HP ZE4805 and old battery

Archived from groups: comp.sys.laptops (More info?)

It sounds more like a bad contact...

<larwe@larwe.com> wrote in message
news:1104674451.653135.130500@z14g2000cwz.googlegroups.com...
>I recently bought a spare battery for my ZE4805 on eBay. It's the same
> HP part# as the original battery shipped with my laptop, but it's about
> a year older and comes from a different OEM (Simplo 916-2310 vs. the
> Dynapack DAK100440-1180 originally shipped with the laptop). The cells
> and gas gauge in it appear to be OK, because it powers the laptop for
> up to 3 hours, and the ACPI capacity info it reports is sensible
> (design capacity 4400mAh, last charged capacity 4224mAh, etc).
>
> However it seems as if the battery periodically loses communication
> with the embedded controller in the laptop. It can be sitting there
> charged, and then the charge LED will go from green to blinking yellow.
> Looking at ACPI when that happens, often I see garbage info. When the
> laptop is turned on, no AC, with that battery, sometimes it thinks the
> battery is empty until I eject and reinsert it.
>
> The symptoms are consistent with the laptop losing communication with
> the battery, and not retrying after one of these upset events until the
> battery is ejected and reinstalled.
>
> I'm wondering if there was something revised in the ZE4805's comms vs.
> earlier machines - e.g. faster bus speed (I'm referring here to the
> battery I2C bus, by the way), different pullup/pulldown requirements on
> the data or clock lines, etc.
>

Archived from groups: comp.sys.laptops (More info?)

larwe@larwe.com wrote:
> I recently bought a spare battery for my ZE4805 on eBay. It's the same
> HP part# as the original battery shipped with my laptop, but it's
> about a year older and comes from a different OEM (Simplo 916-2310
> vs. the Dynapack DAK100440-1180 originally shipped with the laptop).
> The cells and gas gauge in it appear to be OK, because it powers the
> laptop for up to 3 hours, and the ACPI capacity info it reports is
> sensible (design capacity 4400mAh, last charged capacity 4224mAh,
> etc).
>
> However it seems as if the battery periodically loses communication
> with the embedded controller in the laptop. It can be sitting there
> charged, and then the charge LED will go from green to blinking
> yellow. Looking at ACPI when that happens, often I see garbage info.
> When the laptop is turned on, no AC, with that battery, sometimes it
> thinks the battery is empty until I eject and reinsert it.
>
> The symptoms are consistent with the laptop losing communication with
> the battery, and not retrying after one of these upset events until
> the battery is ejected and reinstalled.
>
> I'm wondering if there was something revised in the ZE4805's comms vs.
> earlier machines - e.g. faster bus speed (I'm referring here to the
> battery I2C bus, by the way), different pullup/pulldown requirements
> on the data or clock lines, etc.

I'm curious if removing the ACPI compliant battery in Device Manager and
rebooting when the new battery is installed would have any impact on the
problem. I've never had the opportunity to try this myself, but with
the minor problem you report, there might be a slight parameter problem
that this would correct. OTOH, there is the "law of unintended
consequences" so that given the squirrely nature of replacement
batteries, it could disable the battery completely. Maybe this is the
"if it isn't *completely* broke, don't fix it" situation.

Q

Archived from groups: comp.sys.laptops (More info?)

> It might not be the battery, but some parameter in the Linux boot that
> is responsible (sorry, I didn't see Linux in your original post). I'm

No, no, no... Maybe I didn't explain the situation clearly enough.

1. With original (shipped in the box) battery, there is no problem.
2. The problem can be made to show up when the machine is switched off
(not suspended, but powered down).

It's not an OS problem; it's an issue specific to this particular battery.

--
Here, in a large house, formerly a house of state, lives Mr. Tulkinghorn. It
is let off in sets of chambers now, and in those shrunken fragments of its
greatness, lawyers lie like maggots in nuts.

Archived from groups: comp.sys.laptops (More info?)

>However it seems as if the battery periodically loses
>communication with the embedded controller in the laptop.

Followup on this: I didn't want to open the battery, but the problem
kept recurring so I cracked it today. The layout is a bit unusual - the
connector that goes to the laptop is on a separate PCB, connected to
the controller board via flying leads. I took a close-up picture of
this board which canbe viewed at http://www.larwe.com/pdrm0950.jpg
(large file).

You can see the line marked 'C' (I2C clock) has a poor solder joint (it
looks worse from the other side). I resoldered and the problem appears
to be fixed now.