Riello SEP series USB even in ECO mode identifies as in bypass stating UPS needs attention on pfSense

[ha11oga11o]

Hello all,

i have couple of Riello SEP2200 UPS units and have strange issue. Namely, at pfSense all works fine with rs232 connection but it insists that UPS needs attention when is set on ECO mode. At manual i found info which states
"In order to optimize performance, in ECO mode, the load is normally powered by the bypass. If the mains goes out of the permitted tolerance range, the UPS switches to ON LINE operation. If the mains returns within the permitted tolerance range for at least five minutes, the UPS goes back to powering the load from the bypass."

Indeed UPS is in bypass, but thats ECO feature (page 15). Does nut interpret that "feature" in riello driver, does it UPS send it at all as status, can it be added if not? UPS happily switches ON LINE on power outage, or any other disturbance on power grid. It simply works fine.

Theres also my questioning about it on pfSense forum. Forum username ha11oga11o.

Problem occurring at

2.7.2-RELEASE (amd64)
built on Fri Dec 8 21:55:00 CET 2023
FreeBSD 14.0-CURRENT

nut 2.8.2

System log of successful nut service start:

Dec 29 13:50:22 upsmon 59353 UPS RielloSEP2200 on line power
Dec 29 13:50:17 upsmon 59353 UPS RielloSEP2200: on bypass (powered, not protecting)
Dec 29 13:50:17 upsmon 59353 Communications with UPS RielloSEP2200 established
Dec 29 13:50:16 upsd 60905 Connected to UPS [RielloSEP2200]: riello_ser-RielloSEP2200
Dec 29 13:50:15 riello_ser 87467 Startup successful

From UPS detail page:

ups.status | OL BYPASS CHRG

Is there any way for nut to understand ECO mode for these models, and get it out from alerted state?

Many thnx in advance.

[jimklimov]

Hello, can you please clarify the NUT versions involved? 2.7.2 is antique (in IT terms), and 2.8.2 is only planned.

The description of this ECO mode seems to be what was called "line-interactive" in the early days of UPSes (as opposed to "online", which kept wall and load on separate circuits witha battery and inverter in the middle), maybe except the smart part of waiting for 5 minutes after a hiccup.

As for fixing the driver behavior - sure it is possible, albeit with a coding effort :)

On the one hand, at least a toggle option can be added to "treat situation X as Y on this device", generally speaking. Quite a few precedents in usbhid-ups about "OL+DISCHRG" handling, for example.

Hopefully at some stage criteria would be found (by device model/id or better) to set the default values of such toggles.

On another, I think riello driver authors can still be contacted by mail for insights...

CC @mzampieri70

[ha11oga11o]

Hello,

thanks for fast response. All i can see on pfSense about nut that latest version is 2.8.2. Its package which is installed on demand. Also i have same issue if i connect it on usb interface. Its not isolated on rs232 only.

Highlighted links are

nut version

Package Dependencies

If any other data needed i can provide.

Thanks

P.S. version 2.7.2 is pfSense actually, i stated below its nut version 2.8.2. I think i have latest which pfSense offers.

[jimklimov]

Hm, so it comes from a devel stream, so versions may be a bit off. NUT v2.8.1 was released 2023-10-31, so by the numbers in screenshot they may have an even earlier snapshot. But maybe just the naming is off and they included an errata patch...

CC @gdt : any insights on this one?

[ha11oga11o]

Hello,

just stop buy to see is there any possibility for fixing this? Can it be done at all?

Thanks.

[ha11oga11o]

Hello,

i hope i can bump this again? Can it be done anything about this? I dont wnat to be rude, but any answe would be ok. Thank you