M6809 on battery sometimes just shuts down - Page 4

That's for your wireless card. If set to enable, your wireless card will theoretically always be on when the system is powered off. Disabled = powered off, restore = return to last state it was in the last time the laptop was shut down. In reality, it does whatever the heck it wants to.

Think about how close to boiling that is. Your laptop would be melting at those temperatures, and if you were touching it, your skin would fare worse. Either Sandra isn't reading it correctly, or you have an out of whack sensor.

Sets the start up state for the wi-fi. off and on are pretty easy, auto just returns it to the previous state.

Thanks for the info...

BTW, tonight I had a shutdown that started as I was hitting the Delete button to boot to the bios from a cold start...

Ok, things are getting interesting. Put the laptop to sleep at work yesterday. Later I opened it in battery power, and the fans kicked in and shut down. There has to be a link between the battery and this issue. This is the first time it has happened to me in x64.

I don't think so, seeing as I use my laptop about 95% AC and I have it occur almost weekly. It has to be tied to the temperature sensor because it shuts down whether hot or cold...

I've had this same problem with my m6805 as well. However, for the most part my laptop is stationary so my laptop is always plugged in. Here's what happens for me:
I go to sleep at night listening to music from my laptop, then I let the computer put itself to sleep after 30 minutes of inactivity (basically after winamp is killed). Then when I wake up in the morning I'll turn the laptop on and after about 2 seconds, full blast from the fan and then 5 seconds later power off. Luckily I got the bb warranty, so I might take it to them if I ever decide I want to go without my laptop for a couple weeks.
So the specs on my system are the 0F08 bios (I never noticed the problem before the update, but I didn't notice it right after either) and I also added a gig of Kingston memory. I tested that with memtest 86 for hours with no problems. I'm still running winxp home sp1, no DEP, kerio firewall, and I don't have the wireless radio on when I wake the system.
After reading everyone's post, I'm thinking it is what I had thought it was, and that'd be just a sensor that freaks out and makes the computer turn off. I suppose I should get something to monitor my sensors (I don't think mbm works on the laptop) and check temps if I can when this happens next.
Still though, it is nice knowing that others have this problem too. So when I do take this to bb (right before the warranty is up, if not sooner) and ask em to fix all the problems or replace then I'll have a little bit more leverage.

Ok, let me ask you guys this. Have any of you had this issue happen after getting a BRAND NEW battery (and I don't mean that your battery says it's fully charged.)

I ask, because I'm sitting here at JavaOne, had my old original battery in the machine, Windows was showing it as being at 33% power left, but it suddenly powered off. This is the first this has ever happened. I replacd the battery with my new spare and it's working fine.

I suspect your batteries are just totally shot. Batteries have a tendancy to put out power rather dramatically once they're drained. I've found that the power meter stuff built into batteries RARELY are accurate. As far as it happening under AC, I bet your battery isn't holding a charge and so it also causes the cut out because it's telling the recharge circuit "I'm out, boom" and the power supply tries to feed the battery BEFORE the system.

Anyhow, just a thought. Try a new battery and see if that fixes things at all for you.

--Mike

Sounds like a bad power supply or bad solder at the terminals where you insert the battery. This is plausable since your battery isn't being recharged at times. If you can run the laptop without a battery and just the power connected then the terminals are even more suspect. You can definitely rule out the battery being the culprit.

Well, I spoke too soon. About 15 minutes after I sent that post it happened again on the new battery that was fully charged. Hasn't happened since. However, this is the first time that I've done a lot of putting the machine into and out of suspend mode and hybernation. I wonder if maybe that's the culprit. I did a full reboot and it *seems* to be working fine now... Gonna have to keep an eye on it. Normally I don't work off battery very often so this week will be a big stress test...

--Mike

Well I am also experiencing a shutdown problem. Mine has just started within the last week. It has happend 3-4 times over night. I wake up and it's off. Then it happend this morning when I was listening to music and encoding some audio files. The only thing that I can tell is different is that it gets hotter more often but the fan is on more as well. Everytime it has happened it has been on AC power, so no battery issue.

