Has anyone experienced problems with these controllers? Specifically, where they power-cycle by themselves and get locked into Bootloader Mode?
Three years ago, we installed five new i2 controllers (two i2810's and three i2814's) at a site.
Nearly one year later, one of the i2814's started to power cycle every once in a while (every week or two) by itself, a few months later another one began having the same issue, and a few months after that the third one starting doing it as well. However, these never reset at the same time - they would power cycle independently, rarely on the same day/week.
I know they were power cycling because the maintenance staff would hear the fans they were controlling randomly turn off, the point values on CyberStation would turn into question marks, and then the fans would come back on after for a minute or so and the point values would return to normal. On one of the controllers, before I replaced the battery, it would return with the setpoints set to their default values in the flash memory, so I know they were cold-starting.
Also, every couple of months, after rebooting, they would get stuck in the Bootloader Mode. A quick firmware update would get them running again, but after a couple of times the update would fail and they would not come back. We've replaced each one of the i2814's since they were installed, and all was fine for about a year, but last week, one of them started acting up in the same fashion... I'm now worried that the other two will also follow suit.
When I sent the original ones back to the factory, they stated that high voltage took them out and they're non-repairable, and not covered under warranty. However, all five of these controllers (the two i2810's and the three i2814's) are fed from the same 500VA transformer, and this issue will only occur to one of the controllers at a time, and no other power interruptions/surges in the buildings are witnessed. I haven't stuck a power meter to trend the voltage, but since one controller out of the five will "bug out" randomly, while the others remain completely fine, I'm having a hard believing its a building power issue.
Also, I'm not having these issues with the i2810's - only the i2814's. There's also about a dozen old TCX controllers on-site and I'm not having any issues with them.
One thing I started looking at was the analog outputs on the i2814's. All them are feeding Johnson Controls' EP-8000-2 Electronic-To-Pneumatic transducers. One thing that caught my attention was that the i2814's analog output impedance is 2k ohm minimum (sourcing only), while the input impedance to the E2P transducers is 1k ohm minimum. Some of my colleagues say that the controllers are overloading the E2P's, but with maybe an additional 10mA max, I'm not sure if this would actually be the culprit. None of the internal fuses on the i2814's analog outputs have tripped either.
I've checked all of the wiring and it seems to be okay, the infinity trunk is shielded and grounded at one end. I've also cleaned up their databases, removing any non-essential points and deleted/stopped all trends, and made they programs very simple and straight-forward.
Has anyone ran into a similar issue? Any help would be appreciated.
Thanks.
Three years ago, we installed five new i2 controllers (two i2810's and three i2814's) at a site.
Nearly one year later, one of the i2814's started to power cycle every once in a while (every week or two) by itself, a few months later another one began having the same issue, and a few months after that the third one starting doing it as well. However, these never reset at the same time - they would power cycle independently, rarely on the same day/week.
I know they were power cycling because the maintenance staff would hear the fans they were controlling randomly turn off, the point values on CyberStation would turn into question marks, and then the fans would come back on after for a minute or so and the point values would return to normal. On one of the controllers, before I replaced the battery, it would return with the setpoints set to their default values in the flash memory, so I know they were cold-starting.
Also, every couple of months, after rebooting, they would get stuck in the Bootloader Mode. A quick firmware update would get them running again, but after a couple of times the update would fail and they would not come back. We've replaced each one of the i2814's since they were installed, and all was fine for about a year, but last week, one of them started acting up in the same fashion... I'm now worried that the other two will also follow suit.
When I sent the original ones back to the factory, they stated that high voltage took them out and they're non-repairable, and not covered under warranty. However, all five of these controllers (the two i2810's and the three i2814's) are fed from the same 500VA transformer, and this issue will only occur to one of the controllers at a time, and no other power interruptions/surges in the buildings are witnessed. I haven't stuck a power meter to trend the voltage, but since one controller out of the five will "bug out" randomly, while the others remain completely fine, I'm having a hard believing its a building power issue.
Also, I'm not having these issues with the i2810's - only the i2814's. There's also about a dozen old TCX controllers on-site and I'm not having any issues with them.
One thing I started looking at was the analog outputs on the i2814's. All them are feeding Johnson Controls' EP-8000-2 Electronic-To-Pneumatic transducers. One thing that caught my attention was that the i2814's analog output impedance is 2k ohm minimum (sourcing only), while the input impedance to the E2P transducers is 1k ohm minimum. Some of my colleagues say that the controllers are overloading the E2P's, but with maybe an additional 10mA max, I'm not sure if this would actually be the culprit. None of the internal fuses on the i2814's analog outputs have tripped either.
I've checked all of the wiring and it seems to be okay, the infinity trunk is shielded and grounded at one end. I've also cleaned up their databases, removing any non-essential points and deleted/stopped all trends, and made they programs very simple and straight-forward.
Has anyone ran into a similar issue? Any help would be appreciated.
Thanks.
