I have a couple of the Xracer V1 in some mini quads that work just fine.
They have certainly performed well.
Not sure if they have changed the X2.1 a lot.
Hi Marc and others…
We have made a strong attempt to improve our documentation, starting with the X2.1.
There are no known hardware issues.
See here:
it has CH and CL reversed on the CAN port 
it’s “amazing” that the wiring of ports is so varying, even for “standardized” ports … doesn’t help in establishing an ecosystem
I follow this board from begining, it´s looks nice for my builds but long time to go to market, very few experiences and no answer from developer many times I asked for news, I don´t know if I miss something too but I bought a pixracer just in case…
Olli: It is reversed on the X2.1 or the X2? Or both?
Good to see that Arduplane has worked for others.
X2.1, see the scheme in the mrobotics link given in the above
X2 has a non-functional CAN: UAVCAN for Hobbiests
How did the CAN remain reversed for the X2.1 and yet is correct for the Pixracer? How strange.
Olli – can this not be addressed by swapping around the CAN CH and CL wires on the CAN connector? Obviously not ideal. Maybe I am not understanding fully but I am not sure why you have had to do delicate soldering jobs on your v1 X2 boards rather than swapping the wires.
And given that I am looking for an extra small form factor board for a plane that has CAN functionality I think I may have to pass if I need to do delicate soldering on it. I could live with swapping CAN wires on the CAN connector though…
the X2 is not the X2.1
so what is true for the X2 might be irrelevant to the X2.1, and vice versa
X2.1: here you need to swap the wires in the CAN connector, as you say.
X2: here the PCB is routed incorrectly such that the CAN transceiver can’t work, and if one wants to correct for that one needs to do some delicate soldering. Since this is not something many would do, it effectively makes the CAN on the X2 non-functional.
the X2 thing is irrelevant to the X2.1, and hence OT. I just mentioned it in response to your question, thinking that it may help clarifying the situation. Sorry if it rather added to the confusion.
for the X2.1 holds (and that’s all I wanted to bring to attention):
it has CH and CL reversed on the CAN port
=> needs to be addressed by swapping around the CAN CH and CL wires on the CAN connector
(unless you connect it to a CAN device with the same reversed layout LOL ;))
EDIT: it’s kind of sad since, if you read e.g. Mike Kelly’s posts on UAVCAN, one of the appealing aspects to him is the simplified standardized connection … this practical advantage is a bit broken into pieces
EDIT2: paragraphs 2-4 here: UAVCAN: CANbus for the rest of us
All good. Understood now.
It is not a deal-breaker to have to swap wires, for me anyway, I suppose. Still if we start using CAN for peripherals (and I hope we do) then basically we will need standardization, as you say.
I will say I still like my older X2s. Even though they have had some minor issues, but all of which are still working for me. One of those issues, only relevant now, is a useless CAN port since I now will be playing with CAN peripherals.
Based on some positive feedback here, I may well order one.
I ordered one of these. At $90 including soldered headers, shipping and cables, they have reduced the price by about $20. It seems a better option than a Pix Cube (good but big and pricey) or a Pixhawk 1.
For plane users having the USB by default on flexible cable is a plus.
Are there any recommended STLs for a case?
1 Like
Nice Marc, I follow your experience with this board
I will post my progress. I think $79 is a good price for these boards.
1 Like
hello, olliw42, I’m not an expert , but I just plug my Zubax gps to the CAN port of the Mrobotics X2.1 and it works without swapping anything .
I’m not sure what you expect me to reply now.
I’m not an expert either, and just look at the documentation provided, that is the connection diagram here https://github.com/mRoboticsIO/Hardware/wiki/X2.1 and the electrical scheme here https://github.com/mRoboticsIO/Hardware/blob/master/X2.1/Docs/x21_V2_schematic.pdf. According to both, CL and CH are swapped with respect to e.g. pixracer and P2.1, and what appears to be the standard.
I did not want to deny your remark about CL and CH swapped.
I just wanted to point out that using the board with the provided cable , I did not have to do anything special to have the Zubax Gps to work with the 2.1
Well, ordered one, along with one of their GPS’.
The GPS seems solid. Locks on super quickly, even indoors.
Form factor for the X2.1 is good. It’s very small. I like the USB extension, but will miss the a separate I2C port.
There’s an option for buying a separate I2C splitter with a GPS pass-through, but it’s $29.
However, the biggest thing I missed, is that you cannot power the X2.1 from the servo power rail. This means that with any kind of power failure/brown-out, you’re screwed.
1 Like
Anyone flying the X2.1 on a racing-class copter? I’m curious if it works as well as PixRacer in that environment.
You would have to direct solder the wires. If you use the headers, it actually ends up taking a significant amount of space. You might get away with on a full 250 sized frame, with vertical headers, though.
Sure. I spend a lot of time with racing drones, and soldering straight to pads is pretty much the norm. I haven’t used a header on any racing controllers for 3 or 4 years. I actually remove the RC-out and WiFi headers on PixRacer, as they are too bulky for my build, and really don’t serve any useful function for me. Guess the WiFi would if I used WiFi, but I don’t. Fortunately they provide an option to get the X2.1 without headers installed, which I guess is one of the things that interests me. We’re building these systems in mass, and it’s a bit of labor to remove the headers from PixRacer. I have no complaints on the performance of PixRacer though.
My biggest concern would be sensitivity to vibrations. The PixRacer does really well. I have mine almost hard-mounted to my frame, just using elastomeric male/female spacers from the frame to the PR, and flight logs show a really clean vibration environment for the controller. This is on a stretched frame for 7" props, which tend to be more vibration prone than smaller 5" builds. In fact, the same frame with a generic F4 racing controller cannot fly in stabilized mode, as there is too much noise for the accelerometers. The X2.1 uses almost the same IMU models as the PixRacer, so I suspect it would be similar, but I was just curious if anyone had actually done it and had some feedback. Maybe I’ll be the guinea pig.