configuration for the network card with environmental sensors

[BillP]

We want to get alerts when the humidity goes over the range of 40-60%. When I configure it that way I get an email every 17 to 22 minutes. When I check the log it may be at 61% but on the configuration page it shows 47%. Does anyone have recommendations on how to configure this so that if it is a problem we will be notified but not constantly?

Temperature & Humidity: Temperature Sensor 1
Name:		restricted to ASCII characters only
Location:		restricted to ASCII characters only
Alarm Status:	Normal
Temperature:	64° F
Humidity:	59% RH
Temperature Thresholds		Humidity Thresholds
Maximum:	° F  [32 to 140]	Maximum:	% RH  [10 to 90]
High:	° F  [32 to 140]	High:	% RH  [10 to 90]
Low:	° F  [32 to 140]	Low:	% RH  [10 to 90]
Minimum:	° F  [32 to 140]	Minimum:	% RH  [10 to 90]
Hysteresis:	° F  [0 to 18]	Hysteresis:	% RH  [0 to 20]
Smart-UPS X 2000 Integrated EMU Date  Time  Vmin  Vmax  Vout  Iout  %Wout  %out  FrqOut  %Cap  Vbat  Tups  T1ambF  %Hum1  T2ambF  %Hum2   08/27/2015   13:08:26   115.87   116.70   120.28   6.62   42.75   41.58   59.97   100.00   131.93   68.23   62   54.0           08/27/2015   12:53:26   116.21   116.84   120.29   6.62   42.78   41.62   59.97   99.99   131.96   70.55   66   48.0           08/27/2015   12:38:26   116.03   117.03   120.28   6.62   42.82   41.63   59.97   100.00   131.93   70.26   69   56.0           08/27/2015   12:23:26   115.37   116.78   120.28   6.62   42.75   41.57   59.97   100.00   131.93   68.30   66   61.0           08/27/2015   12:08:26   115.98   116.95   120.28   6.59   42.64   41.48   59.97   100.00   131.96   67.15   62   58.0           08/27/2015   11:53:26   115.18   117.23   120.29   6.59   42.66   41.50   59.97   100.00   131.96   67.71   62   52.0           08/27/2015   11:38:26   116.20   116.93   120.29   6.59   42.58   41.45   59.97   99.99   132.00   68.68   62   49.0           08/27/2015   11:23:26   116.40   117.35   120.28   6.59   42.64   41.46   59.97   100.00   132.00   70.91   68   47.0           08/27/2015   11:08:26   116.56   117.37   120.26   6.62   42.81   41.64   59.97   99.99   132.00   70.41   69   51.0           08/27/2015   10:53:26   116.40   117.32   120.28   6.59   42.67   41.50   59.97   100.00   132.00   68.63   66   57.0           08/27/2015   10:38:26   116.43   117.37   120.28   6.59   42.64   41.48   59.97   100.00   131.96   67.06   62   62.0           08/27/2015   10:23:26   116.98   117.93   120.29   6.59   42.66   41.50   59.97   100.00   132.00   68.10   64   56.0

[Terry_Kennedy_apc]

Electronic humidity sensors aren't the most accurate things (this is a general condition, not specific to the APC ones).

What you probably want to do is to modify the notification defaults for an email recipient. It looks like you're running software 6.x.x. Here's how to do it in 6.2.1: Go to Configuration / Notification / Event Actions / By Event. Select Humidity from Environment Events. Pick one of the messages (for example, "Environment: A high humidity threshold violation exists for integrated Environmental Monitor") and click on it. Now you'll see a page for that event, with one or more email recipients listed. Click on one of them. You'll now see settings for delaying the reporting, how often to repeat the notification, and for how long.

As far as oscillating in / out of alarm conditions, go to Configuration / Universal I/O / Temp & Humidity. Click on a sensor name. Under Humidity Thresholds, configure the Hysteresis setting. The Help menu on that page (search for Hysteresis) does a very good job of explaining it, so I won't repeat it here.

