From owner-freebsd-acpi@FreeBSD.ORG Fri Mar 19 14:44:29 2010 Return-Path: Delivered-To: freebsd-acpi@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1575F106566B for ; Fri, 19 Mar 2010 14:44:29 +0000 (UTC) (envelope-from j@uriah.heep.sax.de) Received: from uriah.heep.sax.de (uriah.heep.sax.de [213.240.137.9]) by mx1.freebsd.org (Postfix) with ESMTP id A150C8FC20 for ; Fri, 19 Mar 2010 14:44:28 +0000 (UTC) Received: by uriah.heep.sax.de (Postfix, from userid 107) id B672E1C; Fri, 19 Mar 2010 15:44:26 +0100 (MET) Date: Fri, 19 Mar 2010 15:44:26 +0100 From: Joerg Wunsch To: freebsd-acpi@freebsd.org Message-ID: <20100319144426.GV52442@uriah.heep.sax.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100319233643.G85436@sola.nimnet.asn.au> <20100320002156.L85436@sola.nimnet.asn.au> <1269000185.47240.28.camel@RabbitsDen> <4BA34277.1000509@obluda.cz> <20100319144604.N85436@sola.nimnet.asn.au> <20100319155246.L85436@sola.nimnet.asn.au> <20100319155246.L85436@sola.nimnet.asn.au> <4BA2EFE8.60004@obluda.cz> X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E X-GPG-Fingerprint: 5E84 F980 C3CA FD4B B584 1070 F48C A81B 69A8 5873 User-Agent: Mutt/1.5.20 (2009-06-14) Subject: Re: Funny battery values (nx6325) X-BeenThere: freebsd-acpi@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: Joerg Wunsch List-Id: ACPI and power management development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Mar 2010 14:44:29 -0000 Thanks for all the suggestions so far, I'm writing a mass-followup to everyone. As Dan Lukes wrote: > And it's filled in C1AC() method. The Design/Last Full Capacity is > filled here: > Store (C164, Local0) > Store (Local0, Index (DerefOf (Index (C1AF, Arg0)), 0x01)) > Store (Local0, Index (DerefOf (Index (C1AF, Arg0)), 0x02)) > > It mean that both numbers are same and come from C164 variable. Ah, that explains the observed behaviour, I think! (See below.) As Ian Smith wrote: > > Out of sheer curiosity... what does acpiconf -i1 say? > > It should't be there, but .. I wondered about that too :) It's available as an entry, but claimed to be "not present" (which is correct). I guess there's an option to stuff a second battery pack into the DVD drive bay, I didn't check HP's website for that. That would at least explain the sheer existance of the second entry. [About re-calibrating the battery pack] > Yes, been through all that. I drained it from the BIOS setup screen > rather than with an OS running, until the power button won't respond > at all. I drained my TP600 battery with FreeBSD running single-user, until the computer eventually turned off. Pinging the ethernet IF remotely, I could monitor by which time it stopped responding. After all, I've got a battery back to 1.5 h lifetime, which is plenty for that machine. > Pretty sure it's done for .. possibly what Joerg experienced, with > at least one cell shot. In my case, the battery was so severely dead, it did not accept any charge, and the TP600 could not run from that battery at all. > I'm really hoping > the charging circuit is ok; even at 8.6V charging 'present rate' is 0. It might simply refuse to charge due to the too low pack voltage. I wouldn't assume you can kill the charging circuit that way. What eventually surprised me is that it seems almost all laptops use some kind of de-facto standard 18650 cylindrical cells, usually arranged in N pairs of cells, with N being either 3 (10.8 V battery pack) or 4 (14.8 V battery pack). I bought 5 identical-type old laptop battery packs in the bay, in order to just use the cells for my ham radio transceiver. This got me 40 18650 cells, and I've been using 16 out of those for the transceiver (these were all in really good shape). One battery pack suffered from a dead cell pair, so this left me with another 3x2 cells in good shape -- which did now populate the old battery pack of that TP600E. ;-) The other 16 cells are also still usable for tinkering, but cannot deliver the higher currents needed for laptops or my ham radio rig, though they would still do pretty well for low-current consumers (general electronics). So bottom line, the good news is that you can actually replace those cells in dead battery packs. :) As Alexandre Sunny Kovalenko wrote: > But this is likely the check for the presence of the second battery: Yes, apparently, as the second batter is always (correctly) announced as being not present. As Ian Smith wrote: > Can you / anyone suggest a good basic tutorial for ASL/AML? The last > time I tried following it from the ACPI specs I nearly drowned :) I guess I'm also in need for one... ;-) > > Still curious, though. > Mmm. It still seems to come down to the wrong Design Capacity > (equals Lastfull Capacity) being reported either by the battery > itself, or being miscalculated by the EC. This value - still > something like 1/18 of the expected capacity - is then propagated to > the 5% and 1% values. I've had a look at the battery pack now, it is declared as 10.8 V, 55 Wh. This would correspond to 5093 mAh. However, now that I know the "design" and "last full" capacity are always set as identical by the BIOS, I'm no longer *that* surprised about the reported value. This machine has been essentially running from mains power for years now, so I don't know when it actually might have been able to calculate real values for "last full capacity" for the last time. > Joerg, so how long does it really run on battery? I'll try that as soon as possible (the machine is the production-level machine of my wife...), and I'll use the single-user method so the battery can actually drain down until it's really empty. Btw., as someone suggested to tinker with the ACPI OS name, I gave that a try, too. When using "Microsoft Windows", it claims I didn't have a battery at all :-o, for all other values (like "Microsoft Windows NT", or "Microsoft Windows 2006") it reports the same 279 mAh as when not modifying the OS name at all. -- cheers, J"org .-.-. --... ...-- -.. . DL8DTL http://www.sax.de/~joerg/ NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)