I've used those motors in the past, and while they worked ok, I wasn't very happy with them. It's true that you get what you pay for, and most brushless motors are in the $50+ range. I burned out a couple, even though I was running them at less than their rated amperage. The 400XT motors I'm using actually work well even above their rated amperage (I've run them at 15a before). I also didn't like the fact that the shaft on them is so small, because it makes using "real" props harder. They work fine if you're using GWS props, but anything else won't have spacers to fit that small shaft.
The speed control in that package can't be set to fixed throttle end points, which is going to cause you a lot of problems. Unless otherwise stated, assume the ESC uses "auto-endpoints" (most do). Basically, the ESC uses the previously seen minimum and maximum throttle values to decide what it's input range is, which means the throttle values you're outputting are being scaled by a constantly changing factor, and they're not necessarily the same for each ESC. The ones I'm using can be programmed to use fixed throttle endpoints (I have them set to use from 1.1ms - 1.9ms). The "acro" series can be programmed with an external programming box, and have a bunch of options including endpoints, braking, acceleration, cell count, timing advance, and more. They're still freakishly cheap. [noparse]:)[/noparse]
I've attached a pic of my board. The stuff I've tinted yellow is left from a previous project - it's not connected, so ignore it. [noparse]:)[/noparse] I left the center ESC pins connected, since they're just regulators - They can be run in parallel without problems. I connected them to the 5v rail on the board, and the 3.3v regulator converts that to power the prop. The R/C receiver is connected to the 5v rail too. Nothing fancy required.
I have collected a few of the 400XT motors, can't remember how many, they are packed up right now. Probably 2-4. I get them at Hobby-Lobby when they go on sale for as low as $14.00. I live only 14 miles from the place so I can go up there and look at them before I buy. I will not have that option in a couple of weeks as I will be moving 900 miles away. [noparse]:([/noparse]
I am seriously looking at the new line of eRC motors that they just picked up.
I'm not impressed with the 20AMP draw on the Speed 400 and 450's though. The 400XT will max out at 10 and will probably live in the 3-5 AMP range on a quad.
I'm going to look at the eRC's by the end of the week.
Jim
JasonDorie said...
Cluso,
I've used those motors in the past, and while they worked ok, I wasn't very happy with them. It's true that you get what you pay for, and most brushless motors are in the $50+ range. I burned out a couple, even though I was running them at less than their rated amperage. The 400XT motors I'm using actually work well even above their rated amperage (I've run them at 15a before). I also didn't like the fact that the shaft on them is so small, because it makes using "real" props harder. They work fine if you're using GWS props, but anything else won't have spacers to fit that small shaft.
The speed control in that package can't be set to fixed throttle end points, which is going to cause you a lot of problems. Unless otherwise stated, assume the ESC uses "auto-endpoints" (most do). Basically, the ESC uses the previously seen minimum and maximum throttle values to decide what it's input range is, which means the throttle values you're outputting are being scaled by a constantly changing factor, and they're not necessarily the same for each ESC. The ones I'm using can be programmed to use fixed throttle endpoints (I have them set to use from 1.1ms - 1.9ms). The "acro" series can be programmed with an external programming box, and have a bunch of options including endpoints, braking, acceleration, cell count, timing advance, and more. They're still freakishly cheap. [noparse]:)[/noparse]
I've attached a pic of my board. The stuff I've tinted yellow is left from a previous project - it's not connected, so ignore it. [noparse]:)[/noparse] I left the center ESC pins connected, since they're just regulators - They can be run in parallel without problems. I connected them to the 5v rail on the board, and the 3.3v regulator converts that to power the prop. The R/C receiver is connected to the 5v rail too. Nothing fancy required.
The speed control in that package can't be set to fixed throttle end points, which is going to cause you a lot of problems. Unless otherwise stated, assume the ESC uses "auto-endpoints" (most do). Basically, the ESC uses the previously seen minimum and maximum throttle values to decide what it's input range is, which means the throttle values you're outputting are being scaled by a constantly changing factor, and they're not necessarily the same for each ESC. The ones I'm using can be programmed to use fixed throttle endpoints (I have them set to use from 1.1ms - 1.9ms). The "acro" series can be programmed with an external programming box, and have a bunch of options including endpoints, braking, acceleration, cell count, timing advance, and more. They're still freakishly cheap. [noparse]:)[/noparse]
Thanks Jason for·the insight into the ESCs. I didn't fully understand what was happening. Also thanks for the motor info. As I only have 1 motor/esc combo for an unbuilt flying wing it's not too much waste. Thanks for the photo - it just brings a little reality to be able to visualise the simple setup.
