AeroQuad Forums - Blogs http://aeroquad.com/blog.php The AeroQuad is an open-source hardware and software project dedicated to the construction of remote controlled four-rotor helicopters, also known as quadcopters or quadrocopters. en Thu, 17 May 2012 22:50:41 GMT vBulletin 60 http://aeroquad.com/images/misc/rss.jpg AeroQuad Forums - Blogs http://aeroquad.com/blog.php Y6 - Why not? http://aeroquad.com/entry.php?265-Y6-Why-not Mon, 23 Apr 2012 21:46:19 GMT

Name: IMG_1671.jpg Views: 661 Size: 96.5 KB

Name: IMG_1673.jpg Views: 661 Size: 94.9 KB

Name: IMG_1674.jpg Views: 662 Size: 98.4 KB

Name: IMG_1675.jpg Views: 661 Size: 95.9 KB

Name: IMG_1676.jpg Views: 659 Size: 97.4 KB
Attached Images
]]> q6600 http://aeroquad.com/entry.php?265-Y6-Why-not Hacking the Thinkgeek Portal 2 Plush Turret - Part 1 http://aeroquad.com/entry.php?264-Hacking-the-Thinkgeek-Portal-2-Plush-Turret-Part-1 Sun, 08 Apr 2012 13:54:16 GMT Hi all, First up, this post is totally unrelated to AeroQuad - I just had nowhere better to post it and figured some people on here may be ... Hi all,

First up, this post is totally unrelated to AeroQuad - I just had nowhere better to post it and figured some people on here may be interested. Cheers to Mikro for letting me use this blog area for an unrelated project - hopefully I won't spam the AQ site too often!

I recently purchased a Portal 2 Plush Turret from Thinkgeek. I'd been wanting a talking plush turret ever since I saw Jonathon Guberman and Leigh Nunan's amazing homemade plush turret. This one used an Arduino + audio shield as the main micro and for sound, a tilt sensor for detecting when the turret is knocked over, a small button on the bottom of one leg for detecting when the turret is picked up and a PIR sensor for detecting people in front of the turret. You can see in the video posted by Jonathon Guberman (on the page linked above) that it worked very well - the turret could detect being picked up/knocked over reliably and it made a good range of sounds. However, once I received the official plush turret from Thinkgeek I found that it wasn't as good as the homemade one. It wouldn't reliably detect when it was picked up or tipped over. It couldn't tell the difference between being picked up and tipped over, so the things it said just didn't make sense half of the time. It made an annoying machine gun noise (it's a plush toy, it doesn't have a machine gun so shouldn't make machine gun noises imo!). And finally, it never said some of my favourites from the game ("no hard feelings" when you knock it over, "please put me down" when you pick it up). So I went to work trying to fix it.

To begin with, I had a look at the existing electronics in the hope it might be reprogrammable - a long shot, but I'd have to take it apart to make new electronics for it anyway. I believe the microcontroller is one of the 'AM4EE Series' produced by Alpha Microelectronics Corp. The micro is covered in epoxy so I don't know which model it is, but the names of various GPIO ports and components are silkscreened onto the board and they match the naming conventions used with the AM4EE Series. Looks like these devices are one-time programmable and I cannot get to the device, so I abandoned the idea of reprogramming the existing electronics.

Instead, I designed new electronics. The new electronics have been placed on a PCB of exactly the same dimensions as the existing PCB, so it's a drop-in replacement. They also use the existing light dependent resistor (LDR) mounted in the front of the turret and the existing red LED mounted as the 'eye' of the turret. I didn't want to change the LDR to a PIR sensor as it'd require pulling the turret apart to get at the front sensor (the electronics are in a plastic box at the base of the turret) and cutting a larger hole in the front of the turret - don't think I could do a good job of it. The new electronics don't use the vibration motor in the turret as I think it's a big waste of power.



