NI 9792 power consumption

[LDBM]

What is the expected power consumption of a NI 9792 controller node that would retreive 32 sensor readings from 8 nodes ( 4 sensors per node), once an hour?

My main concern here is how to minimize the controller's power drain, as it would be deployed running on battery for about a year with a few month of little to none solar assist. Is there any sleep mode where power drain is minimal?

Thx.

L.

[WearWolf]

Hello BSI,

The expected power consumption of the NI WSN 9792 Controller is 9.5 W with a maximum consumption of 15 W. This link to the NI 9792 manual has the power requirements listed for reference. The controller does not have a sleep mode as it is a Real Time gateway. The wireless nodes have ways of limiting the power consumption but the NI 9792 does not.  

[LDBM]

Hmmm... this is a bit of a problem unfortunately...Under 12V this is 800mA, way too much. For reference, other controller products very much used for field deployement (battery powered)  are spec'd with power usage of about 1mA (Max) and 0.7mA in sleep mode....

What would be your opinion on having the 9792 power input controlled with a timer switch to turn on every 3-4 hours? Any potential issue with this approach beside the need to have an additional device?

Thx

L.

[WearWolf]

Hello BSI,

The only potential issue I can foresee is if you have multiple gateways with nodes on the same channel. If the gateway is deactivated then the nodes could establish a link with the new gateway. I would suggest assigning the specific nodes that are associated with each gateway to different channel sets. Other than that this would work to limit the power consumption.

[LDBM]

Thx Wear,

There would only be 1 gateway. Up to 8 end nodes. The gateway would be turned on to get meas. from the nodes. Presumably as indicated earlier in other posts,  the end nodes could potentially be asleep, only awaking once every 55 min (last sleep cycle time of the Fibonacci sequence). Is there a way for the controller gateway to send a "Wake up" signal to the end nodes so it does not have to wait up to 55 minutes to have them all awake??

Thx

L.

[WearWolf]

Hello BSI,

Unfortunately there is no way to wake up the nodes as they disable the radio when they enter the sleep cycle to save power. Each node has a 61 second heartbeat interval in which it uses to maintain a connection to the gateway. If this heartbeat does not detect the gateway it will enter the Fibonacci sleep/wake cycle until establishes a link. The gateway is intended to maintain a connection to the devices. It is possible to setup a node as a router to forward the messages from the WSN nodes to the gateway. These nodes use less power than the WSN gateway and can be used to increase the range of the system.

[LDBM]

Thanks Wear,

Can you clarify the use of the Router Node from your last post please? Would it keep the end nodes awake with its heartbeat? And would it , itself, stay awake ?

Thx

L.

[WearWolf]

Hello BSI,

The Router Node will always be on and can be used to extend the distance of your network. The reason I suggest router node would be to place your Gateway in a location that can be powered reliably so that it can remain on. The router node is suggested to be externally powered but will use less power than the gateway being run off of the battery setup. The router node can be powered effectively with a solar power setup as suggested in this knowledgebase. The Router node would only be able to keep the end node awake during the end nodes heartbeat phase if the router is in communication with the Gateway. This knowledgebase notes the Nodes power and performance benchmarks while this knowledgebase explains more detail regarding the router and WSN topologies.

[LDBM]

Thx Wear. Yes I was aware of these links. For applications that are remote and cannot rely on a stable source of power, solar may help, however can be unreliable at high latitudes... I think there is still a case to deal with the controller power usage, which is not that small.

On more general terms, I think NI should consider some level of "Sleep Mode" for controllers in order to decrease overall power consumption. Competing products for the type of applications I am looking at have such low power consumptions and are harder to displace because of this.

Thx

L.

[WearWolf]

Hi BSI,

Please refer to the NI Idea Exchange and add your ideas as R&D monitors new ideas to possibly implement. Thank you for your suggestions.