Project Logs

Collapse

• Weight Savings Round 1

  Dylan Radcliffe • 03/29/2022 at 23:35 • 0 comments

  So if there's one thing I love, its overbuilding things. Like I've said, this drone is rock solid, but that comes at a cost of efficiency. You can look at some of my original photos from this project to see how bulky some items were. The "T" Braces are particularly bad, and have been for most of this project. They started out as these behemoth literal T braces, that took 4 screws each (Each!) to secure in place to the carbon fiber. 

  
  

  Then, after much consideration, I replaced them with this far improved design. Not only were they significantly lighter, and required less screws, the airflow around the part was much less obstructed. 

  We can do better though. If you look from above, there's still a largeish obstructed area
  

  So I took the metaphorical hack saw to the part, and this is what I came up with.

  From the original part a year ago, we've got half the screws to secure the part, and less than half the size. I haven't attached it yet, but the 3D printing utility suggests a weight savings of 8ish grams each. Totaling a cool 32g, not including the reduce weight from requiring fewer screws to attach. In addition to the weight savings, I suspect there will be significantly improved flow around the part.
  

  While I'm at it, I've slimmed down the landing gear, and removed the legacy camera mount, further reducing the weight of the drone by 106g. By the time I replace the T brackets, I'll have reduced the all up weight by almost 10% Just writing that out feels great!

  The drone will be out of commission for a couple of days sadly, but I'm sure it will all be worth it!! Until next time!
  
  

• Fixing the Voltage Drop

  Dylan Radcliffe • 03/27/2022 at 21:41 • 0 comments

  So, for starters, let me say that this airframe has not been optimized yet, but while in the air, it handles like a champ.  That being said, the voltage drop was unacceptable when I pulled the drone out for a flight the other day. I was getting 3v-4v drop on takeoff, plunging my battery into the red zone. This meant that I could get a flight time of around 2 minutes, which is far less than the theoretical 7ish minutes I'm expecting with the current all up weight and battery combo. So I set to work trying to reduce this massive voltage drop.
  

  It was confounding me a little bit as to why this was happening, my battery should have no issue pumping out the required current, so I figured the problem was elsewhere. I started by resoldering some of the connections on the power distribution "brick" that I've been using. After fixing up some of my poor soldering points, I actually decreased the voltage drop by about 20%. Not bad, but still less than ideal. 

  So I started to have a real good look at my connectors and other solder joints, and then I realized that one of the connectors I was using had 14awg wire! This was definitely a problem, as the main power distribution ideally should have at least 12awg wire, perhaps even 10awg. since the power module used 12awg, I figured I'd start by using that gauge, and adapt if needed. 

  Unfortunately in my town, hobby supply stores leave something to be desired, I was able to find some new XT60 connectors, but there was no wire to be found! I ended up finding some 12awg wire at the TV and Stereo store, but for some reason, it would not accept solder!!!! Man, I was so frustrated.....

  Anyways, I try to avoid Amazon where I can, but I bit the bullet here and just ordered some 12 gauge wire with silicone sleeves.
  
  After installing the new 12awg wire, my flight time skyrocketed, instead of 2ish minutes, I managed a cool 5 minutes. I pushed it a little too far, and my battery voltage dropped off the moment I hit 12.8v. For a 4s battery, that's pretty rough, but hopefully I can avoid that in the future.
  
  5 minutes is still a little bit low for my liking, but I think there's some major improvements I can make in the near future! With the current setup, the theoretical limit is about 7 minutes. In the coming weeks, I'm going to focus some attention on cleaning up the airframe to reduce the weight where I can. Some ideas I have for reducing weight:
  
  - Removing redundant screws (9 I can think of off the top of my head)
  - Reduced size of T connectors.
  - Replacing 3D printed landing gear with carbon fiber plate
  - Redesigning the power brick with a custom PCB
  - Vacuum formed seed hopper
  - A lighter DC/DC converter
  - Integrate battery mount into frame
  - Possibly replace electronics base with carbon fiber plate.
  - Possibly redesign autopilot vibration damper
  
  Once I've done all this, I think it'll be time to think a bit about optimizing the size and / or numbers of the batteries. I'm waiting for a pocket scale and some copper clad to come in the mail, so in the meantime, I'll try to get out for a flight or two to play with the seed spreading capabilities and maybe get a fun little video for you guys. Stay tuned!
  

• Back At It

  Dylan Radcliffe • 03/18/2022 at 21:38 • 0 comments

  Well, its been a while.

  I started a new job last summer, and a lot has happened since then. I'm finding some time and energy again to get back into hobbies and tinkering. So I've picked up the Aspen and we're gonna plant some trees!

  So, I figure a natural next step will be to work on an "Aspen V1.1". In all of the flights I did last year, I identified several "bugs" that I'd like to address. Think of this as more of a "Quality of Life" update. 
  
