" section.

 

[H. Craig Miller]

I've been having an issue where parachutes I've added to one stage are being simulated in another stage. I think it may be related to where the data is stored/cached because restarting openrocket will change the packed length of the parachutes and change which stage the parachuted is simulated in.

This issue has been reported and resolved in the current unstable, which is available to those familiar with the use of .jar files, otherwise this fix will be included in the next release.

 

[JoePfeiffer]

Ok that makes sense. I'm having trouble figuring out what the Override CG and Override CD are doing though when the "Override mass, CG and CD of all subcomponents" is not checked. For example, adding an override of 1" to the CG does not move the CG forward from its current (i.e. calculated) position by 1" on the vehicle which is what *I think* should happen.

It isn't an offset, it's an override. So it sets the CG of the component (or subassembly) to the value you select. The idea is you can measure the actual CG and enter it -- otherwise you'd have to measure it, subtract the calculated value, and enter the result.

 

[RGClark]

In this thread I discussed a short, squat nose cone seemed to be giving higher altitude than a pointed conical one in an OpenRocket sim:

https://www.rocketryforum.com/threa...te-for-a-two-stage-rocket.145740/post-1783250
Can you confirm that with the older Openrocket version. Is this still happening in the newer sim?

Robert Clark

 

[K'Tesh]

In this thread I discussed a short, squat nose cone seemed to be giving higher altitude than a pointed conical one in an OpenRocket sim:

https://www.rocketryforum.com/threa...te-for-a-two-stage-rocket.145740/post-1783250
Can you confirm that with the older Openrocket version. Is this still happening in the newer sim?

Robert Clark

Sometimes they do that... Pointy conical cones are not always the best shape aerodynamically, depending on the speed of the rocket.

 

[H. Craig Miller]

In this thread I discussed a short, squat nose cone seemed to be giving higher altitude than a pointed conical one in an OpenRocket sim:

https://www.rocketryforum.com/threa...te-for-a-two-stage-rocket.145740/post-1783250
Can you confirm that with the older Openrocket version. Is this still happening in the newer sim?

Robert Clark

This reference may help you understand nose cone fluid dynamics, and why different shapes react/perform differently at different speeds: Nose Cone Design

Also, if you post the current OpenRocket design (.ork) file, a comparison can be done between the current unstable version of OpenRocket and RASAero II.

 

[RGClark]

This reference may help you understand nose cone fluid dynamics, and why different shapes react/perform differently at different speeds: Nose Cone Design

Also, if you post the current OpenRocket design (.ork) file, a comparison can be done between the current unstable version of OpenRocket and

Thanks for the ref. The .ork file I sent through the current version of OpenRocket is attached.

Bob Clark

 

[H. Craig Miller]

The .ork file I sent through the current version of OpenRocket is attached.

Unfortunately, RASAero II is somewhat limited on the design parameters it will accept. And your design falls outside of those parameters; RASAero II will not simulate your design as it is. If you would like more information about the changes that would have to be made in order to simulate your design in RASAero II, PM me.

 

[SolarYellow]

Just figured I'd quote this exchange in this thread for people who might be looking for it.

I was trying to figure out how to get the new A3-6T delay into my sims. Here's the answer:

I've seen this behavior in Openrocket for some time, and just verified it in the current 15. 03 version. I don't see the simulation or the plot taking into account the motor delay specified in Motors & Configuration.

When selecting a motor in Motor Configs, you can manually change the delay time, either to an arbitrary number of seconds, or to a preset number of seconds defined by the motor file. The program accepts this for display purposes, but when you run simulations for, say, 6, 8, 10, or 14 second delays, the simulation calculates the deployment velocity as though the motor was the 14 second delay. The plot shows parachute deployment at 14 seconds after burnout. Openrocket calculates the optimal delay, but it won't simulate a flight with a delay set by the user close to the optimal delay.

Is this a known defect? Or am I missing something in the motors & configurations screen?

I find I have to type the override number and then tab away before closing the motor config, or it doesn't seem to stick.
You also have to tell it to use the motor ejection in the recovery section.

 

[4regt4]

Not sure if this has already been mentioned....

Re: Beta 5, Linux version running on a Chromebook.

