USBizi, LPC2388FBD144, CAN bus

We need a can bus communication. I’ve tested the GHI example (NETFM 4.1 SDK) with 2 USBizi – and 2 SN65HVD235 transceivers. The write example runs on USBizi 1 and the read example on USBizi 2. To differentiate the two, the LED went on each time if data was sent. No errors occurred, but there didnt arrive any data if I set the breakpoint in the read event. In the next step I measured the voltage on CAN1 TD during data send. There is a constant voltage of 3.3 V, no peaks. The same procedure works in the SPI example on pin 11 and 13. Is it a normal state on CAN bus?

Best regards
Leonora

Do you have “ground” connected between the 2 boards?

Do you have 120ohm termination resistors on your CAN bus?

“ground” is connected to one board. CANH and CANL terminated 120 Ohm on each transceiver.

looks correct

What is the minimal wiring/installation to see voltage peaks on the line while read or write on the can bus?

Hi,
can you please send us a minimal configuration (hardware wiring) and software (the sample program in the user manual?) that can be used to put messages on the CAN bus and read these messages with another module from the same bus?

At the moment we think that there is a problem with your CAN implementation because we are unable to send and receive messages even with the Fez Domino evaluation board.
Has anyone used the CAN library successfully?
thanks
Sincerely
Holger

We changed the driver with latest SDKs but there are people using it already:
http://www.tinyclr.com/forum/2/2888/

Did you try our example as is? Do you get any errors in the error event?

What is connected to pin number 5 on your CAN transceiver?

We use GHI NETMF v4.1 SDK. The firmware is on both boards the same V. 1.07.

I load the example as is – one board write, the second board read with can_DataReceivedEvent and can_ErrorReceivedEvent.

No errors occur and no data is received.

Pin 5 on our transceiver is not connected.

We cannot detect voltage level changing on CAN1 TD while writing data, only during the board-booting.

I’m betting the next suggestion is to tie pin 5 (the AutoBaud pin) either to VCC or GND. Who knows how that behaves when it’s floating.

Just a thought:
are you sure the Loopback (LBK) pin is tied to ground on the transceivers?

Shouldn’t pin 5 be connected according to datasheet? Or you can use the exact same part we use and connect it the same way we have here http://www.ghielectronics.com/downloads/EMX/EMX_DevSys_sch.pdf

We’ve tied the autobaud pin 5 to ground pin 2 – the same behaviour…

However the transceiver may be connected or even not connected…
If the CAN software is installed correctly and you make a periodical write command you should measure on the Port Pins P0.0 and P0.1 of LPC2388 some traffic.
Am I right?

Not really, if the RX is in “dominant state” then nothing will happen on the bus. Also, many errors on the bus causes CAN to go into “bus off” condition, where it is basically off.

Do you have a CAN tool?

We tried to communicate via CAN between 2 FEZ Domino Boards. We use the transceivers SN65HVD235.
But I am sure that there must be any traffic (data frames) at TX (P0.1 respectively pin 7 of the connector of the board) at least at the very beginning. But there is nothing at all that could be measured!

We tied pin 8 to ground and it works. Thanks to all!

Glad it worked for you but I ave asked you if you copied our schematics already. If you look, we have pin 8 connected to ground! http://www.ghielectronics.com/downloads/EMX/EMX_DevSys_sch.pdf

It’s still our primary implementation, the only difference - pins 5,8 on both transceivers are tied to ground. We don’t use a can tool.