![]() Quote:And back to "why not just use JBM" - I want something that I can run when I'm not on the 'Net. The SD of error for the McDrag program is 11% at transonic, which means +/- 22% error at this speed (95% confidence limits, see above). +/- 10% is very good for tranosnic drag predictions, most aeroprediction tools will not do that good unless they were trained with specific experimental data sets. Quote:Am I correct, interpreting what you're saying as "currently available programs - including McDrag - will not provide CD's & BC's in transonic region with accuracy better than in the ballpark of 10%"? If you ran the same program with this parameter set to different values, it would produce different outputs. Sometimes the programmer hard-codes this to what's best for the application, and sometimes it's left to the user to select. There is an input parameter which has to do with the laminar/turbulent boundary layer transition. I could be mistaken that Geoffrey's program uses McDrag, but I'm sure JBM uses it. Since his site has been down, I couldn't come back and double-check. Originally Posted By: mouse07410When I ran JBM McDrag on your Hybrid 300g bullet, the G7 BC's I got were more different from your experimental data than the BC's produced by Geoffrey's program on Border Barrels Web site. 95 where the SD is 11%, you can be 95% sure that the program is returning answers that are +/-22% accurate. So for supersonic speeds where the SD of error is 3%, you can be 95% sure the program is returning answers that are +/-6% accurate. To put the statements in perspective, +/- 2 standard deviations is the 95% confidence limit. Quote:In Figures 14 through 24, the present t.heory and ex Here's a copy/paste from the above pdf regarding accuracy (didn't realize you could copy/paste text possible, but a little sloppy) You can expect accuracy of +/-10%, better for super/subsonic, worse for transonic. pdf written by the wizard himself (McCoy) who developed the program:īear in mind that the accuracy of these prediction codes is 'better than nothing', but not nearly as good as directly measuring drag for a given bullet. The (Basic, I think) code for McDrag and McGyro can be found here:Īnd here is the. The tabular output from the JBM code makes it easy to import the data into excel to plot, etc. I find I use the JBM as much as my own code now, depending on if I have Matlab open or Chrome. I re-coded the McDrag code in Matlab, and not surprisingly, it returned the same outputs as JBM. ![]() Speaking of re-inventing the wheel, why not just use the implementations already available? The McDrag hosted by JBM is solid. Appreciate your offer to help, and will use it. ![]() Perhaps just add some GUI, or maybe CGI interface would suffice. I'll start with McCoy code for sure - for if it works, why re-invent the wheel. Been meaning to try to tackle this myself, I just haven't found the time. Would try to convert McCoy's code or start from scratch? Let me know if I can be of help. And then perhaps you could set up a Web site to run it. Which means - if & when I have time, I'll try to find or write the code that can do that. Spoke to them - they'd rather not share the code. Originally Posted By: mouse07410Originally Posted By: BearwalkOriginally Posted By: mouse07410 Re: Drawing your own Litz type bullet made easy
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |