Achieving Stable Aerodynamic Flight (6)

(I try to link to the products I used so you can find them more easily. If you purchase them from these links I may receive compensation from affiliate programs. I am not employed or influenced by the manufacturers or distributors.)

Since I started working on this project I’ve had a lot of fun learning about aerodynamics. It’s something that only really matters once you put the parts together, but planning for an aerodynamically rocket is critical for success. There are a TON of great resources to help you learn about the science of aerodynamics. I’m going to include the videos that I found the most informative in here to help me illustrate the concepts that we will need to understand. After we cover the basics, I will show you how we can apply the concepts to our own designs using a rocket simulator as we plan to build the exact vehicle we need.

Basics of Aerodynamic Stability

Before we get started, we should establish the axes of a rocket, so we have a common understanding. The Z axis runs from the tip of the rocket to the tail, and then the X and Y axes run from side to side, sometimes in line with the fins (on a 4-fin model).

The three most important concepts in aerodynamics are the Center of Gravity, Center of Thrust, and Center of Pressure.

Center of Gravity is the point where the weight is evenly distributed on any axis you choose. This will change slightly over time as fuel is burned, which is something to consider in larger rockets with massive fuel tanks, but in smaller model rockets losing mass in the tail will only work to your benefit. On the Y and Z axes you want the CoG to be as close to the origin as possible. Moving the CoG along the Z axis will be how we achieve stability.

Center of Thrust is where the force of the motor(s) or thruster(s) is applied. You can change this using Thrust Vectoring or RCS thrusters to change the trajectory of the rocket. Typically your center of thrust will be at the rear, and optimally the direction of thrust is perfectly aligned with the X axis.

Center of Pressure is the center of the force applied on the body by the air flowing over it. This can be very dynamic, and there are a lot of ways to determine the CoP, some of which will be covered in this video.

Essentially the goal is to align the Center of Thrust with the X axis perfectly, then arrange the rocket body and payload in such a way that the CoG and CoP are at the origin of the Y and Z axes, and the CoG is forward of the CoP. This way, as the rocket flies the pressure forces acting on it will keep it pointing straight. This is usually achieved by adding fins toward the base of the rocket, which move the CoP backwards by giving added surface area to the body.

Unlike with engines (😂), bigger fins aren’t always better. There is a phenomena called Wind Cocking, which is caused when wind blowing along the Y or Z axes exerts more force on the tail of the rocket than the nose, tipping the body, and causing the engine to push the rocket into the wind. To avoid this you want to try to keep your fin size and shape just large enough to provide stability, but not much larger. Here’s a great video about estimating CoG and CoP, and properly balancing your rocket body so it is aerodynamically stable.

Rocket Stability – LabRat Scientific

Software Modeling for Aerodynamic Stability

Now that we understand how aerodynamic flight works, let’s talk about simulating the design. As I mentioned before, I have been using OpenRocket to do all of my virtual modeling so far. OpenRocket is a Java app, so you can run it on any computer, and it has some pretty simple components you can add and customize to suit your design goals.

For the first version of my rocket I had already done some shopping for parts, and settled on a clear payload bay for the 360º camera, a fiberglass engine compartment, and a pre-fabricated ogive nose cone. The fins that I used were made of balsa as well. With most of these components pre-determined, the two biggest variables would be payload placement/weight and fin size/shape.

As we discussed earlier the payload and engine weight and placement will both dramatically change the center of gravity. For the rocket’s maiden voyage I didn’t want to send up the camera, which only left the telemetry package, a mere total of about 60-80 grams, in the front and a large heavy motor in the back. This meant that the fins would have to be somewhat larger, to pull the center of pressure further back.

In the image below you can see what the rocket’s properties would be like with the motor I anticipated using and just the telemetry in the payload bay. I trimmed the tips of the fins to reduce the stability so the rocket would fly better if we encountered wind. Realistically, though, this was still a lot of fin in the face of any wind.

Version 1.0 digital model with an empty payload bay

When I add the 360º camera then the stability goes way up. This would result in a lot of wind-cocking, since the inertia of the heavy payload bay would exaggerate the tipping. This was about as far as I could trim the fins and still have a stable empty flight, though, so it would have to work.

Version 1.0 digital model with camera onboard

One of my favorite features of RocketSim is the ability to plot the model of the flight. There are a few considerations to remember here. The rocket takes off under power, and flies under power for a total of 3.5 seconds. After the motor burns out the rocket is in ballistic flight for about 6 seconds (the current model uses a 6-second delay) after which a charge detonates separating the engine compartment from the payload bay and ejecting the parachute. From this point the vehicle should gently descend back to the earth at a steady rate of ~-10m/s.

