Excel template for crunching Perfectflite data

[Penian]

I put together a template for crunching your downloaded Perfectflite data in Excel. This allows you to put multiple flights on one altitude/time chart. You can also chart velocity in MPH, FPS, or Mach number. The Mach number is altitude corrected.

I've attached a screen shot of a MPH chart with two flights.

The template is located in a zipped file on my web site at https://www.rocketry.penian.com/ , in the technical stuff section. There are instructions included in the workbook.

Let me know how it works for you, suggestions etc.

 

[solrules]

The spreadsheet looks really nice...you have some pretty good Excel skills. Here comes the hurt: you can't do velocity differentiation from altimeter data without huge amounts of error. Plain and simple. I've tried it, others have tried it and came to the same conclusion. When you differentiate the altitude to get velocity, the error between each data point sticks out....unlike integrating accereation, which "smooths out" the error between each data point.

Do you see the "up and down" nature of the Perfectflite: Altutude goes up, stays there for a few samples, goes up again, stays there for a few more samples, repeat. This is destruction for differention subroutines. You get a positive velocity, then 0 velocity, then positive velocity again. Sure, you could average the data points, but that takes out all of the large changes in position, making your velocity innacurate.

To prove my theory that baro altimeters suck for velocity calculations, I flew two perfectflite MAWD's and an ARTS accerometer in the same rocket. When I plugged it into you spreadsheet, I get about 680 MPH for max velocity, which is a little under .9 mach. Compare that to the ARTS data, which shows that it hit 593 MPH. Big difference. Add to the fact that the ARTS samples at 200Hz, but the MAWD samples at 20Hz, the ARTS will give you good data, while the MAWD will give you total junk, no matter what you do with it. I get an average of 20% error on every flight, making it useless for "mach busting" calculations.

Nice try, but so far as I can tell, it can't be done. Spend an extra $100 and buy an ARTS, there is much more data to mess around with.....it's a bundle of fun.

 

[Penian]

Yes, speed calculations between individual data points is useless, the S/N ratio is terrible. The more points you average, the better the accuracy, and less sensitivity to bogus velocity transients.

The handy feature in the Excel tendline function is that you can adjust the number of data points in the rolling average easily. Try ncreasing the number of data points averaged (and reducing the peak speed) and and you can calibrate the Perfectflite velocity against the ARTS velocity.

The ARTS is a nice machine, but you will have touble getting it into a 24mm tube.

 

[aerospike]

I did the same with my PF data for my kid's science fair project (terminal velocity) We averaged and plotted a trendline. We got numbers for max altitude and velocity at the end of the flight that matched very well with sims and hand calcs. for terminal velocity of his rocket (flown w/o a nosecone and allowed to freefall to 300' before popping the main) True it is very noisy but I wouldn't characterise it as crap. Besides, integrating accel, and computing velocity and altitutude ain't all it's cracked up to be either. (200 samps/sec or not)