Some features of the electronics I've designed:
- STM32L152CBT6 micro (LQFP48). Chose this mostly for the DAC with DMA support allowing a high output rate, and ChibiOS support - I've been meaning to learn how to use an RTOS for ages now!
- Freescale MMA8452Q accelerometer. This is for sensing if the turret is still and upright, moving (ie picked up) or tipped over. I chose it as it was cheap, small and has a convenient 'motion sense' interrupt.
- Buffer amplifier and active low-pass filter for the voltage divider used to measure the LDR. Just a single filter pole at the moment.
- microSD slot. This is for storing audio files.
- TI TPA6204A1DRB audio amplifier. This is a fully differential amplifier, so should hopefully be more resistant to the SPI and I2C signals nearby. Its inputs are connected to the two DAC channels on the STM32. It can be used in single-ended mode by pulling a DAC pin to ground - this would simplify the software I'd imagine.
- Header for nRF24L01 2.4GHz radio module. Compatible with this radio. The radio has a high enough data rate to stream audio to the turret in real time, though I suspect it'd be challenging to implement the software to stream audio in real time. I intend to use it to have multiple turrets networked together, so as to all trigger when any one of them detects motion. Or to have the turrets play the 'turret opera'.
- 2x SMD LEDs connected to GPIO pins. The red LED on the front of the turret (the 'eye') is also connected to a timer with a PWM output, so it can fade off when the turret 'dies'.

The micro will be programmed using the factory programmed serial bootloader via USART1. USART1 and the BOOT0 pin are broken out to 0.1" header pins so as to allow serial communications and programming. There's no JTAG/SWD as I couldn't fit a header or route to some pads on the edge of the board. There's no USB as I could not fit a receptacle (and there's no USB-DFU bootloader on the STM32L1 anyway - it'd just save an external USB<->UART converter).

Cost for PCBs is ~$NZ8 each and cost for components is ~$NZ28 (in single quantity). Current schematic and PCB layout are attached as PDFs - don't have Eagle files sorry. Does anyone have ideas for extra features or can see any stuff ups? Will probably get some of these PCBs soon and get started on the software.

Cheers,
Ewan
Attached Images
]]> alamo http://aeroquad.com/entry.php?264-Hacking-the-Thinkgeek-Portal-2-Plush-Turret-Part-1 Unmanned Search and Rescue Quadrocopter http://aeroquad.com/entry.php?263-Unmanned-Search-and-Rescue-Quadrocopter Mon, 02 Apr 2012 05:37:15 GMT Hello All!

This is a University project build. We're designing a quadrocopter to aid in search and rescue operations.

Here is a picture of the balsa wood truss frame.

Name: Chris + Dave 018.jpg Views: 1322 Size: 92.4 KB

Here is a picture of the completion of Phase One.

Name: 2012-03-28 17.24.55.jpg Views: 1320 Size: 95.3 KB

Here is a video of our first flight test. Just a simple serial command to test everything out.



More to come.

Cheers!
Dave ]]> DavidPenny http://aeroquad.com/entry.php?263-Unmanned-Search-and-Rescue-Quadrocopter First Flight http://aeroquad.com/entry.php?261-First-Flight Wed, 22 Feb 2012 00:02:40 GMT So I read all the posts about not trying to fly your quad in your garage and I promptly ignored them. I mean hey, it's winter in Seattle and if I wait for weather good enough to do it outdoors I could be waiting a while.

Here's the video of my first "flight."




And my second "flight."

After I put away the camera I to give it one more try. That ended up hitting a column in my garage and ended my playing for the evening:






After my couple of bumpy "landings" my frame isn't quite as straight as is was an hour ago.


I suspect I might be replacing those arm plates a few times.