Plot of simulation altitude, vertical velocity, and vertical acceleration vs time

You can even adjust parameters like recovery chute delay and launch rod angle to accommodate for windy conditions and predict how far the vehicle will travel from the launchpad.

Plot of simulation altitude vs position

This last plot I found really informative, which shows stability over time. Note the first second or so, where stability is most critical. You don’t want the rocket to be unstable when it leaves the launch rod, because it may tip, and without enough wind passing over the fins it won’t correct itself, and may fly off uncontrollably.

Plot of simulation stability vs time

Finally, I’ll leave you with a couple of videos that I really enjoyed. The first one is very informative about rocketry on many levels, and the other illustrates how active aerodynamic controls works in the context of the F-22 fighter jet. In my next post, I will discuss parts, and the strategy I used that allowed me to assemble the rocket in just a couple of days.

Imaginary Rockets – Modeling, Mocking and Planning (2)

(I try to link to the products I used so you can find them more easily. If you purchase them from these links I may receive compensation from affiliate programs. I am not employed or influenced by the manufacturers or distributors.)

Most people who know me probably wouldn’t be surprised if I skipped the planning phase and just started building a rocket, blindly trusting it would work. Kerbal inspired me, though, and modeling and testing virtual prototypes had its own appeal. I downloaded OpenRocket, and found that on this modeling software, much like in KSP, I could build outrageous designs and test them for feasibility without ever having to worry about the dangers of live testing.

Before I even got started building I figured it would be important to set a few goals of what I would like to accomplish through iterative testing and development:

  • Stable Aerodynamic Flight
  • 1km Apogee
  • 360º Video
  • Aerodynamic control (articulated control surfaces? thrust vectoring?)
  • Remote Telemetry
  • Powered Descent/Landing
  • Maintain Medium Powered Rocket Classification

I’ll go into more detail about each of these objectives in their own posts, but to summarize, I wanted to build a rocket under 1kg that could take a 360º camera up to 1km altitude. Meanwhile, I thought it would be fun if that rocket constantly sent telemetry data back to a computer, and was able to perform some tricks.

At first I thought it might be necessary to build a massive body rocket that can carry larger camera systems, since most 360º cameras are fairly wide, but with some searching I was able to find the Kandao Qoocam 4K – a 360º camera in a stick configuration, which will help greatly reduce the diameter of the rocket, and therefore improve A LOT of the aerodynamics, but I’ll discuss this more in my “Optics” post.

Kandao Qoocam 4K 360º Camera on

A 1km rocket isn’t easy, especially when one key limitation is that I’m not certified to fly anything over a G class motor. We can get further into motor classes and model rocketry certification levels, but suffice to say I was not going to just be able to strap more motors on it like you would in KSP. This time I needed to be a lot more strategic about my design choices.

I spent some time looking around for a suitable rocket modeling program, and settled on OpenRocket because it was straightforward, free, and cross-platform (which meant it would work on my Mac). There are a lot of great modeling programs out there. They range from simplistic, like OpenRocket, to complex all-purpose 3d modeling suites. Choose one based on your level of comfort or ambition, and employ google search liberally.

In OpenRocket you can change each of the design parameters of your craft, including weight, roughness, shape, placement, motors and timing. One major setback is there seems to be a bug when running on OSX that causes it to crash when switching back to “Side View” from other view modes.

Another big setback of OpenRocket, but also almost all model rocketry software I encountered, was that it isn’t built for modeling electronically influenced flight plans, such as off-engine recovery methods, chute delays, dynamic flight control or electronically timed staging. It’s not the end of the world, and I’m probably one of very few people who have this complaint.

The first versions of the MarkDart were very bulky. Before I found the Qoocam, large cameras led to large bodies, lots of engines, and ultimately nowhere near Medium Power classification.

OpenRocket rendering of an early, unfeasible version of the MarkDart

As you can see from the image above, it would have been a BEAST of a rocket, both in size and trajectory. While I would have loved to violate FAA airspace regulations, I’m not entirely confident I’m ready to be able to light 3 motors simultaneously, and the risk of rapid unscheduled disassembly would have been altogether way too high.

Another early/scrapped MarkDart design

Moving to the Qoocam allowed me to be a lot more realistic about the design, ultimately settling for a one-engine vehicle with a 41.6mm diameter body. The rocket would have a clear payload tube for the optics payload (that’s aeronautical engineering lingo for the camera), and a fiberglass engine/parachute compartment.

The final MarkDart1 design before ordering parts

In my next post, I will discuss building a system to understand what the rocket is doing at any given point in time, also known as telemetry.