Is there any type of temp. monitoring software I can run to see what happens during the shutdown? Something that would log the temps.

I'm thinking about sending it to BB to have them check it out.

Has anyone ever had this happen on NS.??.?? version bios's? I heard the mainboard is slightly different and maybe this is why they went to the NS boards. This may be a motherboard design problem that when it ages certain components burn out causing the power circuit to work incorrectly. Again just a guess.

Update since my reply a month ago.

Shortly after posting, I took the unit apart and looked for problems. There was a little dust, but not much. Heatsink seemed a little loose on the CPU, but could be normal. I usually crank mine down pretty good so I'm not used to any play.

Cleaned out the little dust that was inside. Cleaned off CPU/heatsink interface and applied AS5. Reinstalled heatsink and cranked it down firm.

It's run 24/7 for the past month with no problems. Then tonight I'm surfing the web and it shuts down. I restart it, and about 10 minutes later it shuts down again. Now it's been running about 30 minutes or so. This is all on AC power, stock speeds.

It seems like whatever it is gets worse with time. Each time I've made a corrective action (first time was reverting from overclocked to stock speed, second time was applying AS5), it seems to fix the problem for a length of time, but it eventually returns. At this point it seems like the only option left is to deal with it until it gets unbearable, then eBay it for parts and get a new laptop.

Makes me regret purchasing eMachines. Seemed like a great deal at the time. First problem was the cracked hinges. Got replaced under warranty, now the right hinge has cracked again. And this thing is BABIED. Rarely ever leaves the AirDesk, and even then it's to occassionally roam upstairs. Now this problem, and it's no longer under warranty.

As a rule, I never buy any type of ESP on anything, and have 'saved' thousands of dollars over the years and consistently turning it down. Maybe I'll start making an exception for laptops. First one I owned did OK, lasted about 3 years. Second one lasted about a year and a half, then had a problem with the video card. Parted it out and bought the M6805 about a year and a half ago, and it's already unreliable.

This is one of the reasons why I haven't bought a packaged desktop in about a decade... you can't get to any of the BIOS options unless you build your own. This could probably be solved by raising the setpoint or just turning off the auto shutdown on high temp option. Could potentially lead to a fried CPU, but then again, at least I wouldn't have to worry about this problem anymore.

That's rough, man. The last OEM desktop I owned was a Gateway P-166. After support had me get inside the thing a few times to fix various problems, I realized how easy the stuff was to work on, and even identified a few things I considered to be design flaws. I've whiteboxed every desktop I've ever owned since then and have been happy doing it. If something goes bad, I chuck it and have a replacement on the way the next day.

When I was buying my 6809 and the BB rep started talking about the ESP, it was my wife who jumped in and said, "We don't need that, he can fix anything!" Now, bless her heart, my wife thinks I hung the moon. She is the one person in the world I try never to look like a chump in front of. But, I had to look her in the eye and say, "Honey, if this breaks physically, I won't be able to fix it." The look in her eye and the little "oh" that slipped out almost broke my heart.

But, I'm glad I went with the ESP. My wife is too, so I guess I proved that by being humble I was still able to be right. I'm sorry for you man, I hope there is a good solution for this. Just wanted to comment, I thought it was odd you and I had such a similar story but made such a dramatic difference in decisions.

Again, from my experience, I don't think this is a temperature problem. This is a sensor problem (software or hardware).

I would love to know what that shutdown circuit looks like... And where is this temperature sensor located??? Do mobile Athlon 64's have a built-in temperature sensor?

An interesting thing happened the other day... Out of the blue my fan came on full blast... I started to wait for the laptop to power off (I know the routine well) and after about a second, the fan turned off and the laptop stayed on... It usually powers off after about 3 seconds of the fan blowing on high. I didn't do anything out of the ordinary to cause the fan to go on or to stop the laptop from turning off. Weird...

