Shop OBEX P1 Docs P2 Docs Learn Events
Ken/Cluso99/ W9GFO/JasonD's QuadCopter Build Log (updated info ELEV-8 + availability) - Page 5 — Parallax Forums

Ken/Cluso99/ W9GFO/JasonD's QuadCopter Build Log (updated info ELEV-8 + availability)

1235743

Comments

  • Ron CzapalaRon Czapala Posts: 2,418
    edited 2011-08-26 18:10
    W9GFO wrote: »
    That is some incredible flying. I can't believe how long it took him to crash that helicopter.

    There are many similar videos and these YouTube channels:

    http://www.youtube.com/user/CrazyTomVideos

    http://www.youtube.com/user/multistratus (some flying at night!) Also flying electric ones in the same manner...
  • Capt. QuirkCapt. Quirk Posts: 872
    edited 2011-08-27 13:35
    FROM Post #79
    Ken Gracey wrote: »
    It flew very well at sea level on the California coast with lots of wind, but poorly at 7,000 feet above sea level (more on this below where I show the two motors I've been running).

    attachment.php?attachmentid=84016&d=1313364499

    Ken Gracey

    You know with a engine at altitude we would use smaller props, but I
    know that doesn't apply here.

    (1) I realize you have already discussed increasing the blades from 2 to 3. But I was
    curious about the pitch and diameter of the original two blade prop was? and the pitch and diameter of
    the 3 blade---if that is what you are currently using.

    (2) What rpm are the motors turning, and how is the performance now?
    Is the acceleration great or so-so? Do the props make more noise?

    (3) Something I am really curious about is; does the quad-copter have a sweet spot? at a certain
    rpm. Especially since the speed controllers have a set rate of acceleration? If so, has it changed
    with propellers or type of motors?

    On all airplanes there is a "thrust line" for stability and lift, 2 to 3 degrees down works
    well--but many are simply set to zero. Depending on the torque of a engine they may also shim a motor to the left or
    right to compensate for torque.

    (4) Does the Quad-copter have any thrust-line? or is the motors simply mounted
    perpendicular to the booms?

    (5) And two other questions, what is your flying weight and how many volts are supplied to the motor?

    Lastly, don't take this the wrong way, but I still think those motor booms need to go! and replaced with
    something stronger and more durable, without increasing the weight significantly.
  • Cluso99Cluso99 Posts: 18,066
    edited 2011-08-27 13:56
    Ken & All:

    My interest lies in the electronics. I have professional Radio gear bought for the purpose, so that is where I am starting.

    A frame available via the forum would be great. This way, a set of motors and escs can be specified which will work with the frame dimensions. I know there is a lot of experimentation in this area, and a lot of conflicting info. Having one set that works makes an excellent basis.

    My (electronics) solution is to use a prop in a tiny enclosure to protect the electronics. I have 2 pcbs done and back now (just a small prototype run). They just stack together.

    The first pcb is an smt prop pcb (just the prop, eeprom, xtal, 3v3 regulator (from 5-6V), expansion conectors, and an optional transistor reset circuit. It is a modular base prop platform pcb that I will be using for a number of projects and it measures just 45x45mm (1.76"x1.76") and fits a Hammond 1551 R/S box.

    The second pcb is a daughter pcb (t/hole 30x45m) which contains up to 16 servo headers. Each servo header has a separate series resistor (to protect the prop) and a separate enable link for the 5V. The 5V link is required (to be isolated = unconnected) for the ESC connections because many ESC controllers provide 5V for receiver (etc) logic and it is not wise to parallel these. Each servo output is labelled for N,E,S,W (for quadcopter) and extras for NE,SE,SW,NW (for hex/octocopters). Then the input servos (from the receiver) are labelled THO, AIL, ELE, RUD (for throttle, ailerons, elevator and rudder). There are then CH5, CH6, SP1, SP2 (for up to 4 more channels). None of these are mandatory, but provide a reference. These extra 4 channels require a wired connection to the prop pcb. Also, a servo header is provided for power in or out.

    I plan to do an ESC (most likely 4 ESCs on a pcb) replacement pcb, but this will come later. For now, I am just using bought ESCs. This should reduce some of the wiring clutter.

    I may also do a tiny adapter pcb to take the wii gyro and accelerometer pcbs.
    I figure, at least for now, that the gyro and accelerometer are best obtained from Wii Motion Plus and Wii Nunchuck clone devices which are easily and cheaply obtained from eBay. These chips are tiny and difficult to solder, so a pcb that these little salvaged pcbs attach to makes more sense to me.

