Hydrostatic Testing
Hydrostatic testing is when we pressurize our motor case with water to test the seal of our hardware at high psi. Hydrostatic testing allows us to determine the pressure thresholds of our design and give us a chance to calibrate any sensors or Data Acquisition Systems (DAQ) before static fires.
Hydrostatic Setup:
- Check motor
- make sure no phenolic or graphite parts are in motor
- check installation of hydrostatic cap
- Pump
- Fill 2 buckets with water, one for pumping and one for exit
connect reservoir hose to low pressure of pump, unclip hose
open red drain valve on high pressure side, close pressure valve
pump a few times until no bubbles appear in drain bucket
Pressurizing our motor
attach pressure hose to hydrostatic cap (use Teflon tape!)
open pressure valve on high pressure side till water comes out of the pressure transducer hole
attach pressure transducer
attach pressure transducer to DAQ
Hydrostatic Procedure:
Setup cameras, close blast chamber as much as poss.
Start up data collection
pressurize
start with a little bit then build up psi by 250 psi increments
call out pressure milestones, venting, etc.
check to make sure pressure is holding (if not, indicates a leak)
pause 1 min increments to check for leaks
Post Hydrostatic Test:
- open red vent valve to depressurize
- turn off DAQ
- drain water
- leave the place cleaner than you found it
More on testing procedure, emergency protocol, checklist can be found here
...
Static Fire
Why static fire?
Prove that the motor works as an isolated system
"Fail fast" and without blowing up other subteams' work
Gather data to validate your own simulations
Gain information that will help you iterate on your design in the future
Characterization of propellant (*C can be calculated with pressure and burn rate)
Safety materials, fire extinguishers, etc.
test stands, motor, ground stakes, ratchet straps, DAQ, igniters
Hardware like wrenches, screwdrivers, calipers, cameras, grease, shovel
charged computers, proper software
Snacks
Now, we will talk about the static firing procedures. This is a list of actions that must be followed in the following order:
...
Firstly, a higher frame rate is often more important than a larger resolution, but both are good to have. And secondly, trying to get as many different angles as possible is very important.
After shooting various quality videos and taking pre- and post-firing pictures of the equipment, it is important to know how to diagnose common failures. These consist of overpressurization and thermal failure.
Overpressurization occurs when the propellant produces too high of a chamber pressure, rupturing the motor case. Usually, this happens at the startup unless the design is progressive, or a large void is present.
Thermal failure occurs when the insulation protecting the case from combustion fails and the case melts down. Typically this is seen later in the burn when the temperatures have really been high for a long time. Sometimes this failure seems relatively gentle, and other times it looks just like an overpressurization.
Now, we will briefly talk about data acquisition. This system should be able to sample transducer as quickly as possible, at least 100 hz. Since transducers typically output a very small signal, it has to be amplified and a conversion function from voltage to force or pressure must be found. Also, the transducers should be pre-calibrated over the full range that they will see during the firing.
Our second to last topic is propellant characterization. As we know, burn rate and pressure are proportional, so by doing a series of burns with different throat sizes and fitting a curve to the data, it is possible to experimentally calculate the burn rate as a function of the pressure. This is also a good way to get c*. A minor but important point is that throat diameter before and after the burn should be measured if any slag accumulates.
Lastly, there are some general data processing tasks to do. Comparing maximum and average pressure to the simulated values (so that we can check if the characterization needs to be updated), comparing real thrust trace to the simulated values (so that we can check if they match, and think about if effects like erosive burning can explain discrepancies), calculate ISP and c* (so that we can check whether they are in line with what we expect from the propellant formula), and pass the thrust curve to the simulations team.