Home AMX User Forum AMXForums Archive Threads AMX Hardware

Hotfix Release: NI-device side firmware Build v1.30.4 (for device 5001)

I have just solved a problem that i had with this firmware and i just wanted to share it.

The problem that led me to use this firmware as i reported it to AMX Technical Support:

"I have 2 NXC-IRS4 cards on an NI-4100 which sometimes are not working properly. The system works fine but there have been some intermittent problems. First of all, sometimes when i reboot the processor, the cards do not come online. Second, sometimes they go offline while the system is working. Third, sometimes they seem to be online but their ports are not responding to any commands. I would really appreciate it if you could help me solve these problems. What should i be looking for? Could these be hardware or programming related? The master processor is at v3.50.430 firmware, 5001 is at v1.20.7 and the cards are at v1.0.15 firmware."

AMX's Technical Support Response:

"A very common issue that would fit with your description of the problem is the cards not being fully seated in the NI-4100. It requires quite a bit of force to fully seat the cards. You should hear an audible sound when the card is fully seated. It should also take force to remove a card that is fully seated.

5001 should at least be at 1.20.7 and on ftp.amx.com we have a hotfix version of 1.30.4 available. Hotfix master firmware 3.60.447 or later is required.

If you hammer a card with commands it can also potentially knock the device offline or otherwise stop the device from responding. You might check your program and make sure you do not send repeated pulse commands to all four ports of the card. If you need to pulse all four ports together, I would recommend pulsing them in pairs. For example I would pulse ports 1 and 3, then use a WAIT in code before pulsing ports 2 and 4.

Your system should obviously have enough power for all the devices. An underpowered system will act unpredictably. The following is an excellent resource to check how much power your system requires:

http://www.amx.com/techsupport/techNote.asp?id=146

And now for the last item to check. When the system is in a failed state, meaning the NXC-IRS4 cards are not responding to commands, have someone look at the card itself. On the face of the card, (not the 4 red and 4 yellow leds facing out), there are some other leds that should always be off in normal operation. If one of these leds is on or flashing, pull the card and find the number next to it and report this. On a NXC-IRS4 the most common of these is the CR10 led."

I checked everything the technical support suggested and i got the v1.30.4 firmware from them as well.

First firmware upgrade attempt problem as reported to AMX:

"I uploaded the new hotfix firmware (v3_60_447) on the master processor without any problems. Then, I tried to upload the other hotfix firmware (v1_30_4) on the integrated controller (device 5001). This firmware upload was never completed. The file was sent to the device but the NetLinx Studio crashed on the please wait message and so did my laptop. Seeing that there was no response from the software I rebooted the laptop after a while. After the reboot, I refreshed the online tree in NetLinx Studio but 5001 was not coming online. I tried rebooting the processor and cycling the power but 5001 was still offline. I even tried to connect using RS-232 to see if 5001 would appear online but again nothing. As a last resort I tried to re-upload the firmware to 5001 even if it appeared to be offline. The upload failed."

The processor was sent to AMX because it had to be repaired.

Repair notes from Technical Support:

" Fault confirmed. Replaced U22. Reprogrammed HCS12's. Tested ok."

However they did not upgrade the 5001 firmware. The device was still at v1.20.7. So i tried to upgrade the firmware once again..

Same thing happened but this time i called technical support as soon as i realized that the firmware was taking quite a long time to finish (about 20 minutes..). The file was sent but NetLinx Studio was still showing the please wait message.. Technical support told me to cut the processor's power and then plug it back in. They also re-sent me the firmware file in case the one i had was corrupted..

Turns out that the reason the firmware upgrade was never finished was that the “Reboot device” box on the Firmware Transfer window was ticked and not because the file that I had was corrupted!! The file was successfully sent every time but the device could not be rebooted automatically!

Comments

  • ericmedleyericmedley Posts: 4,177
    My workaround for this has been to check for the cards online stat and reboot if not there. This seems to have fixed the problem. I put in a '3 strikes, you're out' counter but I't never hit more than once.
  • chillchill Posts: 186
    I have a 'problem child' project with 19 or so NI masters which would occasionally, randomly become unresponsive. One symptom was the masters apparently hanging - losing network connectivity and not responding to the Program port. The other symptom was that the NI devices, or sometimes the cards in the lone 4100, would stop working. A reboot would often fix it, but often it would take more than one reboot. Eric, I like your 'watchdog' idea.

    Anyway, I've been running the 3.60.447 - 1.30.4 combo for about a month now and it *seems* to be fixed. Of course, with a problem that's random and rare, how long do you wait before you decide it's fixed?

    And just to be extra-painful, the job is a) ridiculously far into the red by now and b) over 100 miles from our shop.
    .
  • chillchill Posts: 186
    Almost forgot...

    As kousios correctly notes, 1.30.4 device firmware *requires* 3.60.447 master firmware, so upgrade the master first. But before upgrading the device, it is semi-required to shut down the NI's RS232 ports, e.g.
    send_command 5001:1:0,"'RXOFF'"
    
    ...and so on for all the NI's RS232 ports. I bricked a couple of masters by not doing this. You can skip this if you're not doing any polling on any RS232 port, but due to past experience I'd do it anyway out of paranoia.
    .
Sign In or Register to comment.