One wire communications w/ MotorMindB
Howler
Posts: 3
I think I probably know what's going on here but I'm wondering if there is any sort of work-around.
I've setup a UART to communicate with a MotorMindB.· This device is capable of one-wire communications, that is sending the the response back on TX line.
Here's my chunk of code....
··· _motorMind.setDirection(Uart.dirTransmit);
··· _motorMind.sendByte(0x55);
··· _motorMind.sendByte(command);
··· while(!_motorMind.sendBufferEmpty());
··· _motorMind.setDirection(Uart.dirReceive);
I then have a simple loop that looks for a response
····· byte data;
····· while(_motorMind.byteAvailable())
····· {
········· data = (byte)_motorMind.receiveByte();
········· System.out.print("From Motor->");
········· System.out.print(pin);
········· System.out.print("<-byte->");
········· System.out.println(data);
····· }
According to the motor mind doc's the response should occur within·500us-1000us of getting the data.
I would appear that commands that return only one byte don't get picked up at all.· Responses that return two bytes only return one byte and that byte is random.
Is it safe to assume that the switch between TX to RX just isn't happening fast enough and I'm droppings some of the return bits?· If so is there any work around?
Right now I'm out of VP's so unless I can create additional ones, I'm really stuck with the one wire solution.
Any help would be appreciated
Kevin...
I've setup a UART to communicate with a MotorMindB.· This device is capable of one-wire communications, that is sending the the response back on TX line.
Here's my chunk of code....
··· _motorMind.setDirection(Uart.dirTransmit);
··· _motorMind.sendByte(0x55);
··· _motorMind.sendByte(command);
··· while(!_motorMind.sendBufferEmpty());
··· _motorMind.setDirection(Uart.dirReceive);
I then have a simple loop that looks for a response
····· byte data;
····· while(_motorMind.byteAvailable())
····· {
········· data = (byte)_motorMind.receiveByte();
········· System.out.print("From Motor->");
········· System.out.print(pin);
········· System.out.print("<-byte->");
········· System.out.println(data);
····· }
According to the motor mind doc's the response should occur within·500us-1000us of getting the data.
I would appear that commands that return only one byte don't get picked up at all.· Responses that return two bytes only return one byte and that byte is random.
Is it safe to assume that the switch between TX to RX just isn't happening fast enough and I'm droppings some of the return bits?· If so is there any work around?
Right now I'm out of VP's so unless I can create additional ones, I'm really stuck with the one wire solution.
Any help would be appreciated
Kevin...
Comments
The way you test could also be wrong if response times are not equal
to all commands.
You use
while (rx.byteAvailable()) {}
directly after sending command.
If at the moment of that test no (complete) data has arrived you still
exit the while loop.
Try using a delay first to be sure no data is received.
CPU.delay(10500); //wait 1 second
while (rx.byteAvailable()) {}
If you then still get no responses then you know for certain
the direction change takes too long.
regards peter
Are there any options like adding an additional VP? I think I tried this early on in the CPU class, but wasn't too successful. Is there any mechasim for programming at a lower level (drop down to assembly language) on the Javelin?
Thanks again
Kevin...
2 uarts, meaning you need to stop another VP temporarily. If you use
another transmit uart, you can safely stop that.
In my code I frequently use only 1 VP slot for all transmit uarts.
Without a look at your code I can't give an exact solution, if any.
regards peter
Kevin...
regards peter