Host Engineering Forum
General Category => Do-more CPUs and Do-more Designer Software => Topic started by: CReese on April 16, 2014, 05:16:37 PM
-
Hello,
I'm using a CAN/serial converter: http://www.adfweb.com/download/filefold/CANbus_from_to_RS232_MN67290_ENG.pdf
I've got it set up to read messages happily using STREAMIN @IntSerial, but after a minute or two it just stops reading. No activity light on the CPU. The program is still cycling as it's logging failed read operations, and CAN device is still transmitting, as if I power down the CPU and then power it back up, I'm again receiving messages. Taking it out of run or stopping the PLC also doesn't do the trick. Any idea what could be happening here?
Thanks,
Colin
-
That sounds familiar. What version of firmware are you running?
-
1.1.0. No asterisk here on NetEdit.
C
-
NetEdit doesn't know how to update Do-more. Not sure what version fixed that, but it sounds like something that was fixed. Not in office, but I'll check later.
-
I updated ALL the things. No love.
Colin
-
What version? The bug was a condition that would cause the receive interrupt to get shut off in the internal serial port. It would run for a few seconds to a few minutes, stop receiving, and required a power cycle to fix. I'm pretty sure it was introduced in 1.1.0, and fixed shortly after.
-
DM: 1.3.0
OS: 1.2.0
Booter: 3.0.5
FPGA: 3.5
Hardware: 7B
I forgot I happen to have a serial module plugged into this guy, so I went ahead and switched over and it's working great. Happy to test and fix the internal serial though, as I don't really NEED an additional three serial ports in this applications.
Colin
-
So 1.2.0, huh? Poo.
What are your port settings, your PLC scan time, and what is the input data typically?
-
I've tried about every port setting there is.
For the CAN protocol converter, I need to use 115200 no parity. I've used 8/1 data/stop, and every combination under the sun for the other two settings. Data comes through fine, just very shortly. Interestingly, changing the serial settings does result in reads working shortly thereafter as well.
-
That is exactly the bug I fixed. After an update it works again because the port is reinitialized. I wonder if the fix got removed somehow...
-
After studying it, I think this is a different issue. It looks like if the port FIFO gets full, I'm shutting off the interrupt and not re-enabling when the data is read. I'll pursue that, but out of curiosity, when you do a DEVCLEAR on the serial port, does it start receiving again?
-
I will definitely check when I get over there. Thanks.
-
I'll test this when I'm back to the test bench, but do you have any more ideas? An uprev on design has eliminated my serial card.
-
It's been fixed in the latest firmware.
-
Great. Thanks.
-
Confirmed fixed for me.