Can someone tell me the differences between a Pixhawk and the Pixhawk 2 that 3DR is marketing as being supplied with the Solo?
I've tried searching, but only found vague references to Pixhawk 2.
Even the 3DR web site is very light with info about Pixhawk 2.
Perhaps the info is being kept quiet until the final release? If so, that's ok, just let us know.
Thanks!
Rich
Replies
Jake,
Nobody kicked me off DIYDrones. I deleted mostly everything I had posted on DIYDrones. I was tired of posting helpful information (i.e., doing the legwork) on DIYDrones and not seeing that it was having an effect. I am busy with other projects, and the fun of multirotors/quadcopters is gone for me (DJI [the Chinese] has taken over). I felt I was posting too many negative comments and not really helping the community. In addition, I'm working on my own UAV company, Thomas Butler Technology, in Rockford, Illinois; cannot be giving help to competitors!
The fact the companies are STILL farming production to China irks me too. Even the PCB engineering is being done outside the US (of the 100,000 American electrical engineers, is there not ONE that can design a simple UAV flight controller? Not naming names but they know who they are. Were it not for the greed of companies like Intel and Apple, the technology in China would not be there and many more people in the US would be employed.
Note that the engineering analysis I did of the PixHawk2 has only been reviewed in the past few days since I came back on DIYDrones (my presence must have triggered a red flag...ha!) and this is what AUGUST 2015! I believe the first post related to PixHawk2 showed up in Novemebr 2014! Yes I deleted the analysis posts. It looked to me like I realty didn't need to be wasting my time doing on other peoples work. They could and should do their own R&D. The PixHawk makers have plenty of R&D money to do it themselves.
In addition, the switch away from Eagle CAD was an obvious attempt at fostering "barrier to entry"; now Altium ($$$) CAD is needed to work with the schematics of the PixHawk2. There is/was no technical reason to not use Eagle.
After looking at the response which only showed up this week, they seem to be on the right track. The response is what I expected, and is positive. One comment was on leaving the hard mounted sensors in place for R&D purposes. This is a silly waste of money for most customers, but it does give a (someday, probably, never day) possibility of a vibration analyzer program being run on a user's PixHawk. However, one could easily connect and hard mounted IMU and get the same data
I'm going to try to post only helpful information and try not to be critical of peoples wild ideas...how about that guy doing acrobatics on America's Got Talent at the end of two hoses attached to a jet ski! Wild innovation!
Back to the salt mine!
Does Thomas Butler Technology has a website?
Chris Anderson just pointed me to this thread.
My name is Philip Rowse, I am the lead Systems Engineer for 3DRobotics, and Hardware lead on the Pixhawk 2
All the files for Pixhawk 2 are available @ https://github.com/3drobotics/Pixhawk_OS_Hardware
In this folder you will also find instructions on how to download and install Altium, to view and search through the files.
I just had a look through Thomas's list, and will point out a few things...
1. We LOVE invensence :) there is 0 desire to get away from them. Their sensors rock!
2. The MS5611 is THE barometer of choice, as its accuracy is predictable.
3. The reason for a Baro on the baseboard and the IMU is for us to be able to study the effects of mounting, heating, and isolation strategies.
4. Isolation... Vibration is only 1 reason for isolating the IMU.
Temperature control, and mechanical strain are the most important.
5. Why the F4 rather than the F7? The F7 looks nice... But ST screwed up. The F7 is only available in a 1M flash size! If you look at both the Copter code and the PX4 flight stack, you can see that is a big concern moving forward.
6. The Flex IMU is attached to the FMU via a locking connector, and is removable, in fact, we have a few Devs planning on flying copters with just the FMU, and no External IMU... Nice and flexible.
7. The 80 pin... This is where controversy starts... And ends...
Trying to keep everyone happy on how connectors should work... Well, that's a loosing battle.
Pixhawk 1 was a connector farm. It allowed for everything, and as such needed to be massive.
With CAN ESC's just around the corner, we will soon see a day where a single 4 pin plug would be enough to get you off the ground...
Then there are OEM applications... Solo being a prime example... Where the new mounting method just makes sense.
For customers, we are looking at two options to start with, a full sized Pro, with more connectors than Pixhawk 1, and a mini, which still will get an Octo in the air, and will give a great idea as to the possibilities with future custom carrier boards!
To top that off, we have already sent the drawings for the carrier boards to a HK hobby supplier, to encourage them to build carrier boards to suit their airframes, rovers, boats etc...
Within the next couple of months I will be releasing the carrier board drawings in this same folder.
Oh, why Altium? Eagle is not suitable for doing 8 layer boards, and it's design rule functions are not as advanced as Eagle. However, I do intend on making an Eagle version of the carrier board reference design, as the carrier is quite simple for anyone to make. Don't be scared by the 80 pin connector, it is quite simple to solder in.
Any questions? PM, Drones Discuss, mumble... Raise an issue on the Git Repo where I put my drawings! That is my live working directory, that's where you will see future versions, and any mods...
For Solo we are currently building Rev D FMU, and Rev F IMU, rev E IM U is my experimental branch.
As far as this being a repackage... This is the beginning of a family of products.. Think about being able to update to the latest autopilot, without having to do any more than swap out a module... And for those into schematics... Can someone tell me, what are the inherent benefits of separating out the Power selection module from the Autopilot? Why might we have done this?
I hope this clears some stuff up.
And I would love to add, Thankyou Thomas for looking into this! It is appreciated and all of your points will be reviewed by the team.
There's no reason for a 8-layer board. This appears to be done specifically so that you can't work on the design with lower cost programs, and can't get the boards made at a reasonable cost unless you have manufacturing connections.
You also talk about "modules" which is just code for disposable, non-repairable, custom stuff you have to buy from one supplier.
It also seems like now the community is one step from being out of the loop. Clone makers in China now get the first look at schematics? WTFF??
It's always been BS that they don't let the community have input on the hardware, then turn out buggy hardware that often get copied by the clone makers. Now we're ASSURED that bugs caused by your hubris will be fully locked in before we even get a chance to see them and point them out.
Yes I'd does the boring PPM...
PPM or S.Bus directly into Pixhawk is our preferred method, we don't recommend a PWM RX to PPM encoder.
The second thing it does, is for your plane... It allows a genuine manual pass through... PPM in, and PWM to servos out. This allows us to have a "get out of jail" option in case of an FMU failure, or a total power system failure. The IO chip will remain alive even if you have the servo rail powered at 10V, well above the voltage that will shut down the FMU.
This however is of minor use on a copter....
Nice explanation Philip!
Thanks.
Would you mind to ask when Pixhawk 2 is going to be available to be purchased standalone and which Rev IMU is going to be used in that?
What goes in the Christmas version will depend on feedback from this first group, as well as Solo users. every log from a solo owner is very useful in this process
-
5
-
6
-
7
-
8
-
9
of 15 Next