    I plan to use open source software and because of time constraints, could certainly do with some help here. I know Jason and others have the basics working on a prop. The prop certainly makes more sense to me than a single core cpu.

    Later options may include: Digital compass ic, pressure/temp ic, GPS, misroSD. I already have all the ICs but they are all QFN chips.

    Some alternative options include:
    Alternative radio control by iPhone/Android devices, or WiFi etc.
    Camera control.

    PS. Ken, sorry to clutter your thread. Thought you would not mind since you mentioned me.
  • Ken GraceyKen Gracey Posts: 7,386
    edited 2011-08-27 20:46
    Cluso99 wrote: »
    PS. Ken, sorry to clutter your thread. Thought you would not mind since you mentioned me.

    It is my hope that you would contribute in this way. Everything is welcome in this thread.

    Ken Gracey
  • Bill HenningBill Henning Posts: 6,445
    edited 2011-08-31 21:32
    I am officially jealous; I need to make some time to build a quad copter...
  • W9GFOW9GFO Posts: 4,010
    edited 2011-08-31 21:52
    You know with a engine at altitude we would use smaller props, but I
    know that doesn't apply here.

    Can you explain?
  • mindrobotsmindrobots Posts: 6,506
    edited 2011-09-01 05:12
    I would think at higher altitudes, with lesser air density, you'd need/want to increase pitch and/or increase prop diameter to get similar low altitude performance.
  • Ken GraceyKen Gracey Posts: 7,386
    edited 2011-09-01 08:39
    mindrobots wrote: »
    I would think at higher altitudes, with lesser air density, you'd need/want to increase pitch and/or increase prop diameter to get similar low altitude performance.

    I think so too. With lower air density it would take much higher RPM to attain the same amount of lift. Further, these little motors move out of their power band at high RPM, so it seems one could compensate nicely with larger propellers. I'm getting by fine with 10" propellers at 6,000 feet, but they fly wonderfully in the middle of their power curve at sea level, with tons of power to spare for fast vertical flight. I'll try some 12" propellers soon at high elevation just for comparison - if I can get sea level performance somewhere in the middle of the motor's power band I'd say it's a good modification.

    Ken Gracey
  • jdoleckijdolecki Posts: 726
    edited 2011-09-01 12:09
    Ken, You know ther is limit on how high you can go and I think its like 400ft. With a COA certificate and in regulated Airspace.

    Dont quote me on the 400 but its around there.

    For now the FAA will allow the RC community to police themselves.

    The is ane editoral in Robot or Servo mag I can rember which one.
  • Roy ElthamRoy Eltham Posts: 2,996
    edited 2011-09-01 14:50
    jdolecki: Ken is talking about being up at higher elevations to start with. As in, up in the mountains. He's not actually flying the quad up to 6000 feet above ground level. :) In any case, the air is thinner up in the mountains and it impacts the quadcopter performance.
  • GadgetmanGadgetman Posts: 2,436
    edited 2011-09-01 15:03
    What about adding ducts around the propellers?
    (Yes, I know the gap between the prop and the duct wall must be very narrow for it to have any effect.)
  • Capt. QuirkCapt. Quirk Posts: 872
    edited 2011-09-01 15:46
    W9GFO

    Unless turbochargers, and/or superchargers are involved; naturally aspirated engines loose power with increases
    in altitude, compression decreases, mixtures must be leaned out and more. An engines cam timing or cylinder port
    layout becomes less effective.The result is an engine simply makes less horsepower. And something I find personally
    amusing, is that most "Race Gas" decreases horsepower further.

    The engine will no longer rev to the same rpm, acceleration is reduced. In some cases the extra load on the engine
    increases combustion temps, and leads to detonation, followed by preignition and finally engine failure. Most efforts to increase
    horsepower for a certain altitude or density altitude will still require a gearing change.

    In order to have the best possible performance at an altitude; is to gear it accordingly by reducing the
    diameter or pitch. The exact same methodology applies to increases in density-altitude too.

    In any type of competitive racing, they will have morning, afternoon, and evening setups to remain competitive throughout a day.
    But this is something most enthusiast never notice.
  • Capt. QuirkCapt. Quirk Posts: 872
    edited 2011-09-01 16:34
    The main reason why I posted was about how http://www.rcgroups.com/forums/showthread.php?t=1006721. was testing the propellers.
    I have only read about a hundred posts or so, but test stands, IMO, is a deeply flawed way to test model airplane engines (or motors) and propellers.
    Especially when responsible or trusted authorities like magazines and manufactures report this kind of data.



