[Nut-upsdev] REGRESSION: New "megatec" driver does't work for UPS that was managed by old fentonups driver

Carlos Rodrigues carlos.efr at mail.telepac.pt
Mon May 19 16:47:28 UTC 2008


On Mon, May 19, 2008 at 4:33 PM, Michel Bouissou
<michel-nut at bouissou.net> wrote:
> Well, a double-scale gives a rather good guess, but with the UPSes I've
> observed, a single scale is plain useless. For the true-life example I
> gave, the same 12V value means "fully charged" when on battery and
> "empty" when on AC...

The problems with the battery charge calculation have been discussed
before in this list, and I'm starting to agree with (I don't remember
exactly who, sorry) who tried to convice me that having any kind of
charge calculation was a bad idea.

I'm thinking if the charge calculation shouldn't be disabled by default...

See, as simple as it is, the currenty way at least gives the user who
can't read voltages some ideia of how charged is the UPS. At least, if
the UPS happens to be on the known list. Complicating the calculation
by adding more values to work with gives little benefit because the
users would have to obtain these values themselves, and it is already
difficult for them to obtain max:min now.

-- 
Carlos Rodrigues



More information about the Nut-upsdev mailing list