11-12-2010 07:51 AM
I have a number of cRIO systems deployed with UPS (uninterruptible power supply) connected to each unit. In the event of a long power failure, I want to somehow let the cRIO know how long the UPS has been on battery power, so that It can go into a safe shut down state.
My UPS is a "APC BR1500" with a USB data port normally used to communicate this status to a computer. http://www.apc.com/resource/include/techspec_index.cfm?base_sku=BR1500
Any ideas on how to do this?
Thanks
11-15-2010 10:34 AM
The features and limitations of the USB port on cRIO's can be found here. To communicate to the cRIO through the USB port you will need to develop your own USB driver for the APC UPS which may or may not be difficult depending on the communication protocols of the APC.
Another option, though not as elegant, is to place a phototransistor over the LED on the front and connect it to a digital IO pin.
Hope this helps!
11-23-2010 09:14 AM
Hi,
how about monitoring supply voltage using a suitable relay and a digital input? Of course this solution would not recognize a UPS fault. And there really isn't a digital output for this purpose on the UPS?
Regards,
Simo
11-23-2010 10:03 AM
Thought about monitoring the supply voltage like you said, but that would only give me whether or not the UPS is on battery power. I would like to know, if possible, when the battery is low and the UPS is about to shut down as well.
I contacted APC and they sent me this diagram for the serial cable that you can buy for this UPS. *see attached*. I played around with it for a bit and using an oscilloscope, was able to see the "ON battery" line go high when I removed live voltage. However, I am having trouble reading a "LOW battery" condition (even running the battery all the way down). I unfortunately was not able to get an schematics from them on the UPS side of this cable.
Ideally, I'd be able to use the serial lines as my digital I/O (unfortunately, the UPS does not have an dedicated DIO on it). I would have preferred to use that.
11-23-2010 10:29 AM
Jim,
Do you have an analog C series card on the cRIO? You could make a little circuit which performs the ADC for you then you could you use digital. What behavior is the "LOW battery" line giving during your testing? The schematic pdf is not the most thorough I have seen.
11-23-2010 10:54 AM
Ben,
Sorry the drawing is not very detailed. I didn't have a lot of time so I quickly put something down to explain what i was doing. The issue is not a ADC problem. The "ON Battery" Line acts like a digital status contact like expected, but the "LOW Batt" line does not change. The "Low batt" line is constant @ 2.4V as shown in the test results table. It remains at 2.4V when the UPS is plugged in, on battery, and even when the battery is drained all the way. It does not change.
It is almost like the "Low batt" line is acting like a voltage-divider. When I change the value of the pull-up resistor, the voltage I measure changes (but remains constant).
Perhaps I am hooking up my test setup incorrectly. I do NOT have the serial cable from APC, I am trying to re-create it, hoping to use the pins as Digital Outs.
11-23-2010 10:56 AM
Jim,
Is there a way you could call APC to clarify on this line? It seems like a strange design to have a pin called LOW battery that does not change despite the battery being low or not.
11-23-2010 11:00 AM
Already tried. They won't provide any detailed support for what this is doing because it is oly "designed" to plug into a serial port/USB port and use their provided software.
11-23-2010 11:38 AM
You may want to try using their software to see if there is a low battery indicator, that way you at least know it is possible.
11-23-2010 01:39 PM
Jim,
I am not sure how the Schottky diodes are being used in this particular diagram but it seems like the cable has no extra signal manipulation other than protection. It seems like the line should go high when the battery gets low, as the name suggests. Why did you choose to add a voltage divider to the testing circuitry and not just attach the oscilliscope directly to the pins?