Author: kdramos

Asterism Final Presentation

 

 

This week was mostly spent on getting the last few particulars of physical assembly done. Tomorrow (Sunday 26 2020) will mostly be spent on getting data and video compiled for the upcoming presentation. As of Saturday 4/25/2020 at 9:16 pm the progress on each individual physical part is as follows:

1. Small digikey bipolar motor driver/controller PCB (fully assembled, only needs to be plugged in [Rpi + motor + housed on physical mount]).
2. Nema 17 unipolar motor driver and power supply (1 set [2 H-Bridges + 1 Power supply + heatsink] fully assembled] 2nd set is being debugged, I might attempt using replacement MOSFETs/BUCK converters to see if that gives better results)
3. Wooden physical stand (each portion [1 ⅝’s wood door hinge assembly + polar scope + camera mounting slot] + [1 MDF declination axis mounting hinge] + [1 MDF turntable Right ascension mounting slot] individually assembled, once debugging is done it should be a matter of joining parts and wires)

On top of that a few visual aids have been set up for debugging and demonstration purposes:

1. Arduino adc logger + matplotlib for plotting logged adc input over time.
2. Simple led circuit for demonstrating test point connection to circuit ground for small digikey motor controller.
3. As discussed previously a 10W-10Ohm resistor for measuring H-bridge current (currently being used for debugging)

 

System Changes:

The main changes done this week basically break down to design changes to accommodate parts trickling in over time. Most of this is adapting the previous eq aligning design to the new barn door tracker. After receiving the universal stepper motor couplers I should be able to assemble the top portion of the diagram above with the exception of the motors since I do not have the pcb’s yet and I would like to thoroughly test them before placing them into the mount. Some things to point out in the diagram below include:

1. The polar scope is manually (by us, not the “user”) aligned parallel to the hinge joint adjacent to the scope.
2. The compensator motor is attached to the declination base and the universal hub is attached to the carriage bolt.
3. The motor is kept from spinning by two #4 threaded rods that are locked into the barn door tracker base by two #4 nuts each. As the cariage bolt actuates the motor is pulled along with it so spacing must be kept in mind and a hole of sufficient size must be bored in the declination base.
4. The base below the main barn door tracker compensator is essentially the same design as the one previously done by Joy for the previous iteration of this project. Some adaptations have been made to fit the new requirements. (As discussed above).

Progress has also been done in adapting logging for the Arduino. Getting the logger right now outputs to csv in real-time via a python script.

Risks:

Unfortunately, the arrival of the incorrect board types set back demo-readiness a fair bit, hopefully once the new boards arrived the new design will be ready for it. So that’s a risk that we could face.

This week I worked on the tangent error calculations as well as the parametric equations for modeling the 3d piece used for hardware error correction.

 

Our new method of barn door tracker has inherent errors due to the way the upper arm is actuated (as the hypotenuse of a right triangle). This means that while the expected rotation rate should be constant the actual rotation that comes from this method is proportional to (dL/dt)where L is the rate of height change for the threaded rod.

 

This means that error from desire rate depends on arm length, thread pitch, and rate of thread rotation:

 

E(t) = (Sidereal Rate) * t – 2 * arcsin(k * t/60s * (thread_pitch) / (2 * arm_length))

 

Where k is rod rotation rate per minute.

Assuming a 200mm arm, a thread pitch of 1.27mm per rotation, and a sidereal rate of 2pi/86164s we get the following error:

 

E(t) = 2pi/86164s * t – 2 * arcsin(k * t/60s * (1.27)/(400mm))

 

Letting t E [0s, 10770.5s] (Or 1/8th of a sidereal day and ~ 3 hours) we get the following graphs for different rates k:

This provides the linear displacement perpendicular from the camera arm.

 

There are two methods for compensating for this: a software method and a 3d printed error compensator method.

 

The software method involves approximating the theta value necessary and working backwards to derive the necessary linear thread displacement. This can be easily implemented on the side along with the other motor drivers and could prove to have a smaller latency than the hardware method.

 

The hardware method for tangent error compensation involves using a constant rate for the thread motor and printing out a curve whose altitude is the negative of the error shown in the graphs above. In turn effectively reducing the error to 0. The only additional modeling requirement for this method is to find the parametric equation for parallel displacement which for pi/4 radians is L*(1 – cos(theta)) over [0, pi/4] radians and time of [0, 10770.5]s as before. Due to the fact that this is a piece that is on the bottom of the actuating arm we require that the error is negative throughout so we would go for constant k E [0.8, 1.6].

 

The advantages for using this method over the software method is that once the piece is printed the sidereal compensating motor only needs to be run at a constant rate and all of the error compensation is contained within the 3d printed piece.

 

Further updates:

• I set up Arduino ADC logger that works through the Arduino IDE’s serial communicator.
• Most items that were shipped before spring break have arrived to my house so the only things that are left are our orders from last Friday that include motor/laser led controllers/drivers and further structural pieces.
• This week I will be setting up for measuring some of the demos possible as well as discussing our options for tangent error compensation.

 

For this week there were some previously unexpected design changes due to the COVID-19 pandemic and the resulting quarantines and public space shutdowns. The more dramatic impacts include: our inability to physically coordinate, lack of access to Techspark, and lack of access to speciality testing or verifying equipment. Due to these changes a few changes had to be done to the “compensator” portion of our design. Many of these changes were listed in detail in the recent SOW.

The main tasks I tried to coordinate this week was getting the supplies necessary for assembly in my home. Since I will be the one doing physical assembly and testing I need access to a few specialized tools. Furthermore, for certain, less accessible or more expensive tools such as an oscilloscope I began researching an alternative in the “Arduino oscilloscope”. As our project partner Yuyi pointed out the frequencies at which the controller works at makes this option viable.

Another thing that I wanted to explore is using rendered images by simulating point light and point light movement in a spherical grid. Using these simulations we can create more accurate mock data as input for the pi. This should provide another alternative to testing which could come in handy in increasing input data fidelity without having to do field testing in the probable upcoming quarantines.

For this week we made sure parts were ordered in hopes that we could start assembly the week after spring break. For the first half of the week we worked on fleshing out the report and budget making sure all part decisions were made. We decided that one of the stepper motor speed reducers we were looking into will probably be affected by shipping delays due to the rising COVID-19 epidemic and one of the backups I chose was for the wrong type of stepper motor so we had to respec the speed reduction ratio to 30:1 for NEMA 17 stepper motor. This means that I had to work on altering the reduction ratio of our compensator gearbox to a 43:2 for an effective ratio of 645:1 which is conveniently closer to our ideal 646:1 ratio than from our previous design although less of the speed reduction is being done with a well fabricated part and more with our own gearbox. For the rest of the week I was in between traveling and testing the polar axis alignment error calculator.