hover1: Thats a serious motor, but then again it's serious work that it's doing!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ Links to other interesting threads:
There are several escs that you can program the range, if you power them up with the pwm signal > min they go into throttle programing mode, the pwm when powered up is the max value and then you take the signal to min and they save that range. After than as long as you power up with pwn set at min they will not change the range. Turnigy escs use that mechanism. Aeroquad configuration tool uses that to program the range during calibration.
I did some looking at how the aeroquad does its calibration and they support the Turnigy calibration method. From·reading around it looks like a number of other ESCs also support this method. The steps·they go through is
1. Connect all the escs and motors to esc (but dont put the props on the motors) and connect the ESCs to the CPU
2. Power up the CPU without powering up the ESCs - so you need an extra power supply to do this.
3. Set the output to all the ESCs to max value - command from PC
4. Connect power to ESCs - you get a set of beeps
5. Set the output to all the ESCs to min value - command from PC
5. Get another set of beeps and ESC throttle range calibration is done.
Their PC config app walks you through the steps above and commands the software to go through the steps, so you dont need the transmitter to do the calibration.
A few people have asked about my quadcopter PC configuration/monitoring program, so for those interested these are archives of the quadcopter code and the quadconsole PC app.
Its a work in progress, there is still lots of stuff that doesn't work but you can monitor/config the quad, capture sensor/imu/motors,etc and chart the output, etc.
The flight system uses a mix of gyros and imu, the gyros stabilize rapid changes and the imu handles more consistent tipping.
In the quadcopter code there are several options - I use the on-board imu - itg3200/adxl345/hmc6843/bmp085 , you can look at the bst project file for the defines I use. It contains code for several other options e.g. SF9 AHRS board, separate RC receiver inputs rather than PPM but most of these options I might have tested the driver but I will not have tried it in a quad.
I will update it as I work on stuff.
I managed to fly the quad into the ground yesterday (pilot error) and so needed to repair it today so I was adding a couple of upgrades - battery monitor and bluetooth while I was working on it.
I'm new to the forum and this thread, and also building my very own quadcopter :smilewinkgrin:!
You've all done a lot of great work so far:)!
I've still one question though: I cannot find out how to power the propeller board with just the BECs from the ESCs.
Can anyone explain me if it is possible and how to connect everything?
It would save a lot of weight if the propeller board doesn't need an extra power supply.
Many thanks in advance!
Depends on the prop board you have. ESC/BEC normally supply 5V, the prop uses 3.3V so you need a regulator for 5V to 3.3V. Many prop boards already have this.
A couple of examples, the prop backpack has an on board 5V to 3.3V regulator so 5V from ESC/BEC can be supplied directly to the board. A prop protoboard has a 7-12V to 5V regulator followed by a 5V to 3.3V regulator, so you can supply 5V from the BEC/ESC to the input to the 5V to 3.3C regulator bypassing the first regulator.
neat idea, a few things.. i havent read every single post, skipped some pages after page 4. With regards to the safety of the quad/hex copter design, any ideas how to keep the props protected from leaves/small tree branches? also, unlike a true R/C heli and gyrocopter, are you planning to code some kind of "low batt" emergency landing routine? As one of the major drawbacks to using BLDC's is that they tend to spin down quickly when main power is lost, where a heli can auto-rotate down and make a relatively safe landing by decending raidly to keep rotor speed up and bleeding off rotor speed in the landing flare. It would really suck to just see $180 of electronics fall from the sky and smack the ground.. or get too close to a tree and have it suck in a leaf and jam one of the motors -however using quality gyros like those used in R/C heli's would probably compensate for the sudden pitch change)
I am experienced in vacuum bagging & regular molding of fiberglass/carbon parts for automotive, also in making custom molds for those parts. I could design and built a design using fiberglass or carbon body, like the quad copter of that austrailian company linked to earlier in this thread with fiberglass bodies using ducted fans starting from around $25/complete shell.. If the majority is interested, i can begin designing the plug(s) for the molds now, that is if you have settled on a final physical layout. It can be with 4 separate ducted fan and single main body or i can make the body one solid peice similar to what is pictured earlier in the thread.
