> I think what you are seeing is an consequence of the fact (as I now understand it) > that the good folks who wrote tripplite_usb had to make it up as they went along.
Yeah, that's right. > They didn't have technical specifications from Tripplite so they had to cleverly > interpret (aka guess) at the correct output; Well, not quite. From a comparison of your data with mine, it's clear the raw data from the units requires either an offset value or a correction factor, or both, in order to obtain the actual voltages present on the input, output, and DC bus. Prior to you kind response, I was unsure whether I simply had a unit which was producing errored raw data, but your unit's data pretty much confirms it is a systemic issue, not an issue with an individual unit. In lieu of documentation from Tripplite, of course, the developers had liitle recourse other than to assume the raw data did not require correction factors, especially since I do not believe they had a unit in hand. > If I'd known how poorly Tripplite supports the nut project I would not have bought > a Tripplite UPS. > I won't buy another one. OK. We might drop them a note. It's possible it could induce them to release their specs to the developers. > Here is my upsc output > battery.charge: 100 > battery.voltage: 14.50 > battery.voltage.nominal: 24 ... > input.voltage: 108.87 > input.voltage.nominal: 120 > output.voltage: 114.0 Clearly, the correction factor and / or offset cannot be the same for the DC voltages as for the AC voltages. I no longer have a variac, and I looked into getting one on ebay, but they are more expensive than I can afford to put into this effort at the moment. I may be able to do some battery testing, however, by simply running a discharge cycle and plotting the points one for one. I may be able to rig something to vary the input voltage, as well. _______________________________________________ Nut-upsuser mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