Now I'm just waiting for good weather so I can fly this thing outdoors. ]]> xtrmtrk http://aeroquad.com/entry.php?261-First-Flight Debugging http://aeroquad.com/entry.php?260-Debugging Tue, 21 Feb 2012 23:04:46 GMT At this point I thought everything was ready to go so I hooked the serial cable up to the Configurator and started Calibrating everything. For the... At this point I thought everything was ready to go so I hooked the serial cable up to the Configurator and started Calibrating everything. For the life of me though I couldn't get the ESCs to calibrate. If I used the Configurator's built-in routine 2 ESCs responded to low-end speeds only while 2 others only had high-end. If I used the instructions in the ESC to calibrate them I knew I was missing some of the control range of the AQ software. And they also wouldn't spin when using the Configurator's Test speed in the Motor Commands section. I was a bit stumped and must have tried doing both calibrations a dozen times or more.

While I was trying to do the calibration, my motors #1 and #2 stopped working. After a lot of debugging and experimenting and chatting in the forums, kha figured out that I'd blown a couple FETs (which control the signal from the AQ to the ESCs) on the Mega. (Thanks again kha, I owe you a beer ;D) It turns out you don't want to cross the PWM signal lines with 5 volts. It's a $70 mistake.

One thing that made debugging a little easier was making a Y jumper so I could see the signals going from the AQ board to the ESCs on a scope. That little side-tail made getting the probes on a signal a lot easier. And the signal looked very different when it had the ESCs attached. In fact I think that's what gave kha the clue he needed to tell me the problem.


I also ended up having to take the shield off several times. My shield fit VERY snuggly to the Arduino and the best way I found to separate them was with a wooden spoon handle. I just went around and around and very slowly and very carefully pried the two pc boards apart.


While I was trying to figure out what was wrong with my 2 motors controls I wrote some code (posted in the forums) to directly control the speed of motors without having to use the Configurator. Once I got my new Mega my little hack made it easy to calibrate the ESCs and I had them all synchronized in about 2 minutes.

At this point, I was read to fly. ]]> xtrmtrk http://aeroquad.com/entry.php?260-Debugging Adding the Electronics http://aeroquad.com/entry.php?259-Adding-the-Electronics Tue, 21 Feb 2012 22:42:31 GMT After adding the ESCs and threading the wires through the arms and up through the frame, you have a pile of spaghetti:... After adding the ESCs and threading the wires through the arms and up through the frame, you have a pile of spaghetti:


It's really worth taking some time to figure this all out. When I started it looked like there was a lot of space available for everything. It turns out you fill it up pretty quickly.

I recommend not putting the battery plate on until you've got all of the motor and ESC wires figured out. I ended up pulling the slack for the motor wires down below the arms to get it out of the way of the ESC plate.

All of this makes integrated ESC power distribution plates like this look pretty inviting: http://wayne.klinkerstein.m-faq.de/Q...2/BILD0281.jpg

The battery plate has foam pads and velcro to keep the dangerous LIPO comfortable.


Next up was mounting my Mega to the Flight Controller board. My kit was missing the screws needed to screw it down but Mikro gave me the specs on them and my friends at Tacoma Screw gave 8 of them to me gratis. The plastic spacers are threaded. Nice touch.


Everything mounted up and wired.



I know there are a few people on the forums confused about how the wiring should be done. It should be clear here that the only wires going into the receiver are from AQ shield and the ESC's BEC lines go to the shield's motor control pins. One of the jumpers going from the receiver to the shield has to have all 3 lines to carry power to the receiver. The others only need the signal line. I made those single line yellow jumpers but later replaced them with 3 wire jumpers because they held a lot more firmly.

At this point I decided to add the legs. It was good to wait on them because it made working on the electronics easier.
The landing legs are pretty cool - you fold them from pieces of the laminate material that have the aluminum on only one side along the scoring lines.









A very cool idea.

At this point I put the top on and thought I was about finished. I put the props on for the picture
]]> xtrmtrk http://aeroquad.com/entry.php?259-Adding-the-Electronics Assembling the Frame http://aeroquad.com/entry.php?258-Assembling-the-Frame Tue, 21 Feb 2012 22:17:10 GMT This part I really enjoyed. The first thing you have to do is insert the anti-vibration grommets into the arm plate. But deciding where they go... This part I really enjoyed.