  

  Problem	Cause	Solution
  Drone naturally drifts to the left	Drone is unbalanced	Relocate / Redesign ESC Module
  ESC Rack is Melting during hot days	ESCs are overworked / undersized	Relocate / Redesign ESC Module
  Less than Optimal Flight Time	Unoptimized Battery Size / Unnecessary Weight / I was poor when I built this thing	- Reduce Weight - Replace some parts with Carbon Fiber - Replace some parts with vacuum molded parts - Model Flight with Different Weight Batteries
  Flying this thing gives me anxiety	High potential to lose telemetry signal and lose the aircraft because I'm an idiot and cheaped out on parts.	Improved telemetry, YAGI Patch antenna and antenna tracker
  I can't see what the drone is doing!	No Camera	Install a camera, possibly FPV? Not sure yet.
  Drone Gets Dirty	None of the parts are covered	Design a canopy to cover everything.

  Allright, so theres some stuff to chew on. Some of these things probably could have been avoided if I knew what the hell I was doing when I started, but alas.....
  

  As much fun as I find doing physical design work, such as the canopy, I think I'm gonna hold off on that until I get some other items sorted. To start, I think I'm going to take the drone out on a series of test flights, this is gonna help me with 2 things.
  

  1. Help me optimize the battery size.

  2. Give me an idea of what the issue is with the ESC and measure the power draw.

  So for my first task, I intend to do 6 flights, with and without a payload. This should give me a decent picture of the power situation on board. At that point I can start thinking about weight optimization of the battery and / or other parts.
  

  Report Back:
  

  So I didn't really have a firm grasp of the power problems that this drone was facing. My battery seemed overpowered for the drone, but I was getting a massive voltage drop right after takeoff, something like 3v. I wasn't sure why this battery was struggling! The problem made the flight time terrible compared to what I should ideally be getting. I went on a long quest to determine the cause of the issue.
  

  I ended up deciding to re solder some of the connections on my PDB. After completing the resolder work, the voltage drop was reduced by about 25%, but still nowhere near what it should be at. After doing some more digging, I realized that one of the xt60 connectors was using 14awg wire. Ideally, this drone should be using 12-10awg wire for the main power, so I'll be working on replacing those connections. xt60 connectors ideally should use a max 12awg wire, so we'll go with that. If the voltage drop is still significant, we'll look into what it would take to set up the main power with 10awg. I may end up designing a custom PCB since I'll need to replace the voltage monitor at the same time.
  

• Quick Little Update

  Dylan Radcliffe • 06/04/2021 at 15:35 • 0 comments

  
  

  So I took a bit of a break from working on the Aspen, and in the meantime I actually secured some employment that I will be starting next month. I wanted to share with you all some of the next steps I'm working on with this project before I release it to the masses. One advantage of getting a job is that I can probably start to afford some of the necessary components to continue experimenting on this platform. Up until this point I've been working on a bit of a shoestring budget, including a slight dip into my savings.
  
  So I've got a laundry list of items that I would like to work on including some that are longer term. Some of them include:

  - Fully enclosed, lightweight canopy (possibly vacuum formed?)
  

  - Expanded Hopper Size (and lightweight material)
  

  - Optimized Flight Time / Battery size
  

  - Obstacle avoidance and precision altitude control (Lidar?)
  

  - Easier battery swapping (Might be challenging)
  

  - Automated / optimized seed site selection
  

  - Swarm (ie. "Flock") capabilities

  - Precision landing control

  - Long range telemetry
  

  Ideally I would like to get the drone to a point where within 10 minutes of arriving at a given site, the drone can be up in the air and planting seeds, possibly with multiple units working simultaneously. Its a bit of a dream to have all the drones working at a single "base station" where they come to receive battery and seed "refills" when either item begins to run low. Using this system I would hope to create a seed planting base station that can cover a 1km-2km radius.
  
  In the medium term I would like to release the design for people to be able to build their own copy of the machine. Are there any features you would like to see included? Let me know what you think!
  

• A "Final" Video

  Dylan Radcliffe • 04/20/2021 at 23:31 • 0 comments

  Its been a long few months since the Hackaday Earth Day challenge was first announced, but I've finally got a working prototype of the Aspen V1. It deployed it's first seeds on a field 2 days ago and has had many successful flights since that time. Please enjoy this "launch" video of my drone and thank you Hackaday for hosting the 2021 earth day challenge!
  

  Transcript:

  This is the Aspen V1, a dedicated environmental restoration drone platform for planting seed on a massive scale.

  On June 5^th 2021, the United Nations will launch the decade on environmental restoration in the aim to prevent, halt and reverse the degradation of ecosystems worldwide. The simple fact of the matter is, simply conserving nature for future generations is no longer enough, we need to be actively repairing the damage brought on this world over centuries past. Species once considered abundant, even not long ago when I was a child are being pushed to the brink. Forests myself and many friends once explored are now housing subdivisions. We need stewardship solutions and tools that scale massively beyond our current capacity. Locally, regionally, nationally, and globally incredibly talented people have come up with tools and plans to restore our natural environment. We know what we need to do, we just need to turn our plans and resources into action.