Or i can do a small production run of maybe 30-40 body kits in exhange for the guts to make my own or i can be paid via paypal, in either case i can use UPS, Fedex, or USPS for shipping depending on the final size of the bodie(s).
Also, thoughts on an alternate landing idea? like using a hook to "catch" the quad copter? Might be a good idea for you Cluso, being on a sailboat most of the time. The hook idea, first experimented with the skunkworks back in the 60's with one of their more parculiar aircraft that took of with caster wheels on the tail & landed vertically on a hook near the nose of the craft.. the design didnt go very far, but the hook idea with a big wire loopmigh be a good idea for landing such a craft on a constanly rolling & pitching platform.. although you could probably use a scheme that uses 3 or 4 ping)))'s or maybe even a CMu cam to detect a pattern on the landing and track it in software to make yaw/roll adjustments on final. i wonder if the PIR distance detectors would work better for something like that. maybe 3-4 ping sensors would be fast enough to detect a moving landing platform?
Anyway, if you and the majority in this thread would be interested in working out a deal with fiberglass or carbon bodies, just PM me
P.S. -personally id rather do the small production run of 20-30 complete body units with or without the molds to make more bodies to whoever in exchange for a complete set of "guts" for one of these.
I have added a battery monitor to my quad, the idea being to have a return to launch capability when the battery gets low though that piece isn't coded at the moment. I am working on a Video OSD using a prop backpack at the moment.
Another way to handle a motor loss is to have 6/8 motor, this also increases the payload capacity of the quad.
Depends on the prop board you have. ESC/BEC normally supply 5V, the prop uses 3.3V so you need a regulator for 5V to 3.3V. Many prop boards already have this.
A couple of examples, the prop backpack has an on board 5V to 3.3V regulator so 5V from ESC/BEC can be supplied directly to the board. A prop protoboard has a 7-12V to 5V regulator followed by a 5V to 3.3V regulator, so you can supply 5V from the BEC/ESC to the input to the 5V to 3.3C regulator bypassing the first regulator.
Thanks for your explanation!
I'll be going to use a USB Protoboard. My BECs can deliver 5V @ 2A each, but to reduce the stress on 1 BEC,
I want to use multiple BECs to deliver the current together, resulting in a lower current per BEC,
so I'll connect them parallel to the 5V input.
One other thing: Would the accelerometer be accurate enough to measure the minor translating movements of the quadcopter when hovering,
or is a gyro more useful, since it seems a bit overdone to take both gyro and accelerometer to have 3 axis,
instead of a single axis gyro (z-axis) and a 3-axis accelerometer. Anyone any experience with that? My setup: (which is still shipping :mad:)
I will check how I did it for the protoboard but in general I connected the escs to the protoboard in the 4 servo positions. Then you can configure the 5V from ESC though the servo power configuration. In terms of using multiple esc power, in theory you shouldn't but I haven't hadn't any problem with doing this.
In terms of acc/gyro - acc is too slow for stabilization, gyro is fast enough, however gyros drift. I use both - gyros possiblity for stablization and accelerometers to correct the gyro drift.
One other thing: Would the accelerometer be accurate enough to measure the minor translating movements of the quadcopter when hovering,
or is a gyro more useful, since it seems a bit overdone to take both gyro and accelerometer to have 3 axis,
instead of a single axis gyro (z-axis) and a 3-axis accelerometer. Anyone any experience with that?
Gyros ony detect/correct angular changes and cheaper ones tend to drift more than good ones, accelerometers are good for detecting/correcting drift.. so yeah i would give my autonimous brain as much info as it needed to successfully navigate. Many here may not be pilots, but flying a plane is probably three times more intense than driving a car, since in a car your just a line follower with speed and periphial detectors.. where in a plane your periphial detectors are now dealing with a sphere instead of more or less just a band of horizon, you have to worry about other aircraft in the area, your altitude, your atitude, airspeed, heading, roll, trim, what the engine is doing, if you need carb heat or not, and communications to control towers when entering restricted airspace. Its about the most complex thing a human can do all at the same time. Your also using your senses, smell, sight, touch, hearing, as well as other senses like your internal sense of 3 axis G and a human version of a gyro for detecting rotational changes on 3 axis.
