SpinorSat — A step along a 50 year journey to the stars

Travis Brashears
7 min readOct 15, 2018

--

We’re a team of 6 Berkeley students aiming to push the frontier of miniature spacecraft that can be mass produced for large networks of satellites. We created a PCBSat, a 35 x 35 x 4 mm satellite that is the size of an Apple Watch screen and is currently less than 10 grams. It has an advanced power management system, radio, and maneuverability system;we’re terming it a SpinorSat. You can learn more about it here.

Pictured are 4 (out of 7) of the boards we are flying

We are flying with KickSat and launching 7 of our own SpinorSats aboard the CubeSat Deployer Created by Zac Manchester!

KickSat Deployer — Designed by Zac Manchester

The idea dates back a couple of years. In early 2017, a student group called Space Technologies at Cal was created by Travis Brashears and Olivia Hsu to create a student space program and push forward innovative space research. Each member dove into their own vertical, such as autonomous vehicles, experimental payloads, and satellite deployment. In the summer of 2018, we started working on the idea of the PCBSat and had a unique opportunity to fly with KickSat to further develop it outside of the student organization. The six of us, all Berkeley seniors collectively majoring in Electrical Engineering & Computer Science, Mechanical Engineering, and Engineering Physics, are now building out our own board design and technology.

Team:

Travis Brashears: Engineering Physics - Technology Lead

Aviral Pandey: EECS - EE Lead

Olivia Hsu: EECS - Board Design and Firmware

Kevin Zheng: EECS - Board Design, Software, Radio

Daniel Shen: MechE - Software and Mechanical

Malhar Patel: EECS - External and Software

Olivia Hsu, Aviral Pandey, Travis Brashears, Kevin Zheng (Left to RIght)

More About our Satellite

Our initial layout design was based off of a Sprite designed by Zac Manchester. We iterated off of his layout in order to reach our current design. Some of the changes include:

PCB Layout
  • We selected a single MSP to be the onboard compute module and then developed our own MPPT algorithm to charge/discharge our capacitors with PVs.
  • We integrated all other electronics, including a 9 axis IMU. However the accelerometer does not work when orbiting the earth due to g being effectively zero. So we only use the gyroscope and magnetometer data.
  • We developed a torquing mechanism to counter the rotation of tumbling and atmospheric drag by using torque coils. We will effectively burst energy through a coil of copper wires via stored energy in the capacitors that will generate a magnetic field which then counters the Earth’s magnetic field and rotates our PCB about its principal axis.

The torque control system forms one of the foundations of our current iteration so we decided to dive a bit more into its functionality. Since our board is effectively a 2-dimensional PCB, but we utilize two external coils embedded on the edges and a third spiral coil within the board. In order to torque, we use a variant of a B-dot controller in order to select the correct axis and power level to torque. This essentially is taking a cross product of the magnetometer data with each unit vector (i.e. torque coil axis) and then applying a dot product to that output and the gyro data to determine a similarity value across all axis’. We then check which axis will apply the largest opposite angular acceleration.

Visualization of magnetorquer selection

For this initial launch, we would be content to just measure the change in rotation of the PCB and communicate down to Earth that we were able to modify the tumbling. Since this torque is on order of 10^-6 Nm, it would be a remarkable accomplishment to measure and use this method of actuation to completely detumble.

Larger Satellite vs Smaller Satellite

Obviously, we cannot utilize the sheer space and mass that large satellites have and are able to do with larger optics, electronics, and antennas, but we push what is possible from cellular technology to allow for connectivity between satellites, which is our primary goal. There are clear limitations of using small electronics, akin to smartphones versus laptops, but there are certain use cases that larger satellites are overkill for. These use cases include satellite network connectivity around specific orbits, creating a swarm of smaller nodes that connect larger satellites, and long distance space travel probes. This is where we apply SpinorSat technology and capitalize on the advantage of low mass satellites.

What’s Up With Space

There has been a cumulative $15.3B poured into space verticals worldwide across 351 companies (Space Angels Q2 investment Report). These companies are working on topics including satellites, launches, and new planetary markets. This monetary flow paired with an increased political focus on advancing space exploration and development through programs like the Space Force place space tech at the forefront of new global initiatives.

Most of these initiatives have centered around reducing the cost of launch and constructing satellites that abide by standard form factors. In the past, there were only a few launch providers so these organizations determined the form factor of satellites, which forced satellites to remain large and bulky, but now the pendulum has shifted towards the satellite makers to determine the form factor of these satellites. This opens up the gateway towards modularization and optimal designs.

Modularization through small electronics have become more pervasive in the past few decades with most technologies fitting on the palms of our hands, such as the first cell phones and now wearables. Larger satellites often use older technology, which expose them to numerous vulnerabilities and constant decay. Additionally, given that their functionality and usability can be mapped onto smaller satellites, it does not make sense to continue to deploy larger satellites with their associated high launch costs.

  • From a launch cost perspective, the cost to launch is significantly decreasing, which makes launching a large number of smaller satellites more viable in comparison to a single CubeSat.
Source: Founders Fund
  • From a futurist perspective, the evolution of long distance space travel is nonsensical for a large mass system. Spacecrafts must be evolved to be on a wafer and a SpinorSat is the first step in that direction. The future of space travel will not be with rocket fuel; it will be a standoff system that delivers propulsion via beamed energy. Under a system that operates under beamed energy propulsion, the speed scales to the 4th power of the spacecraft mass. (Building WaferSat Paper, Evolution of WaferSat Paper) Thus, decreasing spacecraft mass is super advantageous to maximize speed.
http://www.deepspace.ucsb.edu/projects/starlight (Q. Zhang)
  • From a network perspective, architectures tend to have a central base station and smaller nodes that function as the channels for that station. SpinorSats are able to operate as those smaller channels that can be abundantly spread out.
  • Creating a sparse radar array with a large distributed network of PCBSats that communicate to a main CubeSat. This would allow for an effective larger radar that comes at the fraction of the cost.

Next Steps

We already have a small satellite with an onboard power management system and built-in maneuverability mechanisms. Our next steps are to build a custom control scheme and an advanced communication system.

Cost Structure Advantage:

Let’s imagine a world where we were able to align our SpinorSat with another and establish a communication link between multiple SpinorSats. The depreciation of each satellite is quite advantageous. Falling under the assumption that we don’t need a SpinorSat deployer here is the hypothetical cost for a future model that includes advanced comms and alignment tech.

Assumptions:

  • We are able to launch N PCBs without a CubeSat
  • Largest mass of a PCB at 100g (upper-end currently it is about 10g)
  • Launch cost is $3000/1 Kg
  • Mass manufacturing lowering the base cost

For a network of 10,000 SpinorSats the cost is amortized to just about $3K per satellite (which includes launch cost.) I will leave this as an open invitation to spark creativity from the reader on what one could imagine possible with such a cost advantageous spacecraft.

Huge Shout out to Zac Manchester, creator of Kicksat, for making this happen! This wouldn't have been possible without him. Also thanks to everyone in Berkeley’s Supernode for the help along the way and enduring our long hours! And much love to everyone within STAC that helped out!

If you have any thoughts, suggestions or comments about what we’re working on, please feel free to contact us at trbrashears@berkeley.edu and malhar@berkeley.edu. Additionally, if you would like to get involved as a contributor or a sponsor, feel free to contact us as well!

Travis Brashears and Olivia Hsu explain what is innovative about our PCBSat

--

--