[Nut-upsdev] Driver for Cyberpower PR2200

[russ at cb-1.com]

Quoting Doug Reynolds <mav at wastegate.net>:

> Russ Romano wrote:
>
>> Is there anything I can do to facilitate the further development of  
>>  the "cyberpower" driver to cover my model?  I could probably work   
>> around it and hack it for myself... but if I'm going to put the   
>> energy into fixing it, I'd rather do it in the framework of the   
>> overall development so others can have it as well.  Or perhaps it   
>> is already being addressed by someone else.
>
> After looking at the specs for your ups, your ups is more than likely
> supported by the powerpanel ups driver in smart mode and in contact
> closure mode with the genericups driver using "upstype =7".
>
> Please try the powerpanel driver, if it works, we can list it as a
> supported device.  If not, perhaps we can help you out.

I tried the powerpanel driver and it does not work.  It says that  
there is no cyberpower text mode UPS found, or something to that effect.

The cyberpower driver certainly communicates to the UPS well enough  
(so it seems to be talking the right protocol) but it mis-interprets  
the data fields being sent back.

For example... The OEM software detects the UPS status as follows:
utility state : normal
charger state : bypass
battery state : normal
input voltage : 112 volts
output voltage : 112 volts
AVR level : normal
battery capacity : 100 %
load level : 29 %
frequency : 60.0 Hz
temperature : 39 C
backup time : 34 minute 16 seconds.

NUT using the cyberpower driver reports the status as:
battery.charge: 000
driver.name: cyberpower
driver.parameter.port: /dev/ttyS1
driver.version: 2.0.5
driver.version.internal: 1.00
input.frequency: 60.0
input.voltage: 111
ups.firmware: 1.100
ups.load: 064
ups.mfr: CyberPower
ups.model: Unknown model - 22
ups.status: OL
ups.temperature: 7.0

When I pull the plug to the UPS, the status changes to OB, and the  
input voltage  drops to 000.  So... it accurately detects the OL/OB  
status, the input frequency & line voltage, but reports bogus data for  
battery charge, load and temperature.  This could be as simple as  
decoding what fields mean what in the data sent back to the cyberpower  
driver.  I'd be happy to help with that in any way I can.

Is there a debug mode we can put the cyberpower driver in to log what  
it's sending and receiving to help the process?

Thanks!
-Russ