Yes there is.
Yes there is.
Looks awesome. Is there an ETA on availability in the store?
why don't we use some high end sensor like something from AD, you guys probably know the very expensive WKM autopilot system,which cost about $1000, the reason they selling it so expensively is because it use some high accuracy sensors to dramatically slow down the drafting speed, it can even perform an accurate position hold without GPS. Despite it has such high performance and high price, all the IMU sensors it use cost less than $150 :DJI Wookong-M mini-technical-review
Looks really nice Mikro, but I'm missing the LED CI... is it gone?
That looks great.
p0lar == the aeroquadist formerly known as dpackham...
If I can throw in my own 2 bits for shield layout.
I cant say anything about the wiring realities of this stuff - I just know that Im obsessed with killing the really ugly mound of spaghetti we have on our quads... So the following is offered in that perspective.
I would like the board - if you are going to redesign it anyway - to be more suitable to being served by right-angle headers and letting the wiring come off the side.
I dont know what the F#@k im doing in Eagle, but I sketched it out kinda...
The current board suffers from feature creep. If you want heading / alt hold, but otherwise just want a quad that flies - there is way too much there.
For example... Does anybody actually use the prototyping areas? How about usage of the pads? I think your idea of offering those things on daughter boards is a good one. (else why would you put the stackable headers on the thing... We can use a bit of perfboard if we really are inclined.
I am sure that you can make wise judgements about other features as well - I think the LED driver was good to remove also. Focus on what is useful to 80% and let those who want to tweak things figure out how to do that. You know they will anyway...
I'd keep the OSD and the GPS...
Regardless, getting to the point at hand...
I'd avoid overhanging the USB port altoghether - just move the shield down a bit. And then make it so that you can wire your receiver and each ESC off the side. Motors 1 and 5 at the front left, 3 and 7 at the front right, etc.
That will keep all of our wire runs a lot cleaner and allow a lower profile for the board in all.
Anyway.. those are my suggestions... hope it is useful
I'd also like to have the reciever pins in a place where I can drop a ribbon cable off the side to connect the signal pins.
Death to the flying spaghetti monster.
"Don't gain the world and lose your soul, wisdom is better than silver or gold." - Bob Marley
Hi My Name is Franco, I am a multirotor Addict.
see my fleet in my Mad Scientists Lab at http://lab.westernwillow.com
My Latests craziness: Flying into wildfire: http://lab.westernwillow.com/video_list/260
You have some good ideas there Ucdwino. As you know we need to be careful to avoid the USB connectors which is the main reason why the ESC and receiver connections are oriented the way they currently are. For the v2.1 layout I took out I2C ESC connections and the LED's.
Unfortunately, it takes a bit of work to layout the wiring connections, I'd have to check if your proposed layout would work. Sensor sensor comm is so critical, I did my best to shorten the I2C bus comm between the Arduino and the sensors (which is why the level converter and sensors are located where they are). Also I tried to keep common sensor orientations between all the boards to remove yet another layer of complexity.
The new 9DOF sensor is long, so it will require some space on the shield, but will make it simpler for users as it's just installing one board instead of three. The prototyping area was actually just empty space, so I put it there for user convenience. At least in the beginnings of the AeroQuad project, many people liked to be able to easily customize what they want, so I stayed with that philosophy.
The good thing with the shield is that the end user just connects what they want. For example, you don't have to solder up / connect up all the power/ground pins for the transmitter, which would be nearly equivalent to your layout above (obviously the orientation is still different though).
Also, I'm a bit surpised no one asked about it yet, but there are some extra I2C connections near the 9DOF sensor location. They fit the FreeIMU and am planning to work with those guys to produce a lower cost alternative to the Sparkfun IMU boards. But who knows, there is an upcoming MPU6000 from InvenSense which is supposed to have all the IMU sensors together WITH an angle estimation algorithm built in, so all that might change. As you guys know, I like sticking with the shield concept to be flexible for the future.
I've asked Arduino for the pinouts of the upcoming Arduino Due, they had indicated they would, but I haven't received anything yet. I suspect they are still just trying to finalize it first, before doing any further release of schematics/data. I'll take your board considerations when making a potential v3 shield if things need to change drastically. Thanks!
FWIW, I'm not sure the I2C line length is all that critical, it's max clock speed is on 400 kHz. Also if you need the space for other layout orientations, the level converter isn't strictly necessary. I'm running my freeIMU without it. All is good as long as the Arduino's internal pullup's are left disabled.
Hey guys, just test flew the v2.1 shield and all is well... looks like we have a new family member to the AeroQuad shields. I'll be packaging these guys up and putting them on the store as a kit (including sensors and Arduino Mega).
Will bug jihlein to help me fix the code for the new HMC5883L... the code I have right now seems to have some axis backwards or something.