HP Touchsmart 600-1055 Power Loss Issues.

Hello.

Over a month ago, I created a thread on the HP support forums, and have gotten very little assistance with it. Needless to say, I've gotten pretty miffed at the total lack of zero support from the company, since my warranty expired.

I'm not sure if I should just link the thread explaining my problem or C/P it here.

I'll just do both, link and C/P it:

http://h30434.www3.hp.com/t5/TouchS...-Shutdown-and-one-other-extremely/td-p/712077

I purchased a brand new HP Touchsmart 600-1055 All-In-One PC from Amazon last year.

The warranty expired this February, and in the last month, I've had the unfortunate luck of dealing with an annoying problem.



The computer will just shutdown randomly, usually once every day or two. Pressing the power button will not boot the computer until I have unplugged the powerbrick, and plugged it back in.



My first thought is that it is over heating, and being shut off to protect components, so I downloaded and installed Speed Fan, a free utility that monitors the temperatures of the CPU, GPU, Cores, and other things. According to the program, during peak usage, none of the components go above 58-60 degrees celsius. During normal web browsing usage, mid 30's to mid 40's is common such as below.

(Picture in the thread)

I've had the system shutdown during both instances though, be it gaming or web browsing, which leads me to believe it is not an over heating issue. Can I get some assistance as to what can be causing this? I've run a full system scan with MalwareBytes and Avira Antivirus, both of which have turned up nothing, so it's not virus related, AFAIK.



The other minor bug is what I thought was a dead pixel, but it's not. This is difficult to describe, but I hope this is sufficient:

Near the top of the monitor, maybe five inches from the right side of the screen, there is a single blue pixel. However, the only time that blue pixel appears is when there is is bright color on a certain part of the screen, and it's not the same area, but about six inches below that blue pixel.



I have to jury rig an image since "dead" pixels don't show up in screen shots.

(Picture in the other thread)

It doesn't bother me, but if there is a fix for that as well, that would be nice.



Thank you, in advanced for any help.

-------------------------
Further in the thread, I gave this information:

While I appreciate the assistance, I'm still getting the power-offs randomly. Even when doing nothing but sitting at the user log in screen (if I locked it to step away), at the desktop, browsing the internet, or playing games.



I have not pinned down what is causing it, and I'd like some support from some of the people at HP. In a way, I'm not surprised. Products these days are usually designed to work fine until just after a warranty expires, so they don't have to deal with it. :/



I've updated drivers (biting my nails the whole time for fear of a power off in the middle of things.)

I'm terrified of defragging for fear of a power loss and totally screwing up the computer, and the same goes for windows updates.



I have a ton of stuff on my computer, and I really do not want to reset it to factory defaults unless it's absolutely required.



Some actual customer support would be wonderful right now.

------


And this:


Using Windows 7 Home Premium 64 Bit.

No viruses, malware, or rootkits detected. I've never had a problem with any of those as I've always been a careful web user, but I check regularly anyway. The Kaspersky thing revealed nothing either.



Power settings are set to high performance, and I'm still getting completely random powerdowns. They're sudden, as if power has been lost to the system. The light on the power brick goes out a few seconds after the computer does. I have to either unplug the brick from the power strip or just turn off the strip and turn it back on. The computer will then restart. Since it seems totally random when it happens, I don't know how booting in safe mode will help any (I figure that's the next suggestion.)



I've gone for a few days without power loss to a loss within an hour after the last. Vents are clean, fans sound like they're running fine. Temp monitors show that things rarely go above 55 degrees celsius, so it's not overheating (I looked the cores up and those are rated to 90 degrees celsius).



What all could cause this? Perhaps it could help me with troubleshooting. Lemme summarise the problem once more:



Computer powers down suddenly, without warning. Sitting on desktop, watching a DVD, or playing a game have all happened during shutoff.

Behaves as if power is cut to the system. No BSOD, no errors, just off.

To reboot, I must unplug brick from power source or turn of surge protector and turn it back on.

I've used a different surge protector. No help.

I've moved it to another room. No help.

I took it to a friend's house when I stayed for a weekend. Happened there too.

System was purchased on Feb 2010, I think.

Warranty expired Feb 2011.

No malware, spyware, rootkits, bootkits, etc. detected on system.