I say write good code to make sense of all the data you can throw at it and it will be a better autonimous pilot.
Ducted fan motors tend to be more expensive but there is no reason why a standard motor/prop would not work within the protected shell. IIRC that is what the new US quadcopter is using. I think those who are already flying can comment further about a fibreglass shell but I think it would be great. I too have some experience with fibreglass, having helped in the production of my catamaran.
For some time I have thought a 'flying saucer' type shell would look great - with leds around the outer edge. It could be made with two identical halves (top & bottom). The biggest problem is in what size props and the span. There are lots of variations here. We will require a standard for using a fibreglass shell, but this is a great opportunity.
Regarding electronics, I have all the parts (gyro, accelerometer, compass, gps, pressure, prop chip, etc). I have just not completed the pcb for production due to other work pressures.
I checked with way I wired the protoboard and
1. The protoboard has connnection points for 4 servos in top right corner of the board. I put the 4 ESC connectors there.
2. The power connector for the 4 ESCs are connected together and is connected to the power switch, it then goes to a jumper just to the right of the poweer regulators, put the jumper from +5V to jumper, then when to move the power switch to position 2 (servo on position), 5V from the ESCs goes to power the board.
In terms of sensors, I use a 3 axis accelerometer, 3 axis gyro and 3 axis magnetometer. The 3 axis accelerometer and x/y axis gyro are fused using 2 kalman filters to get pitch and roll angles. Then the 3 axis magnetometer and 3 axis accelerometer are used to get a magnetic north bearing angle; that and the z axis gyro are fused with another kalman filter to give a yaw angle.
The flight system uses the x/y gyros directly for short term stablization and the fused pitch/roll/yaw angles for longer term stablizations. The gyros because you get a very fast input when the quad starts to rotate and the fused outputs give you an accurate orientation of the quad to keep it level.
Jason has found that the gyros alone are enough if you are manually flying the quad because you can manually correct for quad orientation issues. I would tend to agree though if you dont have much RC flying experience its hard to fly with just the gyros.
I also have a pressure sensor for altitude control though thats not tied into the flight system yet and a GPS for position control.
Generally the more automated you want the quad to fly the more sensors its going to need.
Thanks for the info. I switched the single-axis gyro for a triple-axis gyro (ITG3200).
(The order wasn't shipped out yet, so I could still change it )
@ Timmoore: When I downloaded your "quadbot1-bst-archive-100731-113807.zip", it seemed to be a broken file.
After repairing it, I still managed to read the files, but it seemed that some object files were not included or missing.
(For example the "itg3200objectLMM" as you described in you "quadbot1.spin" file.)
Do you still have these object files? It'll be very helpfull to get an idea on how to read the gyro outputs.
I have attached it to this reply. I have done a lot of changes, adding functionality and fixing bugs but the new version isn't ready yet. I will do an update once I have had a chance to give it some testing.
just got dental surgery today, recovering next few days, I will have an opportunity to rough out a universal design with two size ducting for the props. just a thought with respect to a moreducted fan approach using regular 3-5 blade props is to order these props oversize like 15-20" props with the tips cut off to fit in an 8-10" short duct. I'll draft a quick design up in google sketchup and post a few different views of what I got in mind. basically a central body using semi-flex wire landing gear, central body will be a few inches thick, motor struts will be tubular & can be either adjustable length telescopic &/or adjustable angle pods via 2-4 servos. The overall contours blending well. I'm on my DROID cuz my internet is down for a while tonight. I'll come up with a few design variations and see which ones are popular.
RinkCustoms: I would think allowance for 10" blades would be good.
I was thinking of a single shell design, with an identical top & bottom. Maybe the overall dia of the shell would be ~30-36". This way the motors could be mounted on 4 aluminium "U" tubes, each mounted just across each opening. This makes the motor mounting simple.
Otherwise, the shell could have this feature inbuilt but would require a little extra fibreglass strengthening here - in fact probably the preferred method - saves cost and time. In fact if the fibreglass struts (i.e. the part remaining when the shell is cut to open the flute for the fan (airway) could be 90 degree opposed making a 4-way strut with a circle in the centre joining the 2 halves. This could be done by rotating the shell 90 degrees top to bottom, so 2 struts would be on the diameter line and 2 struts as a tangent of the circumference. I hope I have explained this - probably a pic would be better so I will try and do when time permits and post.