[BillP]

Terry,

i will do this when I get to work. I greatly appreciate your help. Do you know if the temperature humidity probe that plugs into the PDUs is any more accurate? I know my boss is going to ask about all this. Is the up and down apart of the normal air conditioning process? Thanks again.

[Terry_Kennedy_apc]

It seems to be a general characteristic of the electronic humidity sensors. I may be mistaken and the actual humidity does vary by a large amount over short periods of time, but I'm used to old-style hygrometers which don't respond quickly. Wikipedia says "Humidity measurement is among the more difficult problems in basic metrology." In general, you can control the temperature or the humidity, but not both. Most data centers concentrate on temperature. I've attached 2 graphs (green is temperature in F, blue is relative humidity in %) one from my home server rack and one from a Tier IV datacenter (oddly, Tier IV is the best and Tier I is the worst). I don't do anything to control humidity, but the data center does (but it is secondary to temperature). The 2 locations are only a few miles apart and you can see the correlation between them, although the changes in the datacenter are smaller in magnitude because their cooling system controls humidity as well). The data for both graphs was collected with APC sensors (AP9335TH at home, AP9312TH at the datacenter).

e5KOqPjd4WOlKhhxU2bVUA%3D%3D.png

yxuKUA6HkcEi1coSMzH0fg%3D%3D.png

[BillP]

It make sense to me. For my boss, is there any documentation that you know of where electronic switches or equipment has to be exposed to Humidity of 60% for _____number of days before it starts to effect it? I would even like to know if the only way you could have 70% humidity in a room where the temps were from 50-60 would be if you were boiling water.

I have made the changes you suggested and raised the Humidity levels. I just don't want my boss to fuss at me for "just changing the alarm points" to keep from getting alerts.

[Terry_Kennedy_apc]

Each piece of equipment has its own limits. Things that perform mechanical motion (tape drives, printers, etc.) tend to be the most picky, until you get into esoterica (mass spectrometers, etc.) that you won't find in your computer room. I'd suggest picking a representative sample of the equipment you have in the room (for example, an IBM server, a HP printer, a Cisco Ethernet switch) and check their manuals. That will give you an idea of where your limits are.

Cooling tends to lower the humidity (as water condenses on the air conditioner evaporator coils). Heating also tends to lower humidity, unless you have steam heat with leaky valves. The main problem with humidity that is too low is that equipment is more susceptible to static discharges. If your computer room exchages air with the outside, then your indoor humidity will follow the outdoor humidity. That tends to be more of an issue in modern "green" data centers where they locate the center someplace cool (like Iceland) and just use outdoor air to cool the equipment.

Humorous anecdote: Some years ago I had an air conditioner control failure which resulted in the units running flat out. It got down to 49 degrees in the computer room, and there was fog in there.

[BillP]

Is there a way to know how accurate the sensor is? I mean could you say it could be plus or minus 10% or 15% or 20%?

[BillP]

Here is the output for the weekend. It looks like something is happening on a regular basis where the humidity gets up into the 80% range. I just don't know how much to be concerned by it.

