SUA750I - SMT750I - OID to get output power (Watts)

[Anonymous user]

Hi everyone,

I have SMT-750I and SUA-750I with AP 9630 Network Management Card 2 (FW 6.5) and I need to get the output load power (in watts). I've been trying to do this with an snmpget of 1.3.6.1.4.1.318.1.1.1.4.2.8 OID and I obtained the following results:

SUA 750I - "There is no such variable name in this MIB" 

SMT 750I - = INTEGER : 0

I can correctly get the VA value with snmpget of 1.3.6.1.4.1.318.1.1.1.4.2.3.0 but not the 

Watts value.

When I establish a telnet session with both models of UPS and execute "detstatus -om" I can see that the UPS reports correctly the values of: "Output Watts Percent" and "Output VA Percent".

I want to know if there is any form to get the output load power (in watts) with an snmpget. 

Thanks

[BillP]

Hi,

The OID is 1.3.6.1.4.1.318.1.1.1.4.2.8.0

Description: The total output active power of the UPS system in Watts

[Anonymous user]

Hi,

I didn't put the final .0 in the last post but I'm using the OID you said.  Bellow you'll find the results obtained with detstatus -om and with the snmpget. As you can see the result of the OID 318.1.1.4.2.8.0 is INTEGER: 0 instead of the 27 that 

shows detstatus -om.

BL6F0J35r9zDHQumcVoNlQ%3D%3D.jpeg

kVg8um02bh0f0po6%2Fe2S5A%3D%3D.jpeg

I hope you can find the solution. Thanks

[voidstar_apc]

Hi Alvaro,

Make sure you have the latest SUMX (6.5) on the SMT750I UPS... I think there was an issue with 0 being returned before SUMX 6.4. You mentioned having 6.5 there so it should be fine but it's worth checking.

If it is running the latest version, please attach a debug tarball from About->Support->Generate Logs.

Also worth noting: 318.1.1.1.4.2.3.0 is the highest of the w% and va% readings (ie, max(w%, va%)). So it represents load measurement with the least margin whether that's VA or W.

[Anonymous user]

Hi,

thanks for the information, it was useful to me. Considering what you said about the version of SUMX I tested different combinations of UPS model and FW version with the following results.

QlOThmRQLq7WaTm9KZ%2FcFQ%3D%3D.jpeg

Considering the results and the models involved, Is it correct to conclude that OID 318.1.1.1.4.2.8.0 it's available only for SMT 750I with FW below 6.4 ?

In your opinion, is there any form to obtain the output power (W) in both models by snmpget ?

The reason why I need the W value is because I'm designing an automatized method to compare the runtime reported by the UPS with the theorical runtime provided by APC in the Runtime Graph http://www.apc.com/products/runtimegraph/runtime_graph.cfm?base_sku=SMT750I&chartSize=large

and conclude if the UPS requires or not a calibration.

[voidstar_apc]

Hi Alvaro,

That's a cool table!

I don't think SNMP provides a way to get the wattage on the older SUA units. However, most equipment these days has a power factor of 1 which means Watts = VA. The UPS' max watt rating is always lower than the VA rating. Therefore unless you're powering something really unusual, you can assume 318.1.1.1.4.2.3.0 is % watts.

I'm not clear on the benefits of your automatized method, but maybe I don't have the full picture. The runtime estimate reported should decrease as the battery ages -- the published theoretical runtime graph is for a brand new battery. My understanding is that the UPS' runtime estimate may require calibration when:

 1) The battery is replaced with a new battery rendering the old estimate meaningless

 2) The existing battery's actual runtime decreases but the estimate isn't updated because the UPS hasn't transferred to battery or doesn't have enough load to update the estimate.

For the first case when replacing a battery, the newer SMT/SMX UPSes use the LCD display to ask if the battery has been replaced and update the estimate if so. The older SUA units do not and may require a runtime calibration instead.

For the second case I'm not sure how you would detect that the health of the battery has decreased if the UPS itself hasn't detected it.

You might also be interested in Peukert's law which describes how lead-acid battery capacity varies with load.

[Anonymous user · Accepted solution]

Hi voidstar,

I agree. I found some cases in which the % W it's grater than the % VA, for example the following case:

rJwFsgZaKkDsJR0teXmIIw%3D%3D.jpeg

I suppose the reason why is that in the case of W% the nominal value is 500 W so power: 0,286 x 500 W = 143 W and in the case of VA% the apparent power is: 0,266 x 750 VA = 199,5 VA. Am I right ?

Bellow you'll find an example of the method results applied to a test set of 45 UPSes (Green : Runtime offset fixed with calibration,  Yellow : Reduced runtime, Red : Very reduced runtime). Cases in yellow or red need needs further analysis to determine the need of battery replacement or other actions. 

Thanks a lot for the information provided.

o3jF8T16gSUgEAzLa8UQrA%3D%3D.jpeg