I don't see a need to rotate the arms (motors) or extend them.
A total shell like this would not require landing arms as the shell would be fine to land (on grass). Could get some nice colors with paint or gelcoat. Electronics would be protected in the shell.
Normal fibreglass is fine for GPS etc, but I am not sure about carbon fibre. I would expect this would block reception so external aerials would be required if used. Epoxy resin would be stronger and lighter, but you cannot use gelcoat with epoxy.
@ Clusso99 -while recovering, i whipped up this saucer in sketchup quickly (~7.5hrs). I didn't radius around the ducts because sketchup was freaking out with the radius fillet pluggin -kept corrupting the surfaces. Just imagine them in place, something like 1-2" radius fillets. Small ventillation holes can be drilled at the inward facing part of the ducts near the top and bottom, the differential pressures will circulate fresh air throughout the shell. Alternatively, holes can be put inthe inward facing parts of the ducts below the props and exhaust holes can be drilled around the outer radius of the shell allowing for cross ventilation throughout the shell. Either technique uses slight pressure differentials in each duct to circulate air inside the shell and doesn't require additional tubbing.
Note the sky blue dual purpose motor struts/landing gear underneath, also the motor mounts (BLDC?) being made from fiberglass will deflect to the sides slightly to allow for different motor dia's to be mouned in the ducts. Prop ducts as pictured are 10", shell height in the middle is 4" with a total dia of 32". Motor mounts are triangulated for strength/weight balance and are part of the lower shell. If your planning to use outrunners, i can make the motor mounts a cup shape with an aluminum metal base glued to the top/bottom of the cups for the 4-way bolt pattern on the outrunners.
Electronics area's can fit 4 compartments of 6.25" x 6.25" x 2.5-3" and 1 compartment up to 6.5" x 6.5" x 4". Just guestimating on the weight.. single layer heavy strand /pollyester resin under 2lb I think. Definitely less if the gellcoat is left out.
Maybe tonight i can get a chance to play with a more square design in sketchup, will post if i come up with anything asthetic
RinksCustoms: Yes, thats the idea. Now with the LED strips around the edge to flash various patterns and some nice colors and a good video camera, I suspect a lot more UFOs will be sighted LOL. A PropUFO (TM).
BTW 2lbs sounds a bit heavy - i would think we could get away quite a bit lighter than this. Instead of paint or gelcoat, the colored heat film used on rc plane wings should work - lighter and able to use multiple colors.
If only they weren't so dangerous, laser leds could be used with good effects :shakehead: Just how difficult is it to project a small hologram underneath - a miniature green man LOL
Comments
I've used those motors in the past, and while they worked ok, I wasn't very happy with them. It's true that you get what you pay for, and most brushless motors are in the $50+ range. I burned out a couple, even though I was running them at less than their rated amperage. The 400XT motors I'm using actually work well even above their rated amperage (I've run them at 15a before). I also didn't like the fact that the shaft on them is so small, because it makes using "real" props harder. They work fine if you're using GWS props, but anything else won't have spacers to fit that small shaft.
The speed control in that package can't be set to fixed throttle end points, which is going to cause you a lot of problems. Unless otherwise stated, assume the ESC uses "auto-endpoints" (most do). Basically, the ESC uses the previously seen minimum and maximum throttle values to decide what it's input range is, which means the throttle values you're outputting are being scaled by a constantly changing factor, and they're not necessarily the same for each ESC. The ones I'm using can be programmed to use fixed throttle endpoints (I have them set to use from 1.1ms - 1.9ms). The "acro" series can be programmed with an external programming box, and have a bunch of options including endpoints, braking, acceleration, cell count, timing advance, and more. They're still freakishly cheap. [noparse]:)[/noparse]
I've attached a pic of my board. The stuff I've tinted yellow is left from a previous project - it's not connected, so ignore it. [noparse]:)[/noparse] I left the center ESC pins connected, since they're just regulators - They can be run in parallel without problems. I connected them to the 5v rail on the board, and the 3.3v regulator converts that to power the prop. The R/C receiver is connected to the 5v rail too. Nothing fancy required.
Jason
http://www.hobby-lobby.com/spin_box_programming_control_3082_prd1.htm?pSearchQueryId=753159
plus they have data logging built in and can be displayed on the SPIN Box in the field. BUT, I don't know if I can do $80.00 x 4 for a Quad.