Date	Time	T1ambF	%Hum1
08/31/15	7:58:33	62	58
08/31/15	7:48:33	64	49
08/31/15	7:38:33	68	57
08/31/15	7:28:33	69	65
08/31/15	7:18:33	68	70
08/31/15	7:08:33	66	75
08/31/15	6:58:33	64	73
08/31/15	6:48:33	60	54
08/31/15	6:38:33	62	67
08/31/15	6:28:33	62	49
08/31/15	6:18:33	66	53
08/31/15	6:08:33	69	63
08/31/15	5:58:33	69	71
08/31/15	5:48:33	68	75
08/31/15	5:38:33	66	79
08/31/15	5:28:33	62	72
08/31/15	5:18:33	62	65
08/31/15	5:08:33	60	56
08/31/15	4:58:33	62	66
08/31/15	4:48:33	62	52
08/31/15	4:38:33	64	52
08/31/15	4:28:33	68	61
08/31/15	4:18:33	69	77
08/31/15	4:08:33	68	82
08/31/15	3:58:33	66	86
08/31/15	3:48:33	62	84
08/31/15	3:38:33	59	58
08/31/15	3:28:33	60	56
08/31/15	3:18:33	62	55
08/31/15	3:08:33	64	57
08/31/15	2:58:33	66	59
08/31/15	2:48:33	68	69
08/31/15	2:38:33	69	83
08/31/15	2:28:33	68	86
08/31/15	2:18:33	64	87
08/31/15	2:08:33	60	75
08/31/15	1:58:33	60	61
08/31/15	1:48:32	60	59
08/31/15	1:38:32	62	58
08/31/15	1:28:32	64	57
08/31/15	1:18:32	66	58
08/31/15	1:08:32	69	71
08/31/15	0:58:32	68	83
08/31/15	0:48:32	68	85
08/31/15	0:38:32	64	86
08/31/15	0:28:32	60	69
08/31/15	0:18:32	60	61
08/31/15	0:08:32	60	59
08/30/15	23:58:32	62	59
08/30/15	23:48:32	64	58
08/30/15	23:38:32	66	58
08/30/15	23:28:32	68	68
08/30/15	23:18:32	69	81
08/30/15	23:08:32	68	85
08/30/15	22:58:32	64	86
08/30/15	22:48:32	60	72
08/30/15	22:38:32	60	61
08/30/15	22:28:32	60	60
08/30/15	22:18:32	62	59
08/30/15	22:08:32	62	58
08/30/15	21:58:32	64	58
08/30/15	21:48:32	68	62
08/30/15	21:38:32	69	80
08/30/15	21:28:32	68	84
08/30/15	21:18:32	66	86
08/30/15	21:08:32	62	80
08/30/15	20:58:32	60	61
08/30/15	20:48:32	60	59
08/30/15	20:38:32	62	59
08/30/15	20:28:32	64	57
08/30/15	20:18:32	66	58
08/30/15	20:08:32	68	65
08/30/15	19:58:32	69	80
08/30/15	19:48:32	68	84
08/30/15	19:38:32	66	86
08/30/15	19:28:32	62	80
08/30/15	19:18:32	60	61
08/30/15	19:08:32	60	59
08/30/15	18:58:32	62	58
08/30/15	18:48:32	62	58
08/30/15	18:38:32	66	58
08/30/15	18:28:32	68	65
08/30/15	18:18:32	69	80
08/30/15	18:08:32	68	84
08/30/15	17:58:32	66	86
08/30/15	17:48:32	62	79
08/30/15	17:38:32	60	60
08/30/15	17:28:32	60	58
08/30/15	17:18:32	62	59
08/30/15	17:08:32	64	58
08/30/15	16:58:32	66	58
08/30/15	16:48:32	68	65
08/30/15	16:38:32	69	80
08/30/15	16:28:32	68	83
08/30/15	16:18:32	66	86
08/30/15	16:08:32	62	80
08/30/15	15:58:32	60	60
08/30/15	15:48:32	60	59
08/30/15	15:38:32	62	57
08/30/15	15:28:32	62	57
08/30/15	15:18:32	66	58
08/30/15	15:08:32	68	63
08/30/15	14:58:32	69	79
08/30/15	14:48:32	68	82
08/30/15	14:38:32	66	85
08/30/15	14:28:32	62	79
08/30/15	14:18:32	60	60
08/30/15	14:08:32	60	59
08/30/15	13:58:32	62	58
08/30/15	13:48:32	64	57
08/30/15	13:38:32	66	58
08/30/15	13:28:31	68	65
08/30/15	13:18:31	69	78
08/30/15	13:08:31	68	82
08/30/15	12:58:31	66	85
08/30/15	12:48:31	62	75
08/30/15	12:38:31	60	59
08/30/15	12:28:31	62	58
08/30/15	12:18:31	62	57
08/30/15	12:08:31	64	57
08/30/15	11:58:31	66	57
08/30/15	11:48:31	69	70
08/30/15	11:38:31	68	78
08/30/15	11:28:31	66	82
08/30/15	11:18:31	64	84
08/30/15	11:08:31	60	64
08/30/15	10:58:31	60	59
08/30/15	10:48:31	62	57
08/30/15	10:38:31	64	56
08/30/15	10:28:31	66	56
08/30/15	10:18:31	68	62
08/30/15	10:08:31	69	74
08/30/15	9:58:31	68	79
08/30/15	9:48:31	66	83
08/30/15	9:38:31	62	75
08/30/15	9:28:31	60	58
08/30/15	9:18:31	62	56
08/30/15	9:08:31	62	55
08/30/15	8:58:31	66	56
08/30/15	8:48:31	68	61
08/30/15	8:38:31	69	71
08/30/15	8:28:31	68	76
08/30/15	8:18:31	66	82
08/30/15	8:08:31	62	77