  To celebrate the launch of the UN decade on environmental restoration and Earth Day 2021 I dusted off some of my old notebooks dating back to 2014 to design and build the Aspen V1. A dedicated environmental restoration drone for spreading seed on a massive scale. This platform can be used for rapidly spreading up to 3kg of different seeds including tree, tallgrass, or wildflower seed! The broadcasting wheels is interchangeable and adaptable for a variety of different seed types. The drone is smart and fully autonomous, built to be capable of completely covering a hectare of land in less than 5 minutes.

  It’s not the first seed planting drone ever conceived, since 2014 there have been several successful designs. My desire was to build one for less than $1000 using as many commonly available materials as possible and operated by a minimal crew. Many of the parts are 3D printed using innovative 100% recycled plastic and as much as possible was purchased from local hardware stores. Currently I’m in the process of pushing the limits of what this machine is capable of, but the plan is to put it to work this year if some collaborators are excited to test the machine in partnership.

  My passion is this beautiful earth we live on, and my desire is to dedicate my life to preserving and restoring it’s natural beauty. Working in the not for profit and environmental sector has blessed me with the opportunity to develop a somewhat comically wide skill set. This invention is my way of putting some of those skills to use while I look for work after graduation.

  In the coming years many of us share a dream where we have halted our production of CO2 and are on a path to a more sustainable future. Part of the path is mitigating the ongoing biodiversity crisis we are facing and restoring the natural environment. We know what we need to do, we just need to put the resources and time into making it all happen. The Aspen V1 will hopefully become a small part of the solution to the big problems we are facing.

• Hoppers... Everything I didn't know I needed to know

  Dylan Radcliffe • 04/12/2021 at 21:56 • 0 comments

  Since I had a decent platform for the drone up and running, I figured I needed a seed broadcasting mechanism. Arducopter makes controlling an additional servo initially, all I need to do is pass the RC input to a pin on the servo rail and Bobs you're uncle! I found some decent continuous rotation servos to mail order and strapped them on to an initial hopper design. My initial design I dubbed "the snail" looked awesome and was designed to drop the seeds out of the bottom of the hopper by rotating downwards. It looked great and was destined for success.... until it wasn't.
  
  Unfortunately the cheap continuous rotation servos you can order online don't throw a lot of torque. In my initial tests the wheel would pinch the seeds between the housing and the wheel and instantly stall. Big sad indeed.
  
  Back to the drawing board! I was trying to avoid if at all possible using a gearbox to increase the torque of the motors. The servos don't turn that fast at top speed, and reducing that speed in exchange for torque didn't feel like a good option for a drone that needs decent control over the rate of broadcast. I ended up doing some research and found some cool videos of seed drills online. These designs eliminate any "pinch points" where the seed could be pinched between the wheel and the housing and instead lift the seeds out of the hopper and throw them out. Here's a video of one in action:
  

  
  

  So I set to work putting a design together. I went with a slightly different wheel design (Looks cooler too!) that collects several seeds at once for broadcast and dumps them out of the back of the hopper.

  The size and design of the hopper still needs a bit of work, but currently it can hold ~700g of bird seed for testing.
  
  That's the update for today! Stay tuned to see the design in action.
  

• Some Early Photos of the Aspen V1

  Dylan Radcliffe • 03/29/2021 at 17:38 • 0 comments

  Just for posterity, I thought I'd include some of the earliest photos of the Aspen.
  

• First Flights

  Dylan Radcliffe • 03/29/2021 at 17:23 • 0 comments

  So last week I took the Aspen out for a couple of test flights. PID needs a little work, but its in the air and doing what it was designed to do. You can see in the video and image below the payload attachment / hopper. This drone is intended as a seed dispersal drone for restoring tall grass prairie and potentially treated seeds for reforestation projects. (none of you guessed it!) I'll be submitting it to Hackaday's Earth Day Challenge this month.

  This weekend I added some landing gear to the drone. I found that it sat way to close to the ground for my liking, so I came up with a pretty smart looking solution that only required minor modifications to the existing frame. My partner says it looks a bit like some bridge in Philly? (I dunno, I've only been to the states once) I don't think I'll be adding any more to the frame at this point except maybe a minimalist protective cover for the PDB (Power distribution brick? :P) and ESCs. Right now it has a dry weight of slightly less than 2KG which is actually slightly less than it was 2 weeks ago. I have some ideas to hopefully shave off ~100g.

  Here you can see the drone fully assembled and up close.

  Thanks! That's this week's update. Let me know what you think! I've never designed anything this complex before so I'd love to hear your thoughts. Your feedback thus far has been invaluable. Cheers!

View all 8 project logs