Castle Creations ESC's are nice because they can be programmed via USB from the PC.
http://www.castlecreations.com/products/castle_link.html
·I only have a Phoenix-10 with UBS programing card right now for testing. But those would be $50.00 x 4.
http://www.hobby-lobby.com/phoenix_10a_esc_7147_prd1.htm
http://www.hobby-lobby.com/castle_link_programmer_7146_prd1.htm
Since you mentioned "Acro" I did a search for those. Wow! a nice USB programmable ESC with data logging for 28 bucks! Awsome!
http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=10940
I have got to spend more time on the HobbyKing site. Over the weekend I found one of my Hovercraft friends built a 10' craft and powered it with two of these:
http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=5142&Product_Name=Turnigy_80-100-B_130Kv_Brushless_Outrunner_(eq:_70-55)
(One for lift, one for thrust)
Two Li-Po's each and propelled a·100 pound craft and a 200 pound·man at 10 MPH for 8 minutes!
What wonderful tools we have to work with!
Jim
hover1: Thats a serious motor, but then again it's serious work that it's doing!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
1. Connect all the escs and motors to esc (but dont put the props on the motors) and connect the ESCs to the CPU
2. Power up the CPU without powering up the ESCs - so you need an extra power supply to do this.
3. Set the output to all the ESCs to max value - command from PC
4. Connect power to ESCs - you get a set of beeps
5. Set the output to all the ESCs to min value - command from PC
5. Get another set of beeps and ESC throttle range calibration is done.
Their PC config app walks you through the steps above and commands the software to go through the steps, so you dont need the transmitter to do the calibration.
Could you attach the schematic and source code?
Thanks,
Robert.
rc.runryder.com/helicopter/t598358p1/
Rich H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Simple Servo Tester, a kit from Gadget Gangster.
Its a work in progress, there is still lots of stuff that doesn't work but you can monitor/config the quad, capture sensor/imu/motors,etc and chart the output, etc.
The flight system uses a mix of gyros and imu, the gyros stabilize rapid changes and the imu handles more consistent tipping.
In the quadcopter code there are several options - I use the on-board imu - itg3200/adxl345/hmc6843/bmp085 , you can look at the bst project file for the defines I use. It contains code for several other options e.g. SF9 AHRS board, separate RC receiver inputs rather than PPM but most of these options I might have tested the driver but I will not have tried it in a quad.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
I managed to fly the quad into the ground yesterday (pilot error) and so needed to repair it today so I was adding a couple of upgrades - battery monitor and bluetooth while I was working on it.
I'm new to the forum and this thread, and also building my very own quadcopter :smilewinkgrin:!
You've all done a lot of great work so far:)!
I've still one question though: I cannot find out how to power the propeller board with just the BECs from the ESCs.
Can anyone explain me if it is possible and how to connect everything?
It would save a lot of weight if the propeller board doesn't need an extra power supply.
Many thanks in advance!
Jesse
A couple of examples, the prop backpack has an on board 5V to 3.3V regulator so 5V from ESC/BEC can be supplied directly to the board. A prop protoboard has a 7-12V to 5V regulator followed by a 5V to 3.3V regulator, so you can supply 5V from the BEC/ESC to the input to the 5V to 3.3C regulator bypassing the first regulator.
You will need a 3V3 regulator to reduce the supply to the prop. The prop will happily drive ESCs from 3V3.
I am experienced in vacuum bagging & regular molding of fiberglass/carbon parts for automotive, also in making custom molds for those parts. I could design and built a design using fiberglass or carbon body, like the quad copter of that austrailian company linked to earlier in this thread with fiberglass bodies using ducted fans starting from around $25/complete shell.. If the majority is interested, i can begin designing the plug(s) for the molds now, that is if you have settled on a final physical layout. It can be with 4 separate ducted fan and single main body or i can make the body one solid peice similar to what is pictured earlier in the thread.
