[Nut-upsuser] CyberPower CPS1500AVR (rack mount) options

[Arjen de Korte]

> This unit works fine with the "cyberpower" driver (which reports it as
> "Unknown model - 15"). It reads this much from it:
>
> battery.charge 		: 	100
> driver.name 		: 	cyberpower
> driver.parameter.pollinterval 	: 	2
> driver.parameter.port 	: 	/dev/ttyS0
> driver.version 		: 	2.2.0-
> driver.version.internal : 	1.00
> input.frequency 	: 	60.2
> input.voltage 		: 	121
> ups.firmware 		: 	5.100
> ups.load 		: 	024
> ups.mfr 		: 	CyberPower
> ups.model 		: 	Unknown model - 15
> ups.status 		: 	OL
> ups.temperature 	: 	36.0
>
> That's of course the driver recommended for it.

Indeed.

> But since it comes with
> PowerPanel Plus for the primary serial port, I thought I'd try the
> powerpanel driver. That looks like this on startup:
>
> CyberPower binary protocol UPS on /dev/ttyS0 detected
> warning: [16] unknown value for [unknown.variable.0x3d]!
> warning: [0] unknown value for [unknown.variable.0x29]!
> warning: [254] unknown value for [unknown.variable.0x2b]!

You can safely ignore the above values. We couldn't figure out what they
mean by looking at just the one UPS we used for deciphering the binary
protocol used.

> And then produces these sometimes-obviously-off values:
>
> battery.charge 		: 	206
> battery.charge.low 	: 	45
> driver.name 		: 	powerpanel
> driver.parameter.pollinterval 	: 	2
> driver.parameter.port 	: 	/dev/ttyS0
> driver.version 		: 	2.2.0-
> driver.version.internal : 	0.22
> input.frequency 	: 	70.8
> input.frequency.nominal : 	60
> input.transfer.high 	: 	147
> input.transfer.low 	: 	88
> input.voltage 		: 	131
> input.voltage.nominal 	: 	120
> output.voltage 		: 	0
> output.voltage.nominal 	: 	120
> ups.beeper.status 	: 	enabled
> ups.coldstart 		: 	disabled
> ups.firmware 		: 	5.100
> ups.load 		: 	11
> ups.mfr 		: 	CyberPower
> ups.model 		: 	OP1500
> ups.serial 		: 	[unknown]
> ups.status 		: 	OL TRIM
> ups.temperature 	: 	158
>
> My guess is this simply has more battery than the 1500AVR-HO model the
> powerplanel driver expects, so the 206% battery charge?

Not at all. The CyberPower binary protocol comes in (at least) two
flavors. The NUT 'cyberpower' driver is for the older one that requires a
wacky conversion of the values read from the UPS. In contrast, the NUT
'powerpanel' driver displays the values as they are received without
conversion. Unfortunately, we currently have no idea when to use which
version of the binary protocol. There must be some way, since the bundled
software can also tell them apart, but so far we have not enough
information to conclusively tell.

> The input frequency and voltage are obviously off. The temperature is way
> off.

We know.

> Would it be simple to add enough to the powerpanel driver code to work
> with this ups?

No, that would break the driver for newer CyberPower UPS'es that 'speak'
the binary protocol.

> Afraid I don't have Windows to see how PowerPanel Plus talks to
> this unit.

In that case, I'm afraid there is nothing you can do to help us improve
this driver.

> But is knowing how far off the input.frequency, input.voltage,
> and ups.temperature are enough to correct somehow to the right values?

No. We already know the conversion functions (from the NUT 'cyberpower'
driver) required to do that. What we're missing is a way how to tell if we
need to use them or not. Running the bundled UPS software and snooping the
communication might help, but if that's not an option there is nothing we
can do. I still have some hope that eventually we would be able to create
one driver for the serially connected CyberPower units though. But that
would require much more feedback/reports from people using these.

Best regards, Arjen
-- 
Eindhoven - The Netherlands
Key fingerprint - 66 4E 03 2C 9D B5 CB 9B  7A FE 7E C1 EE 88 BC 57