No overheating.


A C/P of the Event Log at the times of the power offs:


Error 7/14/2011 16:26:22 Dhcp-Client 1001 Address Configuration State Event
Error 7/14/2011 16:26:19 Dhcp-Client 1001 Address Configuration State Event
Error 7/14/2011 16:24:36 CAPI2 4107 None
Error 7/14/2011 16:24:36 CAPI2 4107 None
Warning 7/14/2011 16:24:21 Wininit 11 None
Error 7/14/2011 16:24:19 Eventlog 1101 Event processing
Error 7/14/2011 16:24:16 EventLog 6008 None
Critical 7/14/2011 16:24:03 Kernel-Power 41 (63)
Error 7/14/2011 15:12:01 Dhcp-Client 1001 Address Configuration State Event
Error 7/14/2011 15:11:57 Dhcp-Client 1001 Address Configuration State Event
Warning 7/14/2011 15:10:06 Wininit 11 None
Error 7/14/2011 15:10:06 CAPI2 4107 None
Error 7/14/2011 15:10:06 CAPI2 4107 None
Warning 7/14/2011 15:09:24 WLAN-AutoConfig 10002 None
Warning 7/14/2011 15:09:24 WLAN-AutoConfig 4001 None
Error 7/14/2011 14:48:16 Dhcp-Client 1001 Address Configuration State Event
Error 7/14/2011 14:48:15 Dhcp-Client 1001 Address Configuration State Event
Error 7/14/2011 14:46:28 CAPI2 4107 None
Error 7/14/2011 14:46:28 CAPI2 4107 None
Warning 7/14/2011 14:46:22 Wininit 11 None
Error 7/14/2011 14:46:15 Eventlog 1101 Event processing
Error 7/14/2011 14:46:13 EventLog 6008 None
Critical 7/14/2011 14:46:06 Kernel-Power 41 (63)
Error 7/14/2011 13:10:17 Dhcp-Client 1001 Address Configuration State Event
Error 7/14/2011 13:10:16 Dhcp-Client 1001 Address Configuration State Event
Warning 7/14/2011 13:08:35 DNS Client Events 1014 None
Error 7/14/2011 13:08:26 CAPI2 4107 None
Error 7/14/2011 13:08:26 CAPI2 4107 None
Warning 7/14/2011 13:08:21 Wininit 11 None
Error 7/14/2011 13:08:14 Eventlog 1101 Event processing
Error 7/14/2011 13:08:12 EventLog 6008 None
Critical 7/14/2011 13:08:03 Kernel-Power 41 (63)
Error 7/14/2011 10:46:49 Dhcp-Client 1001 Address Configuration State Event


Maybe I can get some assistance here since there's not much there.

 

I have never used one of these PCs but I would speculate its either Power Supply or RAM related as a first stab. Try testing your RAMwith Memtest86 to start with and see what that does.

I would also try using a Ubuntu live CD to rule out any software issues.

I assume that it has an external Power Supply like a laptop? This can be replaced but they are very expensive to test with. If you can get one from somewhere that will let you return it if it doesnt work then It may be worth giving that a go.

Are you UK or US or other based?

 

I tested RAM with the Windows Memory Diagnostic Tool. Is that the same as Memtest86? The results of that was fine. No problems were found.


As for Ubuntu, I've heard of it but never used it. A free OS? That doesn't even sound... safe to use. I wouldn't know where to begin on using it and whether it would completely erase my Win 7 operating system.

The way it's set up, there is a cable that runs from the outlet to the large power brick. That cable plugs into the power brick which runs into the back of the system. It's built very much like a laptop, since it's an all in one PC.

http://www.google.com/products/cata...a=X&ei=PIUlTp3CN4Xx0gGjgb31Cg&ved=0CG8Q8wIwAQ

All the components are behind the monitor. There's not even any documentation on how to open it, since I guess it's not designed to be, for me to clean it out if it needs it. Vents are clear, and have always been clear of debris and dust, but it may have accumulated inside over the year.

United States based.

 

After a pass which took over 26 minutes, it detected no errors.

 

I reckon an OS seemed more major than an anti-virus is all. Don't expect those to be free, y'know?