    Ken,

    This project is reminiscent of your Quad Rover project.
    Is this another project biased towards an engineering universities for a lifting contest, or for carrying a camera in a uav contest?

    Even if it isn't, Is Parallax working on an BLDC motor app-note? I am sure the idea of the Prop2 controlling 4 or more BLDC motors is driving
    more people crazy, than just me!
  • W9GFOW9GFO Posts: 4,010
    edited 2011-09-01 16:58
    Why do you not like the test stand? Multi-rotors spend most of their time hovering - or nearly so. The test stand appears to be a very close simulation to what the craft experiences in flight (hover).
  • Capt. QuirkCapt. Quirk Posts: 872
    edited 2011-09-01 19:21
    Because a single airborne propeller perform differently than they do on test stands for many reasons.

    Next is how the data is collected; in order to compare data from a test stand of any kind, it should be
    collected all on the same day, with the same motor, and completed within a few hours after starting.
    Best results would be from testing completed while air temp and humidity is close to being the
    same throughout the test.
  • W9GFOW9GFO Posts: 4,010
    edited 2011-09-01 19:53
    I would be interested to hear what your solution is for making motor/propeller/esc comparisons.
  • Capt. QuirkCapt. Quirk Posts: 872
    edited 2011-09-01 22:44
    Test them individually to graphically represent each components strengths and weakness.

    1. An electric motor brake style dyno to test the actual output of a motor combined with a given
    speed controller. Unfortunately this is dependent on a commercial speed controller. Ideally independent
    motor results would require a custom built speed controller capable testing any motor, and graphically
    represent normal data, and data that would allow someone to optimize motor / speed controller combinations.

    2. A speed controller dyno. I am not certain about how this would be designed, but it needs to quantify the
    differences between other speed controllers. Then graphically represent the strengths and weakness of
    that particular controller. It needs to provide the data that manufactures will not supply. A test that enables a
    magazine, or an informed consumer to recommend reasonable controller / motor combinations.

    3. Every propeller is going to have an individual torque requirement at a given rpm. Depending on the airfoil
    used, the twist, the material, etc. A propellers torque requirements are going to change with rpm, and I doubt these
    graphs will look similar to any normal dyno graph. They should show the areas of inefficiency vs efficiency throughout
    an rpm range. The areas in which the prop is efficient should correspond directly to a torque requirement, or possibly
    a rpm/volt number.

    For gasoline powered giant scale model airplanes I have a dyno, and I needed a prop dyno. Now I need to build a
    motor and speed control dyno.

    Bill M.
  • jdoleckijdolecki Posts: 726
    edited 2011-09-02 04:11
    Oh. That would of been pretty cool thought.
  • Cats92Cats92 Posts: 149
    edited 2011-09-03 01:59
    Hello,

    building a quadcopter is not easy also I like the idea of going step by step.

    This frame kit is a very nice first step to have something solid to start with (thank you Ken).

    Using the Hoverfly sport you can make it fly quickly (I have an Aeroquad frame with Turnigy 2217 + Hoverfly Sport and begin to make it fly)

    But after that Propeller users are hoping a board with open source software (It seems Jason and Roy are on something ? ).

    By the way : what lipo batteries are you using wit the frame (mAh, volts ?) and how many minutes can you fly ?

    Jean Paul
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-09-03 20:45
    Ken, will you make this robotic with some degree of AI? I'm working on a mini wireless helicopter and it takes some skill to fly it. After understanding it and flight piloting details and nuances, I plan to transfer various "robotic" functions over to Propeller chips. This will obviate working with the wireless controller and hacking it. The Propeller could eventually become the controller. I'm always thinking of future projects and techniques of robotics and swarming are now possible with a second copter just purchased.

    http://forums.parallax.com/showthread.php?124495-Fill-the-Big-Brain&p=1033113&viewfull=1#post1033113

    These new mini's are great, with gyroscopic flight and the capacity for light payloads such as Propeller chips, sensors, and cameras. The overall investment can be low for hobbyists, as low as $17 for a mini, $8 for the camera and a few dollars for sensors. It makes it possible for someone to get started on a shoestring budget. It could be a nice start leading to quads.

