Help Developing New Android Software - Drag Coefficients

[Laura Garland]

Hi,

I'm new to the rocket forum but not so much to rockets as we have been shooting with our three boys for the past 10 years of so. Yes, I have my own rockets including scratch built ones. I don't just watch. My favorite is a G powered scratch built we call Krazy Kitty. It has bling all over it and cat heads.

Anyway, my husband and I are developing a new rocket software package for Android that is near completion. He did a short release a few years ago but decided it wasn't ready and pulled it down. He has had it on Amazon for the past couple of years to get input (much slower than play store). Now we have pulled it down to make last tweaks and intend to release on the Playstore for Android before the end of the year.

I already does a lot but we would like to do one last thing before we release it.

We have a box on the rocket screen that let's you set the drag coefficient. It defaults to .75 as everyone including NASA seems to indicate that is a good place to start. You can change the drag coefficient on each rocket and it changes the graphed apogee altitude.

We already have about 1000 manufacturers rockets to select from or you can add your own. We will default the drag coefficient to .75 but would like to include actual coefficients for any of the models where we can find em.

He is the EE/programmer and I am the AE/graphic artist and this is my job to find as I am the math minor on the team and I handle most of the formulas.

I have looked and can't find much.

NASA says, "Coefficient Cd contains all the complex dependencies and is usually determined experimentally."

We hoped coefficients were known for some of the more common rocket kits available over the years. He has been flying alphas and other common models for over 40 years so surely somebody has that info available.

If we can't find this we will release with the default of .75 for all.

Any help is appreciated.

Laura

 

[samb]

Welcome to the forum and congrats on a very interesting project. Actually I'd be very surprised if a database of rocket kit drag coefficients exists. My understanding is that variables like finish, alignment, and overall workmanship have a dramatic effect on any particular design. One of the foundational documents, TR-11 by Dr. Gerald Gregorek, did contain a pretty in-depth analysis of the Estes Alpha. Hope this helps.

https://www.oldrocketplans.com/pubs/Estes/estTR-11/TR-11.pdf

 

[Laura Garland]

I haven't seen that one but it is similar to other stuff I find where there is a lot of variation in build. That is why we want to avoid trying to provide a Cd calculator in the app but at least try to provide a more realistic starting point.

My husband is going to write some code to try to strip the contents out of ork and rkt files just to see if he can build a list of Cds for a lot of the common rockets. If so, we can throw a bunch of those in a directory, scan the Cds out of each and update our 1000 rocket database. We'll see if that works.

Thanks for sending that.

 

[Alan15578]

I haven't seen that one but it is similar to other stuff I find where there is a lot of variation in build. That is why we want to avoid trying to provide a Cd calculator in the app but at least try to provide a more realistic starting point.

My husband is going to write some code to try to strip the contents out of ork and rkt files just to see if he can build a list of Cds for a lot of the common rockets. If so, we can throw a bunch of those in a directory, scan the Cds out of each and update our 1000 rocket database. We'll see if that works.

Thanks for sending that.

That would be outstanding if you could provide a database of Cds for common model rockets. I don't think the Cd is contained in .ork, or .rkt files, rather it is estimated in simulation programs from all of the data in said files. You could of course run all those designs files through their respective simulators and then use their estimated Cd. The ideal procedure would be to build each model and measure its Cd in a wind tunnel. As a practical matter you could just use 0.75, but allow the user to enter their own estimate. It would also be good to show results for say +/- 20% of the Cd so that the user knows the sensitivity of performance to estimated Cd.

 

[Laura Garland]

That was/is our intention. You are very much correct that Cd is not in those files. Hubby's cussing confirmed it. Now he is trying to automate those simulations so we can get a lot of Cds really quick. He will spend time to write code before he spends time running simulations.

If we can get this, we will certainly make that list available.

 

[Buckeye]

CD is not one single number. It is a function of velocity and changes noticably around Mach 1 and with flight profile. That is why is it calculated at simulation time by OR and RS.

Drag coefficients of similar rockets are very similar, or not, depending on the build and finish. Trying to catalog 1000 rockets is probably not worth the effort nor very accurate. If this is a simple simulation app, then an average value of 0.75 is probably good enough.

 

[beantownJPL]

I'd say for most rocketeers, Cd is a hand-wavy number that most won't try and figure out for their particular model. Very, very few of us would have the equipment to determine a more reliable number than the 0.75 value.

Simulators are great for getting ballpark numbers for a model's performance. If a modeler has done the homework to actually get the Cd of his particular build ... he probably doesn't need to sim the rocket's flight profile. More than likely that's dialed in at this point.

What would be interesting from a sim is to get a range of values around that 0.75 default. I don't know what's reasonable, maybe +/- 20%? The ability to enter the starting value of Cd has merit though, as it'd allow a user to vary the simulation based upon things that are known only to them. Also useful and accessible to most of us would be to have defaults for and ability to override things like overall mass, CG, CP, etc...

I also assume that 0.75 value is only applicable to 3/4 FNC models. Will you be able to model any oddrocs, tube fins, strap-on boosters, etc...?

It's a laudable to compile a database of Cd's, but I think you'll find a dearth of information about it since the equipment/technology to actually measure it is out of reach to most folks in the hobby. And those who do have access to it also probably have access to much better software!

Another interesting thought would be to crowdsource data. Add a flight log feature, and folks who have the equipment to measure their flight performance could log their flight data on each model they own. Over time you can compile that database yourself and estimate the range of Cd for any given model from the collected flight performance data. Of course, your performance data will trend toward the higher end, as folks who are investing in data collection devices are also probably flying more well-built models, but you could apply some statistical techniques to round out the left side of the bell curve.