08/30/15	7:58:31	60	58
08/30/15	7:48:31	62	56
08/30/15	7:38:31	62	55
08/30/15	7:28:31	66	56
08/30/15	7:18:31	68	59
08/30/15	7:08:31	69	69
08/30/15	6:58:31	68	74
08/30/15	6:48:31	66	81
08/30/15	6:38:31	62	76
08/30/15	6:28:31	60	56
08/30/15	6:18:31	62	55
08/30/15	6:08:31	64	54
08/30/15	5:58:31	66	55
08/30/15	5:48:31	69	60
08/30/15	5:38:31	69	68
08/30/15	5:28:31	68	72
08/30/15	5:18:31	66	79
08/30/15	5:08:31	62	73
08/30/15	4:58:31	60	55
08/30/15	4:48:31	62	55
08/30/15	4:38:31	64	54
08/30/15	4:28:31	66	55
08/30/15	4:18:31	69	62
08/30/15	4:08:31	68	66
08/30/15	3:58:31	68	71
08/30/15	3:48:31	64	78
08/30/15	3:38:31	60	68
08/30/15	3:28:31	60	51
08/30/15	3:18:31	62	51
08/30/15	3:08:31	64	51
08/30/15	2:58:31	68	55
08/30/15	2:48:31	69	62
08/30/15	2:38:31	68	66
08/30/15	2:28:30	66	72
08/30/15	2:18:30	62	77
08/30/15	2:08:30	60	52
08/30/15	1:58:30	60	50
08/30/15	1:48:30	64	50
08/30/15	1:38:30	66	54
08/30/15	1:28:30	69	57
08/30/15	1:18:30	68	60
08/30/15	1:08:30	68	65
08/30/15	0:58:30	64	73
08/30/15	0:48:30	60	62
08/30/15	0:38:30	60	53
08/30/15	0:28:30	62	51
08/30/15	0:18:30	64	52
08/30/15	0:08:30	68	53
08/29/15	23:58:30	69	56
08/29/15	23:48:30	68	60
08/29/15	23:38:30	66	67
08/29/15	23:28:30	62	72
08/29/15	23:18:30	60	53
08/29/15	23:08:30	62	51
08/29/15	22:58:30	64	51
08/29/15	22:48:30	68	50
08/29/15	22:38:30	69	54
08/29/15	22:28:30	68	57
08/29/15	22:18:30	66	61
08/29/15	22:08:30	64	69
08/29/15	21:58:30	60	52
08/29/15	21:48:30	62	51
08/29/15	21:38:30	64	50
08/29/15	21:28:30	66	50
08/29/15	21:18:30	69	51
08/29/15	21:08:30	68	54
08/29/15	20:58:30	68	57
08/29/15	20:48:30	64	62
08/29/15	20:38:30	60	59
08/29/15	20:28:30	60	50
08/29/15	20:18:30	62	49
08/29/15	20:08:30	66	49
08/29/15	19:58:30	69	47
08/29/15	19:48:30	69	50
08/29/15	19:38:30	68	53
08/29/15	19:28:30	66	58
08/29/15	19:18:30	62	57
08/29/15	19:08:30	62	57
08/29/15	18:58:30	62	48
08/29/15	18:48:30	64	46
08/29/15	18:38:30	68	45
08/29/15	18:28:30	69	48
08/29/15	18:18:30	68	51
08/29/15	18:08:30	66	56
08/29/15	17:58:30	62	60
08/29/15	17:48:30	62	52
08/29/15	17:38:30	62	58
08/29/15	17:28:30	64	45
08/29/15	17:18:30	68	45
08/29/15	17:08:30	69	47
08/29/15	16:58:30	68	50
08/29/15	16:48:30	66	54
08/29/15	16:38:30	64	59
08/29/15	16:28:30	62	48
08/29/15	16:18:30	64	54
08/29/15	16:08:30	64	45
08/29/15	15:58:30	68	46
08/29/15	15:48:30	69	47
08/29/15	15:38:30	68	50
08/29/15	15:28:30	66	55
08/29/15	15:18:30	64	59
08/29/15	15:08:29	60	48
08/29/15	14:58:29	64	54
08/29/15	14:48:29	62	45
08/29/15	14:38:29	66	46
08/29/15	14:28:29	69	48
08/29/15	14:18:29	68	51
08/29/15	14:08:29	66	55
08/29/15	13:58:29	64	60
08/29/15	13:48:29	60	51
08/29/15	13:38:29	64	53
08/29/15	13:28:29	62	47
08/29/15	13:18:29	66	47
08/29/15	13:08:29	69	49
08/29/15	12:58:29	69	52
08/29/15	12:48:29	68	56
08/29/15	12:38:29	64	60
08/29/15	12:28:29	62	56
08/29/15	12:18:29	62	54
08/29/15	12:08:29	62	54
08/29/15	11:58:29	66	48
08/29/15	11:48:29	69	49
08/29/15	11:38:29	69	52
08/29/15	11:28:29	68	56
08/29/15	11:18:29	66	61
08/29/15	11:08:29	62	64
08/29/15	10:58:29	62	53
08/29/15	10:48:29	64	62
08/29/15	10:38:29	64	47
08/29/15	10:28:29	68	48
08/29/15	10:18:29	69	53
08/29/15	10:08:29	68	57
08/29/15	9:58:29	66	61
08/29/15	9:48:29	64	67
08/29/15	9:38:29	62	50
08/29/15	9:28:29	64	60
08/29/15	9:18:29	62	46
08/29/15	9:08:29	66	49
08/29/15	8:58:29	69	56
08/29/15	8:48:29	68	58
08/29/15	8:38:29	66	63
08/29/15	8:28:29	64	69
08/29/15	8:18:29	60	54
08/29/15	8:08:29	62	59