The first thing you have to do is insert the anti-vibration grommets into the arm plate. But deciding where they go is a bit confusing. Included in the kit is a diagram showing the different arm configurations. I wanted to build a "+" configuration but if you do that according to the diagram it's not clear where the grommets go (and thus where you'll attach the arm plates to the rest of the frame). I decided to go with the "x" configuration for the plate knowing I could rotate the ESC plate above it to get everything back in the "+" layout. Kind of confusing.


Inserting the grommets is probably the hardest part of the build. I didn't want to damage them but they're a PITA to get in.


I ended up using a mechanical pencil with no lead sticking out to push it in.


At this point I thought I had the arm plates complete. Later I found the arms were pretty wobbly. Thanks to some posts in the forum I discovered my stupid mistake - I left 2 bolts out of each arm.


It's pretty cool how the arm plate vibration is isolated from the rest of the frame using the soft rubber grommets. Afterwards Mikro suggested I remove the extra black rubber washer in there.


Note - this is being viewed upside down to have a clear view of the grommets.

I like how all of the nuts are the nylon insert lock nuts.

Next up was assembling the motor mounts. All of the machining of the parts was really great in the kit except for one part of one motor mount which had kind of sloppy bolt holes. Totally usable but the only blemish in a really great kit.


That mount must have been jinxed because in the same bag was a bent bolt:

I'm wasn't worried about it though it was totally functional and my first good crash will probably bend a whole lot more.


A finished mount:


Threadlock was used on all of the motor screws


It was time to attach the motors to the arms. Two things though made me nervous though. I didn't like the wires wrapped around the metal edge and I didn't like the idea of the wire connectors bouncing around inside a conductive metal rod. So I taped up the connections and put some clear shrink wrap around the wires where they go around the bend of the mount: ]]> xtrmtrk http://aeroquad.com/entry.php?258-Assembling-the-Frame Unboxing http://aeroquad.com/entry.php?257-Unboxing Tue, 21 Feb 2012 21:48:52 GMT Yay! When I saw on the tracking that my frame was going to arrive on Thursday, I took the day off work to play. It finally arrived:... Yay!

When I saw on the tracking that my frame was going to arrive on Thursday, I took the day off work to play.

It finally arrived:


Opened - I bought a few spare props. Now I know I'll need a lot more.




Everything laid out:


The first thing I noticed was the construction of the frame. The plates are made out of an aluminum laminate. There's a layer of black plastic between the two layers of aluminum.



There are two thickness used, one thicker for the arm, top, and bottom plates, the other thiner for the other 3 middle plates.
]]> xtrmtrk http://aeroquad.com/entry.php?257-Unboxing LIPOs are Scary http://aeroquad.com/entry.php?256-LIPOs-are-Scary Tue, 21 Feb 2012 21:42:43 GMT If you read much about LIPO batteries online pretty soon you'll decide you never want them in your house or in your car. Seems like they're always blowing up.

I learned just how powerful they are making a very stupid mistake. The Sky LIPO batteries I bought come with banana jacks. I needed to switch them to EC3s to match the AeroQuad harness. So I removed the plastic housing from the leads and laid it down on my desk with the bare bananas just sitting there. A few seconds later I knocked the battery accidentally and the leads crossed and arced. There was a loud noise a bright blue light and my smoke alarm went off. I wasn't quite sure what to do thinking the battery might explode at any second so I grabbed a pencil and separated the leads.

Here's a picture of the aftermath. I work on a cork surface and it still has the burn marks.



I'm not sure if that battery is safe to use. I hooked it up to my kind-of-smart balancing charger and it takes a charge OK. Regardless, I bought one of these to store the LIPOs in. Too many stories online of batteries spontaneously combusting in people's houses and cars.