    Here's a motion sensor project for a Micro Space Airport.
    http://forums.parallax.com/showthread.php?124495-Fill-the-Big-Brain&p=1032658&viewfull=1#post1032658
  • idbruceidbruce Posts: 6,197
    edited 2011-09-04 04:15
    Very cool project. Hope you have a lot of fun with it.

    Bruce
  • ypapelisypapelis Posts: 99
    edited 2011-09-08 12:50
    Ken,

    Well done on this project, I really enjoy seeing folks taking on engineering projects as entertainment, it makes me feel less alone in the world! When it is done so competently and professionally, it is a real treat, thank you!

    Depending on a variety of factors (mostly outside my control), my research team may end up needing to build a small fleet of quad copters for R&D purposes and I am very interested in the progress of your project, and also contributing as much as I can as well.

    But to start, I have a simple question regarding the frame. Did you consider using a carbon fiber frame in lieu of aluminum, and if so, what made you go with aluminum?
  • Ken GraceyKen Gracey Posts: 7,386
    edited 2011-09-08 16:00
    Hey all,

    I apologize for the delay in making replies to this thread. We're really excited to have some interested folks contribute. Let me bring you up to speed in what we're doing about quadcopters at Parallax, hopefully answering some of your questions along the way.

    The "Product" Approach

    Quadcopters are something you can build on your own. With few exceptions from companies who specialize in building whole systems, these are usually built from scratch by collecting up the hardware on your own. And it makes sense in this way. From a business perspective, a business can only charge a significant profit to a product if they add lots of value. In this case the value comes from parties other than Parallax: Spektrum or Futaba (R/C radios); Hoverfly (the Propeller control board); HobbyKing (Turnigy motors and ESCs) and your local hobby shop. The only value Parallax can add is the frame, and the whole project gets expensive really quickly even without us trying to burden it with a profit.

    Therefore, from a product perspective it isn't practical to justify a major investment in design, documentation (more work than the design), packaging, and marketing because a company can't charge $500 for a whole kit because you can buy the pieces on your own for far less.

    And further, these aren't for everybody. If you're into R/C, some prior experience helps. If you want to make it a UAV, then you need to program the Propeller. It's a niche audience, but it's a fast growing one and we need to be there. There is no better processor for a quadcopter than the P8X32A!

    But this environment makes it perfect for us to contribute! Parallax will have an informal, community-driven approach to the quadcopter kit: totally open, low-cost, and CC3.0 attribution on everything we offer. This isn't a profit center for us, but since we have all the manufacturing machines required to make the parts, kitting facilities to box everything, and an interested community we'll make it happen.

    Hoverfly OPEN Board and the kit

    Parallax's kit will include everything except the battery and radio. You can source these on your own, because all we'd do is make them more expensive. We're out making arrangements with APC, HobbyKing and local suppliers for low-cost material. The cost of the kit would be about the same as if you bought the parts on your own (plus a small margin for our tools).

    The best part of the kit is the Hoverfly SPORT and OPEN control system. Initially we will sell the Hoverfly SPORT board, but by November we will stock a totally new version made specifically for you called the Hoverfly OPEN. This will be a simple design with gyro, but external connections for GPS, accelerometer, and barometric pressure sensor. You can write your own code or use the code from Hoverfly.

    Documentation would consist of a well-designed forum thread. No book, no PDF, it'll all be right here.

    Current Airframe Design Issues

    The airframe has been flown quite a few times and it has the following issues that we are correcting in another internal beta test with three more units:
    • Bottom motor mount hole is too small for Turnigy 2217 shaft collars. Revise drawings to increase diameter of hole.
    • Top/bottom of motor mounts have broke twice on crash landings at the standoff, but remainder of airframe has had no damage. Slight increase to outer diameter of top and bottom motor mount should fix this problem.
    • Hoverfly SPORT/OPEN/PRO mounting holes are off. Put mounting holes in the right location on top/bottom of frame plates and booms.
    • Power distribution system currently uses pre-made harness. We will decide to continue purchasing the Aeroquad harness or build our own.
    • Landing gear design is rigid and on the outter booms, intensifying center frame pressure on crash and making liftoff more difficult. Design landing gear which extends from the center frame outward, providing some reflex and accommodating a GoPro camera or additional electronics for FPV flight.
    • Battery attachment is currently done by zip tie. Provide a better way of quick-change battery installation/removal, maybe with velcro loops.
    • Frame is a bit heavy and a few opportunities for lightening it up exist. Frame will be lightened with 0.090 Delrin instead of 0.125 Delrin. Replace aluminum standoffs with plastic spacers.

