Verification and Validation Flashcards
What are some common tests for sensors?
Sine sweep to determine bandwidth with frequency domain analysis (use FFT on data), impulse response to determine rise time, settling time, overshoot.
What does phase shift indicate in a frequency response plot for a sensor?
Phase shift indicates time delay between input and output signals at a particular frequency
What are decibels?
Unity of measurement to express the ratio of two signals on a logarithmic scale. A decibel is 10log10(P / P0) where P is the power of interest and P0 is a reference power. 3db is 2x, 10db is 10x
How do you go about creating a test plan for verifying a feature?
I like to break my test plan up into unit and integration tests, and i like to then break each of those up into phases, the first phase being safety (both human and hardware) related tests (like isolation and continuity, or fault handling) for instance and then into functionality. This way I can be sure that I keep people and hardware safe. I also am sure to include relevant subject matter experts as needed.
Validation vs Verification
Validation determines if a spec meets customer’s requirements, whereas verification is the process of ensuring a feature meets the spec
Functional vs Nonfunctional requirements
Functional define what the product does and non functional describe how the products works. Features vs properties
Example of troubleshooting a thermal issue
I wrote nearly all the software responsible for controlling the flight MCBs for Spider, as well as the test scripting for flight MCB actuator qualification. So I would frequently be contacted by test engineers to help them debug issues with the software or the MCBs. At one point I was contacted because a test engineer kept getting an overtemp fault reported by one of the MCBs. I had architected interface software such that it would report all hardware faults up to the RCC UI which I talked about earlier, so I knew from the fault-code that the overtemp was occurring on the motor brake.
I also knew that this brake was electromagnet, and that a common reason for high temperatures on these types of brakes is overvoltage. I took a look at the brake voltage telemetry from the MCB in the test logs on TADA and noted that, while the brake pull out voltage was correct, it was never stepped down. This is atypical as electromagnetic brakes in these flight motors typically have an inital high pull-out voltage to enable the motor, but the holding voltage. I advised the test engineer to reconfigure the MCB to drop the voltage after initial pull-in to match the brake holding voltage spec and we no longer had overtemp issues!