I tried - for the first time - to use the optimization function. Was trying to see if a light rocket could use more mass. ( It didn't...) But after running the optimization, every time I tried to just run a sim on an existing configuration (say, after making a minor change on the rocket), it went FMM. That's Full Molasses Mode. The sims ran VERY slowly. Like nearly a minute. Had to save the file, close OR altogether, and restart. Then everything was back to normal.

FWIW, I had many, maybe 8 or 10, different files open. Closing the one that I ran the optimization on didn't make any difference. Had to shut everything down.

Hans.

Edit: It didn't make any difference which of the open files I tried to run a sim on, they all ran slow.

 

[thzero]

Not sure if this has already been mentioned....

Re: Beta 5, Linux version running on a Chromebook.

I tried - for the first time - to use the optimization function. Was trying to see if a light rocket could use more mass. ( It didn't...) But after running the optimization, every time I tried to just run a sim on an existing configuration (say, after making a minor change on the rocket), it went FMM. That's Full Molasses Mode. The sims ran VERY slowly. Like nearly a minute. Had to save the file, close OR altogether, and restart. Then everything was back to normal.

FWIW, I had many, maybe 8 or 10, different files open. Closing the one that I ran the optimization on didn't make any difference. Had to shut everything down.

Hans.

Edit: It didn't make any difference which of the open files I tried to run a sim on, they all ran slow.

How much RAM does your Chromebook have?
Also did you try with 15.03? Because curious if the same thing happens with the older version.

 

[4regt4]

Samsung Chromebook has 4Gb ram. Haven't tried 15.03. Basically, the Chromebook is my rocket computer - take it to launches for email and OR. Don't use it for anything else.

Before shutting OR altogether, I closed all but one of the open rocket files. Didn't speed it up at all.

Hans.

 

[SiboVG]

Not sure if this has already been mentioned....

Re: Beta 5, Linux version running on a Chromebook.

I tried - for the first time - to use the optimization function. Was trying to see if a light rocket could use more mass. ( It didn't...) But after running the optimization, every time I tried to just run a sim on an existing configuration (say, after making a minor change on the rocket), it went FMM. That's Full Molasses Mode. The sims ran VERY slowly. Like nearly a minute. Had to save the file, close OR altogether, and restart. Then everything was back to normal.

FWIW, I had many, maybe 8 or 10, different files open. Closing the one that I ran the optimization on didn't make any difference. Had to shut everything down.

Hans.

Edit: It didn't make any difference which of the open files I tried to run a sim on, they all ran slow.

Memory management in OR is not great... It can easily take a Gigabyte or two of RAM. My feeling is that opening that many files piles up the RAM usage of OpenRocket, and that closing files doesn't help because the file's memory is not properly cleaned when closing the file.

This is a know issue, and we really have to do some memory optimization after the next release...

 

[neil_w]

Yeah, such problems are usually due to memory thrashing, and with 8 designs open on a 4 GB machine you're asking for trouble. For whatever reason, memory does not seem to free up when closing designs, so I'm not surprised that didn't help.

 

[JoePfeiffer]

Just figured I'd quote this exchange in this thread for people who might be looking for it.

I was trying to figure out how to get the new A3-6T delay into my sims. Here's the answer:

Are you still seeing this with the current beta?

 

[thzero]

Yeah, such problems are usually due to memory thrashing, and with 8 designs open on a 4 GB machine you're asking for trouble. For whatever reason, memory does not seem to free up when closing designs, so I'm not surprised that didn't help.

Not to mention, running Java application is not really an OOB experience on Chromebooks.

 

[4regt4]

Not completely sure about the Linux implementation on the Chromebook, but it's not normally present. Requires a re-partition of the hard drive, and an installation of Debian 10. At least that's what the instructions said. Google themselves actually outline the procedure for the install. You have to go into "Advanced settings" and enable "Developer" mode to do the install. The partition and system download takes about 20 minutes. I'm by no means a Linux guru, though.

Hans.

 

[CalebJ]

Not sure if this should be a general OpenRocket question, but I'll ask here since I'm using the beta release...

When modeling a multi stage rocket, where can you see the ground hit velocity for booster stages? I can infer it by running a plot from the simulation tab and looking at the graph for a particular stage, but is there a better place that doesn't require those steps? Thanks!