    The question about aluminum instead of carbon fiber. The AL6061 5/8" tubing has a small wall of 0.035" - it is very light, easily machinable, and has no tooling costs to manufacture. Carbon fiber is the optimum material for this project, but it has a significantly higher cost and lower machinability. None of the aluminum booms have been bent in the 20+ crashes I have experienced (many of these are intentional, flying around in the office at high speeds).

    The total weight has not been an issue, even at elevations of 6,000 feet carrying GoPro cameras. I don't know the payload, but it'll carry a Boe-Bot or S2 nicely [or a small fleet of Penguins].

    This airframe is intentionally simple. There's not much here beyond motor mounts, booms, and body plates. You can drill it, attach your own parts, and take it apart easily. We didn't want to make a design which isn't understood at first sight, or which requires specialized parts to repair. The most commonly broken parts are propellers and motor mounts - this is easily addressed by throwing extra sets in the kit. I *hate* being captive to the E-Flite helicopter parts wall at the hobby store - thankfully this project will cause no such dependency on us to stay in the air. Your ground to airtime ratio will be very high, I assure you.

    Next Steps: Build Three Kits for Internal Use and then 25 for Forum Members

    Yesterday we had a meeting (Jessica, Matt G, Nick, and Kevin) to discuss how to proceed. We will make changes to the design and build another three kits. Pending positive results we will make 25 kits for the forum members. These kinds of products evolve, so there will be changes in the future. If you want to replace certain parts in the future we will ensure backward compatibility as far as hole mounts are concerned, allowing you to make an upgrade or replace certain parts. If you buy a SPORT board but want the OPEN when it is released, then we will provide the OPEN at a tremendous discount.

    The individual parts will be available in October. The whole kit should be available by the end of October.

    The Quadcopter Contest: June 2012

    Parallax is working with OBC to significantly increase the UPEW event at Parallax to be a larger electronics exposition. As part of this event we will host a UAV contest at our office. You will be able to use any craft and any processor. We want to encourage participation from everybody. More on this to follow.

    I see Parallax being a good facilitator of this activity as a hobby, for research, or for UAV purposes.

    I'm sure there's more to say, many questions unanswered, etc.

    Ken Gracey
  • Ken GraceyKen Gracey Posts: 7,386
    edited 2011-09-08 16:08
    Cats92 wrote: »
    Hello,

    building a quadcopter is not easy also I like the idea of going step by step.

    This frame kit is a very nice first step to have something solid to start with (thank you Ken).

    Using the Hoverfly sport you can make it fly quickly (I have an Aeroquad frame with Turnigy 2217 + Hoverfly Sport and begin to make it fly)

    But after that Propeller users are hoping a board with open source software (It seems Jason and Roy are on something ? ).

    By the way : what lipo batteries are you using wit the frame (mAh, volts ?) and how many minutes can you fly ?

    Jean Paul

    Hey Jean Paul, I'm using a 3S 3300 mAH battery and it is providing 10-15 minutes of flight, easily. It depends on how you fly - stunts and vertical flight draw a lot of current. Hovering around and getting familiar with the quadcopter makes the battery last twice as long. And yes, I think there will be other efforts for control systems to mount on this frame. Cluso99, Roy and Jason all seem to have something in the works.

    Welcome to the action!

    Ken Gracey
  • ypapelisypapelis Posts: 99
    edited 2011-09-08 17:07
    Ken,

    This is great news. I would definitely line up for one of the kits, and if things go well in our project, I may PM you for possibility of getting a batch made for our group.
  • W9GFOW9GFO Posts: 4,010
    edited 2011-09-08 17:24
    This is good news, I would like to be included in this.
  • vanmunchvanmunch Posts: 568
    edited 2011-09-08 18:21
    The Quadcopter Contest: June 2012

    Parallax is working with OBC to significantly increase the UPEW event at Parallax to be a larger electronics exposition. As part of this event we will host a UAV contest at our office. You will be able to use any craft and any processor. We want to encourage participation from everybody. More on this to follow.

    This sounds really exciting! Count me in!
  • Cluso99Cluso99 Posts: 18,066
    edited 2011-09-08 21:54
    Great news Ken. Yes, I have pcbs done. But I have not started the software yet. The open system is great. Unfortunately I had to go interstate for >1 week which has dampened my progress.
  • Cats92Cats92 Posts: 149
    edited 2011-09-09 00:42
    Good news,

    Count me for one of the kits.

    Jean Paul
Sign In or Register to comment.