Mark

Hello all,
just wanted to chime in with another case of the shutdown problem. I'm experiencing random shutdowns (no discernible pattern to when, just whenever it feels like it apparently) on AC power. Haven't run on battery for more than a few minutes since I returned from school, so can't tell if it happens on that too.
Anyway, like the poster above me said, I'm almost positive this isn't a temperature problem. I already suffered through overheating (and the automatic shutdowns from it that occured just like this, only accompanied by actual overheating) and fixed that by swapping the thermal pad with some artic silver and voiding whatever warranty I had left at the time in about fifteen different ways. After that procedure, everything was running fine (except for the flicker problem) until recently, when I've started getting random shutdowns, just like when I used to overheat - but there are no other signs of overheating, so I'm almost positive that can't be it.
Faulty temperature sensors sound like they may be the real problem, but I'm not that good and don't have a way to figure that out. Until someone comes up with what sounds like is probably a software (or BIOS?) solution, if there is one, I think I'm SOL. Next time I'll say screw it to laptops and just build a SFF myself.

(Notes: running WinXP Pro SP2 on 0F08 BIOS. Standard clocks on CPU, overclocked core/mem on the Radeon chip but standard clocks when not in 3d, and shutdown still occurs. 768 mb RAM, everything else stock minus some AS5. About to tear out my DVD burner too due to it getting finicky about taking media types. Absolutely latest drivers for everything, with exception of Broadcomm drivers. Anyone out there tried that alternate NSF04 BIOS?)

This all sounds like memory mismatched or other hardware problem. Try taking out your new memory, DVD drive and wireless card and see if the same things happen. Just run it on the internal DIMM. Just a guess but worth the trial.

Hmm, suppose it could be the memory - but the SODIMM module I got was Transcend, which it seems like most people have had success with. But it's worth a shot - would this be something a memtest program could turn up, or just trial and error?

By the way - hinge on my friend's M6809 just exploded today. Left hinge just fell apart, lid no longer seats properly. Methinks I might try and reinforce those cracked hinges on mine now.

The only way to know is to remove all three parts and see which one is the culprit. If the laptop runs the same way with shutdowns then I would reinstall the OS. If it still happens then you probably have a bad motherboard component. Memtest will tell you alot too. Go ahead and run it first.

It's not software

Having been plagued by this problem for about 3 weeks (and now using W98 on an old machine as a result) I am 100% certain this is not a software or virus issue. I have an M6807.

I decided to restore an system backup from April, way before the problem started. I put in a different hard drive (so no boot sector virus) and booted my Acronis True Image CD (so no boot sector virus, and because it runs Linux, no other viruses and most certainly not the same drivers or anything else). Part way through the restore, the fans go to flat out and five seconds later, it shuts off. Yup, exactly like it's been doing when booted from XP or Linux on the hard drive. I start Acronis True Image again, and part way through the restore, it does it again.

I've cleaned out this machine's heatsinks, it sits on a Targa fan pad, and currently it has a Vornado fan pointed at it. This is not some regular overheating problem then!

Since the problem started I flashed the official eMachines BIOS update into it, so that should eliminate corrupt BIOS settings too.

Now comes the interesting bit. In an endeavour to reduce heat still further, I took off the RAM door and removed my extra 1GB, also the LAN card door, the SD slot filler and the CF slot filler, and tried True Image again. This time it restored ok. (About 2 hours in the battery light started flashing: guess who'd forgotten to plug it in? I quickly plugged it in and all was ok.)

The old restored XP is now running and has been for a while. I'm now wondering if the RAM has something to do with it. It is the dreaded Kingmax, which seemed to be working perfectly (and Memtest is happy with), but now I'm beginning to wonder.

(Note that I have subsequently found it occurs even without the KingMax memory installed.)

I'll leave it on for a while and see what happens. If it stays alive I'll try restoring my most recent backup and see if that's reliable with just 256MB (I must go find the other original 256MB I've stashed away somewhere).

Graham.