]]> xtrmtrk http://aeroquad.com/entry.php?256-LIPOs-are-Scary Waiting http://aeroquad.com/entry.php?255-Waiting Tue, 21 Feb 2012 21:27:56 GMT When I first decided I wanted to build a quad Mikro was out of them in the AeroQuad store so I ordered everything except the frame and started... When I first decided I wanted to build a quad Mikro was out of them in the AeroQuad store so I ordered everything except the frame and started assembling.

Here's the board after I completed it:



Two gotchas: 1) I learned later that I really wanted to measure the actual resistance on those two resistors I soldered to the board. I had to unsolder them, measure them, and put the back on. 2) I put the headers for the ICSP connection on upside down. I see a couple people have done that before me. I thought about just leaving it as it was but then found that an XBee uses that for power. I knew eventually I'd want an XBee to collect telemetry info so I fixed it.

Here's an up-close pic of my resoldering jobs. Ugly.






I'm not convinced the foam over the barometer is very effective. When I just wave my hand over it the barometer responds, even with the foam covering. I've heard of people putting tubes on them.... ]]> xtrmtrk http://aeroquad.com/entry.php?255-Waiting Building my first quad - a Cyclone 2 http://aeroquad.com/entry.php?254-Building-my-first-quad-a-Cyclone-2 Tue, 21 Feb 2012 21:16:29 GMT OK, as promised I'm going to document my unboxing and build of my new Cyclone 2. It wasn't quite as straightforward as I thought it would be but it was and is a ton of fun.

I'm not sure if I put all of this in a single post or break it up. I think I'll break it up to make it easier to edit. ]]> xtrmtrk http://aeroquad.com/entry.php?254-Building-my-first-quad-a-Cyclone-2 MiniQuad build http://aeroquad.com/entry.php?253-MiniQuad-build Sun, 12 Feb 2012 04:24:40 GMT Follow along with my build of a small quadcopter (8" motor to motor distance) here! Constructing these things is almost as fun as flying them, but all the waiting for items to ship, often from Hong Kong, is excruciating! Oh well, everything has shipped now so it shouldn't be too long before miniQuad hits the skies! ]]> wooden http://aeroquad.com/entry.php?253-MiniQuad-build Pics - As promised http://aeroquad.com/entry.php?252-Pics-As-promised Sun, 12 Feb 2012 00:45:53 GMT Here are photos of my first frame and quad build. To save space, I won't embed the images in this post, but you can see them here. ]]> sciman3 http://aeroquad.com/entry.php?252-Pics-As-promised Flight! :D http://aeroquad.com/entry.php?251-Flight!-D Thu, 09 Feb 2012 02:28:37 GMT Last night and today I took my first ever flights with my new quad. Being winter outside, the ground is rather hard, so it is far from ideal to be... Last night and today I took my first ever flights with my new quad. Being winter outside, the ground is rather hard, so it is far from ideal to be learning on. I sadly don't have any daytime flight footage, and I cant get any until my new props arrive (the ground is really, really hard :P). Flying this quad is turning out to be a lot harder than I thought it would be, but I have managed to get a few longer (10-15 sec) hovers. I think I still have quite a lot of PID tuning to do, as it is currently quite unstable. I may wait till it warms up, I don't have the money to chug through too many blades :P

I will post some pics of my quad once I take some. ]]> sciman3 http://aeroquad.com/entry.php?251-Flight!-D <![CDATA[I couldn't wait any longer....]]> http://aeroquad.com/entry.php?248-I-couldn-t-wait-any-longer Fri, 06 Jan 2012 08:17:54 GMT I just couldn't stand having the parts, but seeing them work yet, so I through together a quick rig, to give one motor and esc a test run. Although perhaps not terribly interesting, here is a video of my test: http://vimeo.com/34649657. Although moving the plank of wood was cool, I will remember to be very carefull when I have a prop on my motor!! ]]> sciman3 http://aeroquad.com/entry.php?248-I-couldn-t-wait-any-longer