I was reading an article on JKontherun, and noticed something on the comment section.

Posted by: frank j | May 08, 2006 at 09:46 AM

JK responded:

Posted by: jk |

OK, what is going on? Why is JK getting 2 hours with a prototype, and none of the people who have received EOs, are getting any more then 1:25. Some reviewer I read, under heavy usage was only getting an hour!!!

Is there a logical explanation for this?

Back in his reviews of the device, did JK ever mention the heat issue?  I wonder if the two issues could be related.  If something in the produciton units is causing them to generate a lot more heat than the prototypes did, could it be causing the the fan to work harder and drain the battery faster?

-Jeff

I'm a bit suspicious of these battery life tests so far.

In the detailed testing i've done with Ctitanic we see two problems.
1 - We cant get the CPU to go into C-states. (sleep states) when idle. We need to test this more though
2 - His battery drain appears to be higher than a 1.5ghz VIA C7-M 12" laptop that Van Smith tested with on ViaArena. Thats not right!

Will post updated info on the Carrypad Journal as we go on with the testing and if we find the answer.

I've also opened a ticket with TabletKiosk's technical support team and they sent back a note yesterday afternoon that they are investigating the battery life issues and will get back to me soon.

-Jeff

Good call.
Asia is in bed by now I guess so it could be a few hours before you get an answer.
Good luck with that route.
Steve.

I was also thinking that the processor might not be speed-stepping correctly and/or using the various power moding states when it's suppose to. I was able to get 1hr40min with WLAN/BT off, screen at 20-25%, power setting to MAX Battery and reading Zinio magazine and playing Solitaire only.

Intel has a software package to look into the speed stepping of it's processors but I can't find anything like that for VIA??

I guess it's possible there is a firmware / driver issue causing this?  Or maybe there was a PCBoard or component change between the pre-production and production units and there is a component inadvertently being placed into a high-power drain mode. (seen this on controls we design at times).

I hope we figure this out... otherwise I might have to look at the LS800 instead (but at a much higher cost). In the current very limited battery life of the eo, it's not really able to do what I needed it to. ;-(

Just posted an update at http://www.carrypad.com/journal.

We're working hard on it!

Steve.

Glad you are keeping tabs via your journal.

Don't forget that our VIA CPUs are only the 3.5W - 5.0W flavor: http://www.via.com.tw/en/products/processors/c7-m_ulv/index.jsp 

I design embedded electronic controls for a living (with emphasis on power electronics), so I will help out if/where I can.

BTW... how were you able to determine that the speed stepping was actually taking place?

I hope we get this figured out... otherwise I might have to send the eo back for an LS800. :-(

 

Thanks for your offer Mr_g@dgeTT.
If you have any info on 7" LCD panel power utilisation (including controller) it might help. One of the suspicions is that there's a cheap/old/heavyweight LCD part in there. Whats the max it could take in terms of power? I think 10-12 watts. I'm hoping its going to be around the 3-5W mark though.

We determined speed-stepping by using COSBI Cpu speed 1.1 (results shown in the Origamiportal topic.

Regards
Steve.

I just forwarded a link to this thread and your blog posts to the support person at TabletKiosk that responded to my support request.  Hopefully they will take a look at the information that is out there now.

Please feel free to let me know if there are any tests you want me to run on my eo or if I can help out in any other way.  I admit, though, that power consumption isn't really my area of expertise.  I am a software engineer, though, so I should be able to at least follow technical instructions. :)

-Jeff

I'm available too if you guys need any additional testing. One thing I want to do is drain/recharge the battery in BIOS screen loaded only (or DOS) two times to see if it is a calibration issue.

//TB

docbliny.

Thats a good idea. I'm also wondering if the battery needs conditioning or if the BIOS is set up with the wrong battery power information. I'll be honest. I don't know how the process of determining a batteries remaining capacity works.

Post an update when you can.

Regards
Steve.

They probably have an IC (intergrated circuit) performing coloumb charge counting and then the processor reads the counts over one of the interface/data buses.

My guess is there is a very small resistance "current sense" resistor in series with the battery. By taking a differential voltage measurement across the sensor resistor they can calculate the current flowing out/in to the battery at that instant. I would think they sample this instantaneous current many times a second to determine average loading and thus predict remaining battery life, etc.

There are some other tricks to calibrate, but I won't go into here.

I actually ran the eo until is shut itself off... however I was able to power it back on after it shut off. I haven't determined if Windows issues the command to shutdown, if the BIOS issued the shutdown command, or the battery voltage fell off and placed the digital core into reset which powered everything off.

JeffGr:

I've also opened a ticket with TabletKiosk's technical support team and they sent back a note yesterday afternoon that they are investigating the battery life issues and will get back to me soon.

-Jeff

Please. keep me informed.

JeffGr:

I just forwarded a link to this thread and your blog posts to the support person at TabletKiosk that responded to my support request.  Hopefully they will take a look at the information that is out there now.

Please feel free to let me know if there are any tests you want me to run on my eo or if I can help out in any other way.  I admit, though, that power consumption isn't really my area of expertise.  I am a software engineer, though, so I should be able to at least follow technical instructions. :)

-Jeff

Jeff, here is the link that you should give  them. This forum is very hard to track  so I'm posting more technical stuff there.

http://www.origamiportal.com/modules/newbb/viewtopic.php?topic_id=334&start=20#forumpost1809