About the Internship
The Peregrine project is research and testing for a new type of hybrid rocket that uses paraffin wax instead of conventional solid fuels. The end goal of this is to use this type of rocket to return samples of mars rocks, collected by a new rover, to earth in the 2030’s. This rocket has only been hot-fired a handful of times, but very little data has been collected during these tests. The purpose of my internship was to build a system for data collection for the hot fire test, and two subsequent, suborbital test launches.
Goal and Objectives
To design, build, program and test a data acquisition system (DAQ) for the Peregrine sounding rocket that is capable of recording and streaming data, from a couple of sensors, to an off-the-shelf radio transmitter. The system should also sample at or above the nyquist limit on all of the accelerometer and pressure channels, to allow full reconstruction in the frequency domain.
Methodology used
The data acquisition was split between two DAQs, the forward and interstage boards, to minimize the data rate for any individual board, allow a single, simpler system to be reused with slight variations in both situations, and to reduce the number of long, analog connections between the forward and aft of the rocket (the connection between the boards is digital, only the accelerometer inputs are run from the forward all the way to the back). The boards were designed by the other intern on the same project (Adam Rising) and I, in Altium, using Github to allow both of us to work on it simultaneously. The data acquisition itself was split into two sections, one handled by the single ADC and the other by the simultaneous ADC. The simultaneous ADC read all of the accelerometers (it samples all channels simultaneously, hence the naming), the single ADC read everything else through a multiplexer (thus, only ever reading a single channel at any time). For the most part, Adam worked on the simultaneous ADC components, schematic, etc. and I worked on the single ADC.
Results and Conclusions
The delivered result was a nearly completed DAQ system, an issue in getting the main controller (a board developed by NASA for data acquisition on Cubesats, called the Sphinx) meant that some portions of the project could not be tested. All of the single ADC inputs, on both the forward and interstage boards, are sampled and recorded by an ARM microcontroller on the DAQ board, and saved to an SD card, as well as streamed over serial, to be recorded by the Sphinx. The code for the microcontroller, each of the analog inputs to it (including the low pass filters), was tested as well, and all worked.
Lessons Learned
-Check that the rates given are actually sample rates, not the nyquist limit for a brick wall filter before redoing filter calculations for the twelfth time.
-Double check the datasheet. (Opamp input bias current was several uA for a couple of filters, and the filter resistors were in 10s of K, which ended up as a few volts off, and also meant I had to do the filter calculations AGAIN).
-Things happen, the same reason it’s hard to get whipped cream right now (a nitrous oxide plant exploded) caused a currently indefinite delay of the hot-fire test and launch of the rocket, since it uses nitrous oxide as the oxidizer.