Anyways, it says 32Bit Ubuntu is recommended, but should I get 64 bit since I have a 64 bit system?

 

Where did you see this? Either should work fine. I probably suggest 64 bit. but I dont think it matters

 

http://i.imgur.com/prbmi.jpg


But I'm thinking it may be a hardware problem, and not OS related. Main reason I think so is that it shut off earlier before the windows logo even appeared. Maybe three seconds after powering it up.

 

Sorry for bump but it won't let me edit:

Also, how hot should a power brick get and how long should it take to cool down? I notice that the brick is extremely hot when I unplug and replug it back in. Not hot enough I can't handle it, but close.

 

Power supplies can get pretty hot. That said if it is hardware related my money is on the power supply.

 

So is the power supply that needs to be replaced (if that is the case), the large brick on the outside? Or is that an inside component that would require disassembly?

I'm sorry, but I really don't know a lot about computer components.
One person I spoke to said the PSU is inside the computer, where the power cord plugs into it. Yet you say power supplies get hot to the touch, when I referenced the brick.

 

you only have one power supply. In a desktop PC you have a box on the inside of the case which converts your mains power down to 12v and 5v. In the case laptops and yours the "brick" has the transformer inside the brick. Therefore the "brick" does the same thing as an internal PSU.

If you have a desktop PC then the person would be right and you would need to replace the box in the back of the PC. However you should just need to replace the brick.

 

What sort of problem could it be? Is it something that could be repaired myself or is it likely a thing that would need replacing?

4 screws hold it together, one under each rubber grip pad, but I haven't opened it and I don't think it'd be a dust and hair problem as it seems tightly sealed.

 

Thank you for your assistance. I'll avoid using this computer unless I must, and try to buy a new adapter whenever I can scrounge up the funds for it.

 

personally I would still run Ubuntu for a while. Im not convinced that it is harware as memtest ran stable. If Ubuntu runs fine then you know that it is likely to be either driver or OS based. ie rule out all the free fixes before you spend money.

New power cords are not cheap due to their high power output

 

Hi Plodr,

I am inclined to agree that the power brick is not the problem, however what concerns me is that Memtest was able to run successfully. If there was a hardware fault I would expect it to power off during memtest too. If, as suspected the problem is in the wiring in the back of the PC then I suspect there is no fix? Whats your thoughts?

 

I think they are fair points. I have had a dodgey power connector on my Toshiba notebook but because you have a battery the system wont power off when the connection breaks

Might be worth OP wiggling the cable on the back to see if it is a bad connection?

There are usually two ways the power socket can be fitted. Either the socket is soldered directly onto the motherboard or it is on a cable that plugs onto the MB.

My Toshiba was on a cable but the replacement I bought had the wrong connector so I had to rewire the connector from the old plug on to the new one.

Not too sure what OP will be able to do if it is a bad connector. Will depend how it is wired and how brave OP is feeling.

 

No bad connections.

Wiggled the cable where it enters the PC, where it connects to the power adapter, and where it plugs in at the wall. Nothing powered off.

 

It's possible, and I'm thinking my temp monitor (I've used SpeedFan and Core Temp, both of which have had similar results.) has been inaccurate.

The following is a screenshot of CPUID Hardware Monitor's readout:

http://i.imgur.com/pzzkU.jpg

It's very hot here, but it's hot everyday, lately. My room, according to a digital thermometer is 33.4C / 92.1F. I also don't remember it being this hot back in May, when this first started happening. And it's also happened in the morning before it got this hot, and once within 10 seconds of startup.

 

Try setting it lower and see if that helps.

Many HP users are complaining about the heat issue whether it's PC's, laptops, all in ones. Here's a link you may want to see. There's a hardware fix (metal plate) but it doesn't state it in detail in how to do it. http://h30434.www3.hp.com/t5/TX-TM-...ting-and-shutting-down-constantly/td-p/189955

PS: Just yesterday I was looking and very interested about the newest HP 610q series but since I read your thread and googled info about complaints about HP, I won't even come near a HP. So THANKS!!

 

your 12v rail is only receiving 10.5v This means that you are not getting your full power on your 12v rail (your about 15% down) This could cause problems as the load increases your internal components aren't getting enough juice so the system might be powering off.

What are the thoughts of others on this?