Or i can do a small production run of maybe 30-40 body kits in exhange for the guts to make my own
Also, thoughts on an alternate landing idea? like using a hook to "catch" the quad copter? Might be a good idea for you Cluso, being on a sailboat most of the time. The hook idea, first experimented with the skunkworks back in the 60's with one of their more parculiar aircraft that took of with caster wheels on the tail & landed vertically on a hook near the nose of the craft.. the design didnt go very far, but the hook idea with a big wire loopmigh be a good idea for landing such a craft on a constanly rolling & pitching platform.. although you could probably use a scheme that uses 3 or 4 ping)))'s or maybe even a CMu cam to detect a pattern on the landing and track it in software to make yaw/roll adjustments on final. i wonder if the PIR distance detectors would work better for something like that. maybe 3-4 ping sensors would be fast enough to detect a moving landing platform?
Anyway, if you and the majority in this thread would be interested in working out a deal with fiberglass or carbon bodies, just PM me
P.S. -personally id rather do the small production run of 20-30 complete body units with or without the molds to make more bodies to whoever in exchange for a complete set of "guts" for one of these.
Another way to handle a motor loss is to have 6/8 motor, this also increases the payload capacity of the quad.
Thanks for your explanation!
I'll be going to use a USB Protoboard. My BECs can deliver 5V @ 2A each, but to reduce the stress on 1 BEC,
I want to use multiple BECs to deliver the current together, resulting in a lower current per BEC,
so I'll connect them parallel to the 5V input.
One other thing: Would the accelerometer be accurate enough to measure the minor translating movements of the quadcopter when hovering,
or is a gyro more useful, since it seems a bit overdone to take both gyro and accelerometer to have 3 axis,
instead of a single axis gyro (z-axis) and a 3-axis accelerometer. Anyone any experience with that?
My setup: (which is still shipping :mad:)
Motors: hexTronik DT750 Brushless Outrunner 750kv
ESCs: Mystery 20A Brushless Speed Controller (Blue Series)
Battery: Turnigy nano-tech 5000mah 3S 45~90C Lipo Pack
Props: 12x3.8 & 12x4.5 (CR & CCR pairs)
Mainboard: Parallax Proto USB Board
Gyro: Parallax Single Axis Gyroscope
Accelerometer: Parallax Tri-Axis Accelerometer
LEDs: 4x8 LED bars indication direction and warnings.
PING: 4x Ping Sensors to fly autonomously.
R/C Transmitter: KDS K-7x 2.4 GHz.
R/C Receiver: KDS K-8X 2.4 GHz.
Cam: Live feed cam controlled by 2 servos for tilt and rotate.
LCD Glasses: VGA Video glasses to view the live feed from the cam.
Diagonal span length: 60 cm (no props), 90 cm (with props)
Estimated weight: 1500-1700 grams
Estimated flight time: 10 minutes
In terms of acc/gyro - acc is too slow for stabilization, gyro is fast enough, however gyros drift. I use both - gyros possiblity for stablization and accelerometers to correct the gyro drift.
Gyros ony detect/correct angular changes and cheaper ones tend to drift more than good ones, accelerometers are good for detecting/correcting drift.. so yeah i would give my autonimous brain as much info as it needed to successfully navigate. Many here may not be pilots, but flying a plane is probably three times more intense than driving a car, since in a car your just a line follower with speed and periphial detectors.. where in a plane your periphial detectors are now dealing with a sphere instead of more or less just a band of horizon, you have to worry about other aircraft in the area, your altitude, your atitude, airspeed, heading, roll, trim, what the engine is doing, if you need carb heat or not, and communications to control towers when entering restricted airspace. Its about the most complex thing a human can do all at the same time. Your also using your senses, smell, sight, touch, hearing, as well as other senses like your internal sense of 3 axis G and a human version of a gyro for detecting rotational changes on 3 axis.
I say write good code to make sense of all the data you can throw at it and it will be a better autonimous pilot.
Ducted fan motors tend to be more expensive but there is no reason why a standard motor/prop would not work within the protected shell. IIRC that is what the new US quadcopter is using. I think those who are already flying can comment further about a fibreglass shell but I think it would be great. I too have some experience with fibreglass, having helped in the production of my catamaran.
For some time I have thought a 'flying saucer' type shell would look great - with leds around the outer edge. It could be made with two identical halves (top & bottom). The biggest problem is in what size props and the span. There are lots of variations here. We will require a standard for using a fibreglass shell, but this is a great opportunity.
Regarding electronics, I have all the parts (gyro, accelerometer, compass, gps, pressure, prop chip, etc). I have just not completed the pcb for production due to other work pressures.
1. The protoboard has connnection points for 4 servos in top right corner of the board. I put the 4 ESC connectors there.
