Thanks for the input. I was reading some summaries about socketcan. I think it is a good basis for the project. I've not much knowledge about this piece of software yet, but i think the timestamps are not integrated because of the (quite expensive) PAEK Systems hardware. I think they are added through a PC-internal system clock. But it would be great if somebody can confirm that assumption.jannis wrote:When using an USB-based CAN-adapter (from PEAK Systems), the device logs a timestamp for every received frame and that is transfered to the userspace-software via the socketcan network-interface.
It's like I said: the hardware makes the timestamp. Of course it's not an absolute one since the USB-CAN-adapter doesn't know the current time but it's a relative and high-precise one. You can take a look at their driver's code if you want. And the timestamping is one of the reasons why the PEAK-adapters aren't really cheap.Basti wrote:Thanks for the input. I was reading some summaries about socketcan. I think it is a good basis for the project. I've not much knowledge about this piece of software yet, but i think the timestamps are not integrated because of the (quite expensive) PAEK Systems hardware. I think they are added through a PC-internal system clock. But it would be great if somebody can confirm that assumption.
About the RTC you can have a look at this device:Basti wrote:Layouting a first version of a CANbus-shield with CAN-controller, RTC, isolation, CAN-transciever, 5V/3.3V power supply.
Amazing, I didn't know of this chip and I instantly ordered a sample. Thanks for the tip.Wolfram23 wrote:In that regards, while the MCP2515 is a great CAN chip, why not go with a more "open hardware" friendly STN1110, which is still lightning fast with CAN but also compliant with other protocols? Additionally, it supports the "AT" command set (basically the same as used with modems) so I'm sure programming for it wouldn't be too difficult for someone of moderate skill (not me, sadly). Also, you can get a free sample (1 to 3). I asked for one and apparently it will arrive at my door in a few weeks. Can't beat free.
You better take a look at OpenStreetMap then. The maps are better then Google's in some parts of the world and they are 100% free.Wolfram23 wrote:Since there is talk of GPS, I've done a limited amount of digging. For example here is another Arduino project that data logs both OBD II data and GPS information:
http://www.practicalarduino.com/project ... y-platform
I'd prefer real-time GPS/Navigation, though. I know it won't be easy. It seems like there's a decent selection of USB GPS devices.
There's also a selection of free Linux open source GPS software. I know for sure Google allows use of their maps and such as long as the software is open source. Ideally I'd like to have local maps pre downloaded to remove the need for a 3G component (city/province/state/country/continent/world/whatever you need)
http://tuxmobil.org/linux_gps_navigatio ... tions.html
http://www.fsckin.com/2008/04/06/review ... -packages/
That's great! Thanks. I'll definitely check it out. Sadly my only way to try stuff on a Pi right now is with QEMU emulation at home, but maybe I'll buy a USB GPS device and see if I can get it to work in that environment.jannis wrote: You better take a look at OpenStreetMap then. The maps are better then Google's in some parts of the world and they are 100% free.
Software for navigation using Offline (pre-downloaded) maps can be found in the wiki:
https://wiki.openstreetmap.org/wiki/Rou ... ineRouters
Not sure what you're getting at. The MCP2515 has been discussed already. The STN1110 is a "real" CAN interface too, but also works with all other OBD II protocols which makes it useful for a lot more people like myself, driving a car older than 2008.bertr2d2 wrote:stn1110 seem to be interesting - but less than 5 bucks you could get a "real" CAN interface:
Users browsing this forum: No registered users and 8 guests