 

I really don't know which to believe, if any, whatever these numbers may mean.

Speedfan 4.44 says:

3Vsb= 3.34V
Vbat= 0.99V
Vcore= 1.51V
+12V= 9.98V
AVcc= 3.33V
3Vcc= 3.33V

CPUID says:
CPU VCORE= 1.51 V
VIN1= 1.51 V
+3.3V= 3.33 V
+5V= 4.99 V
+12V= 10.54 V
VIN6= 0.80 V

(Graphics Card Voltage)
VIN0= 0.90 V

Both of these were just now run at the same time, and they both conflict on temperatures. Speedfan says nothing is above 54C which I believe is false, given the ambient temperature alone being 34C. CPUID shows things above 64C, which is more believeable for me. Fan speeds are the only thing concurrent between them (they just don't update at the exact same time so they differ slightly.

Picture below, just for clarification of the above.

http://i.imgur.com/8qlVA.jpg

 

Either way they are both only reporting ~10v on the 12v Rail. Thus some of your core components arent getting enough juice. This is made worse when put under load and the system starts trying to draw more power.

 

So... is that a PSU problem? Something I'd have to replace?

 

Also, I got compressed air today, and fired a few quick shots into the vents. It seems to have helped temps stay a hair lower, but not much, but that may be due to not actually cleaning anything in the process.

I found documentation on how to open the Touchsmart, but I'm afraid of risking total computer destruction by popping it open and cleaning it more thoroughly without knowing what I'm doing.
http://bizsupport1.austin.hp.com/bi...Document.jsp?objectID=c01962064&printver=true

 

Un fortunately we are getting a bit out of my understandings here. In a desktop PC you have 12v and 5v supply from the box. In the case of laptops and all in ones your powerblock only has one voltage (12v). Therefore my assumption is that the motherboard will have some sort of power circuitry that will provide the other voltages required. Your problem may be either the circuitry on the motherboard not working correctly or it may be the actual block that is not providing the full 12v

 

Could undervolting help me in this situation?
I was looking at ways to conserve power (low power mode, lowering screen brightness to 10-15%, etc.) but none of those seem to help.

I came across this site: http://forum.notebookreview.com/har...arket-upgrades/235824-undervolting-guide.html

And I'm wondering if it could help to keep the PC from requiring more voltage that may be causing a power loss.

 

Its debateable. What you are doing is involentary undervolting. Without know 1. whether the low voltage supply is causing the problem and 2. which component is not getting enough juice I suspect you could make the situation worse. The only real solution is to fix the problem but I cant say whether it is the power block or internal circuitry. If you can find an online retailer who will accept a return on a powerblock no questions asked then buy a new block and see if the voltage increases. If it doesnt send it back and I think you will have to get on to HP to get the internal circuitry looked at.

 

And I just don't have the funds for a repair. :/

I found this powerbrick at Amazon for about $70.
http://www.amazon.com/HP-TouchSmart-Adapter-AL192AAR-585010-001/dp/B003Y6NHNO

I'm sure HP would charge even more considering I'm no longer under warranty. I'm just hoping to find something I can do, personally, to solve or mitigate the problem until I have the monetary means for a professional repair.

 

I found this thread when searching for a solution to the same problem. The random power off has gotten to the point now that the machine will not stay on long enough to complete the login process. I have tried many approaches and I am at a total loss.

 

It's going on two weeks now since I undervolted my CPUs, and I'm rather pleased that I've only had one power loss in that time. And I am not even positive if that power loss was because of the same thing (cat may have barely unplugged cord from the back). Dropped the voltages down to 1.1125v (the minimum they can each go to) on the x6, x7, and x8 CPU FIDs shown in that thread I linked to, with no visible ill effects.

While it may not be the "best" fix, at least for the time being, it is a fix until I can afford to send it in for servicing or wing it and buy a new power supply.

 

Thanks I will give that a shot. I am assuming this approach will work equally well for a HP TouchSmart 300-1003 with the same power problem. The main problem is keeping the system running long enough to install the program and make the change. :-D

 

I really don't want to give anyone any technical advice, to further screw up their system. But it couldn't hurt to at least try and get a CPUID HWMonitor readout to see if yours looks similar to the screenshots I posted up earlier in the thread.