2. The power connector for the 4 ESCs are connected together and is connected to the power switch, it then goes to a jumper just to the right of the poweer regulators, put the jumper from +5V to jumper, then when to move the power switch to position 2 (servo on position), 5V from the ESCs goes to power the board.
In terms of sensors, I use a 3 axis accelerometer, 3 axis gyro and 3 axis magnetometer. The 3 axis accelerometer and x/y axis gyro are fused using 2 kalman filters to get pitch and roll angles. Then the 3 axis magnetometer and 3 axis accelerometer are used to get a magnetic north bearing angle; that and the z axis gyro are fused with another kalman filter to give a yaw angle.
The flight system uses the x/y gyros directly for short term stablization and the fused pitch/roll/yaw angles for longer term stablizations. The gyros because you get a very fast input when the quad starts to rotate and the fused outputs give you an accurate orientation of the quad to keep it level.
Jason has found that the gyros alone are enough if you are manually flying the quad because you can manually correct for quad orientation issues. I would tend to agree though if you dont have much RC flying experience its hard to fly with just the gyros.
I also have a pressure sensor for altitude control though thats not tied into the flight system yet and a GPS for position control.
Generally the more automated you want the quad to fly the more sensors its going to need.
(The order wasn't shipped out yet, so I could still change it
@ Timmoore: When I downloaded your "quadbot1-bst-archive-100731-113807.zip", it seemed to be a broken file.
After repairing it, I still managed to read the files, but it seemed that some object files were not included or missing.
(For example the "itg3200objectLMM" as you described in you "quadbot1.spin" file.)
Do you still have these object files? It'll be very helpfull to get an idea on how to read the gyro outputs.
I was thinking of a single shell design, with an identical top & bottom. Maybe the overall dia of the shell would be ~30-36". This way the motors could be mounted on 4 aluminium "U" tubes, each mounted just across each opening. This makes the motor mounting simple.
Otherwise, the shell could have this feature inbuilt but would require a little extra fibreglass strengthening here - in fact probably the preferred method - saves cost and time. In fact if the fibreglass struts (i.e. the part remaining when the shell is cut to open the flute for the fan (airway) could be 90 degree opposed making a 4-way strut with a circle in the centre joining the 2 halves. This could be done by rotating the shell 90 degrees top to bottom, so 2 struts would be on the diameter line and 2 struts as a tangent of the circumference. I hope I have explained this - probably a pic would be better so I will try and do when time permits and post.
I don't see a need to rotate the arms (motors) or extend them.
A total shell like this would not require landing arms as the shell would be fine to land (on grass). Could get some nice colors with paint or gelcoat. Electronics would be protected in the shell.
Normal fibreglass is fine for GPS etc, but I am not sure about carbon fibre. I would expect this would block reception so external aerials would be required if used. Epoxy resin would be stronger and lighter, but you cannot use gelcoat with epoxy.
Does this make sense??
BTW the ITG3200 is what I have and will use.
Note the sky blue dual purpose motor struts/landing gear underneath, also the motor mounts (BLDC?) being made from fiberglass will deflect to the sides slightly to allow for different motor dia's to be mouned in the ducts. Prop ducts as pictured are 10", shell height in the middle is 4" with a total dia of 32". Motor mounts are triangulated for strength/weight balance and are part of the lower shell. If your planning to use outrunners, i can make the motor mounts a cup shape with an aluminum metal base glued to the top/bottom of the cups for the 4-way bolt pattern on the outrunners.
Electronics area's can fit 4 compartments of 6.25" x 6.25" x 2.5-3" and 1 compartment up to 6.5" x 6.5" x 4". Just guestimating on the weight.. single layer heavy strand /pollyester resin under 2lb I think. Definitely less if the gellcoat is left out.
Maybe tonight i can get a chance to play with a more square design in sketchup, will post if i come up with anything asthetic
Isn't the large surface area of such a design not going to make the thing a lot more susceptible to turbulence?
BTW 2lbs sounds a bit heavy - i would think we could get away quite a bit lighter than this. Instead of paint or gelcoat, the colored heat film used on rc plane wings should work - lighter and able to use multiple colors.
If only they weren't so dangerous, laser leds could be used with good effects :shakehead: Just how difficult is it to project a small hologram underneath - a miniature green man LOL