Configuration Suggestions

[Anonymous user]

My company has a very large number of SUM1500RMXL2U & SUM3000RMXL2U units that are deployed into remote sites across the country. These are running servers that are placed in dedicated computer rooms. We are aware that in most, if not all, of these computer rooms that the power is not up to par with data center standards and is often host to voltage swings and dirty power. Since installing these units we seem to be seeing a lot more failures then we had expected from our previous Smart-Ups (usually 1500 but an older models) had. In many cases we are seeing that the units are turning themselves off (NMC card posts messages of turning all the port groups off and then going to UPS > Control the only option is to turn the UPS on). Some of the messages most common messages we've seen accompanied with this are AVR Relay Fault, & Self-test failed UPS overloaded.

It is my belief that these units are much more sensitive then the UPSs we had used in the past and I'm looking for some configuration tips that might avoid these types of issues. We have already experimenting some with lowering the sensitivity to low and it's unclear as of yet if that has helped or not, but we have definitely had issues with devices set to low. I've also been toying with the idea of widening the high/low transfer voltage range, but am not sure of if that will help the situation or if it's advised.

It has been baffling to us that these units are turning themselves off and not coming back on (in many cases we may find one that has been powered off for a week and now has a 100% full battery and works fine once turned back on). In this off state they are preventing power from reaching the devices. Does anyone have any good tips on what setting changes we might be able to make to change help these issues or what's causing these devices to turn off.

Thanks

[BillP]

Good day Bryan,

AVR Relay Fault, & Self-test failed UPS overloaded aren't good. Do you have any idea if these UPS would show some LED indicators or beeps? A UPS would fail a self-test if the battery is bad and the replace battery indicator would be solid red. For the AVR relay fault, the UPS would show that the 6 center LEDs are flashing and in some cases that I have encountered it drops the load.

It would be best to have someone on site and take a good look on the UPS and do an inventory of what are the errors each UPS is doing while they are on the phone with our Tech Support.

[Anonymous user]

Is there anyway to tell which lights are on remotely? I've never seen anywhere on the NMC where it shows that (though it would be extremely useful!).

The problem with checking the lights (unless there is a way to check them remotely) is that we don't have qualified technical people in these locations. What usually happens is we find one of these in this state and we either try to fix it or dispatch a technician. We obviously would rather not incur a cost just to have someone check lights, and it's been my experience that when these errors happen 75% of the time I can just turn the UPS back on and it's fine (the other 25% of the time it fails again in a less than a week and we dispatch a technician). My concern is why the 75% are shutting off and requiring action to turn them back on if they are fine, which seems to be the case when they don't fail again.

To give an example, on Monday I looked at 12 different UPSs where the devices attached were reporting partial power (we split power between 2 UPSs to prevent the device failing if a UPS fails) Of those 12, 2 reported AVR relay fault and 10 were Self-test failed UPS overloaded. All 12 powered back on fine and as of this morning are still running fine from all accounts. Obviously we don't want to incur the cost of 12 site visits (all were different locations) just to have someone report what lights are lit on the device when turning it back on is the only thing needed to fix the issue.

[BillP]

can you post a few of the event and data logs from the NMC for us to take a look at. The AVR relay may weld due to prolonged use and generally will not unweld most of the time. It may clear momentarily but fail again the next time the UPS wants to use the AVR. AVR messages in the log would help show that we are having a constant issue at the site with higher or lower voltage than normal and it should be addressed to avoid causing the failures on the UPS. The UPSs are designed to boost and trim when needed for an hour or two here and there but not 24/7. We could look at the logs and see if there is any info on there relating to that situation.

to download the event and data logs from the NMC, [use these instructions|http://nam-en.apc.com/app/answers/detail/a_id/9321].

this will give us an idea on any recommended configuration changes.

[Anonymous user]

That should be all.

[Anonymous user]

I've made the change to our setup documentation to have low sensitivity set to any new or replacement device some time ago. I'm also piloting a widened range in a few places and I haven't seen any problems from that yet. I still don't completely understand that changing that range is doing, that's just changing when it uses utility power and when it uses battery power, but it still boosts/trims?

This part is the most concerning part of all the issues we've seen:
"i think that your equipment is probably turning off when the UPS is in bypass due to a fault and then utility power is lost or below a certain level, thus the UPS is shutting off cause it cant run. bypass means that there is no battery back up and the UPS is passing utility power to the load."
I can understand the unit going into bypass and I full expect that to happen, but turning it self off effectively blocks all output voltage (battery or utility) sometimes for a temporary spike that goes way. This seems to contradict the concept of having a UPS to me because we put them in place so that the devices don't loose power. I still don't understand why these are powering off and what can be done to stop them.

[BillP · Accepted solution]

when the UPS is in bypass, it is not acting like a UPS, most likely because something is faulting so it cant pass power. if you see this problem, we need to look into why the unit is going into bypass because there is a problem causing it to not function properly.

it could be faulting because of all of the power issues and boosting and trimming and everything.