[BillP]

Hi,

For the AP9335TH

Temperature accuracy ± 3ºF (±2 ºC), from 32 to 104ºF (0 to 40ºC)

Humidity accuracy ± 8% RH, 30 to 80% RH, from 59 to 95ºF (15 to 30ºC)  ± 4% RH, 20 to 90% RH, at 77ºF (25ºC)

Humidity accuracy data is based on a curve and the closer to the outer temperature extremes the less accurate the unit is. At 25C accuracy is ± 4% when you get down to 15C or up to 30C accuracy is at ± 8%.

[Terry_Kennedy_apc]

What concerns me are the large swings between consecutive samples. There's at least one case of a 26% increase in a 10 minute period. I find graphing the data make visualizing what's happening a lot easier. I use MRTG for this.

Either there's something very strange going on (do the cleaning staff come in and mop the floors?) or you have a flaky sensor.

[BillP · Accepted solution]

There should not be a cleaning staff that comes in the room at all. The sensor in near the floor. The ductless air conditioner mounted at about 6 feet is off the floor on a wall that is 9 feet away from the opposite wall. I didn't know if the AC blew on or next to the sensor if it would have an effect either when it is cycling or not. There is also a open space below the door into the room that I'm having weather stripping installed on to seal up the room.

Do you think a sensor that you plug into the PDU and can be remotely located away from the AC would be better?

The only other thing I can think of is to get a recording hygrometer and see if the sensor is way off. I'm at a loss.