Fix it When You Can
Before a spacecraft is sent to space it is important to test the spacecraft in a space-like environment because it is way easier (read: actually possible) to fix it before launch than once it is in orbit. One of the greatest advantages of high altitude ballooning is that it is so affordable and easy to do that it can actually serve as the pre-launch test for small spacecraft, sensor packages, or other payloads. Whether or not the balloon launch is your mission or just another testing step on the way to space, you can and should still do some basic testing before your launch to make sure you are making the most of all of your big efforts. Some early testing could mean that you never suffer a scrubbed or failed balloon flight because of a faulty calculation, loose wire, or over-simplification of a complex problem.
The principle behind testing is to expose your system to as many elements of the mission environment as possible before the full mission actually happens. For example, car designers always design their cars to operate in all the conditions it could reasonably (even unreasonably) encounter but it is extremely important to make sure the calculation were done right before the cars are produced en masse. By exposing the system in a controlled way to elements of the environment it will operate in, you can find if the system behaves as you expect and make improvements before the “big game.” This begs the question then, what do we need to test? Let’s look at the main aspects of the mission and see what we can test and what risks we must accept.
What to Test
What is there to test? Your payload will go to a near vacuum but the difficulty in finding a vacuum chamber to use is probably greater than that of just doing your launch so this is effectively off the table. Some other thing you can test are exposing your craft to cold and running it for a period of time similar to what you expect your mission to be.
Simulating the Near Space Environment
Let’s start with the doozy. Your HAB will experience a vanishingly small air pressure at the maximum altitude. Can vacuum be achieved on the surface of the Earth? Absolutely yes. Is it easy? Not really.
Spacecraft and their sensors can be tested in vacuum in specialized testing facilities but it would be reasonable to expect access to these chambers to be limited, or at least expensive. You could have better luck asking around at a university to test in a chamber they might have, especially any school with an engineering department. However, the difficulty in securing a vacuum chamber to test would likely approach the amount of effort needed to just do a launch in the first place so I would tend to skip this testing step unless my payload there was a particular risk with going to a vacuum, like damage to an expensive and/or sensitive payload where flawless operation in a vacuum is critical. Even then, it may be worth considering ways to mitigate this risk over multiple missions or explorations.
Before I test my new bicycle brakes on a massive downhill section I will test them out on my driveway. Parachutes should be very reliable, especially ones that you don’t build yourself, but should still test your payload train or a simulated equivalent to make sure that your setup is not especially prone to collapsing the parachute or any other possible in-flight issues.
Doing calculation to ensure sufficient battery power is a great start, and it can be aided by looking up the battery specifications or datasheet, but your batteries are going to be exposed to a lot of cold and maybe even a lot of heat depending on what you are doing. This is going to be stressful for your batteries and you want to make sure that you have some confidence that what you have calculated for your mission. Measure the voltage before and after your flight to see how much of the battery you have used up but consider that the voltage drop is not necessarily linearly related to how much of the battery has been used up; look up the voltage/discharge chart for your battery to try drawing out some conclusions about usage. You might also consider finding a reliable way to draw down used batteries with a current counter to get a better feel for how much of the battery you are using. Even still, you might run your mission off of, say, half the planned number of batteries, measure their endurance, and then extrapolate out.
The greatest driver of the design and operation of your system will be the cold experienced in the upper atmosphere. For this reason, it is very important to know that your system will continue to function, both the electronics and the structure/mechanisms. An easy place to start is with the freezer in your kitchen or garage because even though this is not quite as cold as where you will go in the atmosphere it has no additional cost past the electricity you were already going to use. Once everything passes in your freezer then you can move on to testing in a cooler filled with dry ice. The dry ice is actually colder than the air where your craft will go so it is a good test being a little more extreme than the expected. Dry ice can sometime be a little tricky to find depending on where you live but at grocery stores that carry it it can be as affordable as about $1/pound. For testing one small payload you shouldn’t need any more than about five to ten pounds of dry ice placed in a cooler with your craft. Once the dry ice and payload are in the same place then you should test it for at least about 90 minutes, the length of a typical mission, but going longer, say for two hours, would be even better so you can test both extreme temperature and operation for a longer-than-planned mission.
This test as describe will also answer questions like if your camera will work in the cold, if your batteries are big enough and cold-tolerant enough, and if your memory cards are big enough.
This is an interesting testing area with two main approaches. I think it goes without saying that you will want to make sure that pieces of your payload will not readily move, fall off, or break with a few moments of vigorous shaking, like when the balloon pops and the payload train does some tumbling as it starts its high-speed U-turn back to Earth. With that situation out of the way, the decision you need to make is whether to take on a more intense testing campaign in this area. If you have a lot of time to test and your payload is more delicate then it would probably make sense for you to do a lot of impact testing, like dropping it off the roof or rolling it down a lot of stairs. This can quickly highlight where you have over- or under-built the enclosure or internal attachments. It will also almost necessarily require a re-build of at least parts of your system. On the other hand, if you have a simple payload that only needs to survive in mostly one piece until touchdown then you might as well just preserve your original and fly that. For context, a semester-long aerospace engineering class would probably do destructive (well, at least “damaging”) testing of their payloads while a week-long summer camp does not have the time or need to damage the first vehicle they build just because it might get damaged on landing; as long as the SD card survives (incredibly likely) then you are at least going to get your data back.
It is very interesting to know what is happening outside of the payload enclosure, but it is maybe even more important to know what is going on inside in case something goes wrong. In most cases it could be interesting to at least know what temperature the batteries are at (attach a temperature sensor directly onto the battery and the battery voltage (use a voltage divider circuit with your logging device). If possible, place temperature sensors at other interesting locations, like on the heater, logging device, camera, antenna/GPS, and any other critical parts. If anything happens during your testing or during the mission the data from these sensors could help you pinpoint what went wrong. Heater placement can be a guessing game so having a collection of temperature readings could help direct future heater placement.
If you are using an antenna that may produce enough heat for you to not need heaters on their own. Also, if you are simply running an Arduino you may find that the system is tolerant to cold and the current draw is not high enough to stress the batteries past the ability during flight. To include or exclude heaters, or where to place them and how many are not simple decisions or processes so I encourage you to pursue a rigorous testing campaign that will bring clarity to this area.
If you include a radio in your craft you should test it at a distance similar to the distance (both from altitude and downrange travel) as you expect during your mission. You can run into an issue with the curvature of the Earth in this case but if you utilize at least one tall mountain nearby then you should be able to get this range of at least 20 miles.
Even if you are using an off-the-shelf solution (Delorme inReach, SPOT, etc.) you should do what you can to emulate its operating environment. Will it transmit through a wall of your payload? Could it land upside down? Could it turn off when it gets cold or hits the ground and will it turn back on?
It is important to practice the fill procedure and every other part of the launch day because one thing left at home or one mistake while filling can delay you by hours if it doesn’t scrub the entire launch and/or cost you in terms of balloon or lifting gas replacement.