Jump to content


 


Register a free account to unlock additional features at BleepingComputer.com
Welcome to BleepingComputer, a free community where people like yourself come together to discuss and learn how to use their computers. Using the site is easy and fun. As a guest, you can browse and view the various discussions in the forums, but can not create a new topic or reply to an existing one unless you are logged in. Other benefits of registering an account are subscribing to topics and forums, creating a blog, and having no ads shown anywhere on the site.


Click here to Register a free account now! or read our Welcome Guide to learn how to use this site.

Photo

PC Start-up - Power Cycling On & Off


  • Please log in to reply
17 replies to this topic

#1 SAESeeker

SAESeeker

  • Members
  • 7 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Toronto, Canada
  • Local time:09:35 AM

Posted 28 January 2014 - 01:35 PM

This is my first post and it is kinda long.  To help with readability, I divided it into 2 parts: 1) the bare essentials, followed by 2) more depth and history "Long Version".  My PC details are at the end.

Short Version:

In my efforts to diagnose my PC start-up problem, I removed almost everything inside the PC.  Only the motherboard, CPU, CPU fan, the 3 system fans, front header connections, PSU, and the case remain together.  I press the ON/OFF button, next all the phase lights inside the PC light up and the CPU and system fans also turn on ... for 10 secs, then these shut down for 5 secs, start up for 10 secs, ad nauseum!  The monitor starts off blank and remains blank (black) no colour, text whatever, never gets to POST screen. Pressing the ON/OFF button has no effect.

 

Does this cycling indicate a problem with the mobo or CPU?  I tested the PSU and it appears fine (See Long Version for detail).

If the components were all fine, what should I expect when I turn the power on?

 

If all the components were fine, my thought is that the power LEDs should light and fans start, without cycling.  I think there is a problem with either the mobo or CPU but could it be something else   I am familiar with PCs but am not really technically proficient, I thought it would be helpful to ask for advice from others more knowledgeable than I.

Long Version:

 

This actually started March 2013.  System logs showed a "major event" happened in March.  The PC powered on intermittently.  Took me a while to figure out the PSU was a problem.  Bought a new 720v CoolerMaster.and installed in early September (I couldn't decide what to get!!).  Powered on reliably with new PSU but...

 

PC would go through POST, Microsoft start-up screen appeared, screen went black, stayed black.  (In the good ol' days, the black screen would take about 10 secs then the account log-on screen would appear (I have 2 accounts on my PC)).  Pressing the ON/OFF button was ineffective.  Pressing the reset button, I could boot up into Start-up Repair. 

Running Start-up Repair options: the Restore only found the "windows.old" folder, the Memory Test was fine.  Startup repair details usually indicated a bad patch, all the other tested parameters were OK.  

 

I researched various sites for possible solutions and fixes ... did complete system virus scan, scanned with MBAR, installed new drivers, tried rolling back a bad patch (DISM), checked BIOS settings (toggled IDE/AHCI SATA controller), tried fixing Master Boot Record, finally did a Repair Install of Windows 7 Professional.  Nothing I tried has worked. Just recently I read about replacing the CMOS battery; last week, I bought and installed a new battery. I turned the PC on and it booted up right into the account log-on screen!!

 

I was amazed that a lowly battery could have such an impact.  With bated breath, I shut down the PC and started up again.  The computer started up for a few seconds, then went off, then went on again, then went off, etc. until I turned off the power bar (pressing ON/OFF button had no effect).  More specifically, the phase LEDs and fans come on for 10 secs, the lights and fans go off for 5 secs, then this continues in the same synchronicity until I turn off the power.

 

I have gradually removed the innards of my PC to see if I could get a beep code or any sound whatever or get the cycling to stop.  I am now down to pretty much the bare essentials (as described above).   This morning it still does the exact same cycling.  An hour ago, I tested the PSU with my PSU tester.  The results were dead on or within range (the 12v1 reading being 11.5 and the PG was 360ms).  I also tested the PSU with the paperclip and the fans started running.  So I don't think it is the PSU.  I suppose it could be the front header but it didn't seem likely.

 

With this stripped down system, I THINK the fans OUGHT to go on and stay on, just stay running, until I cut the power.  I think the cycling on & off confirms that I have a mobo and/or CPU problem?

 

Thanks for taking the time to look at this.  Your thoughts are very much appreciated.

 

R

 

SPECS:  I built this in 2009 based on a Tech Report system build

Motherboard: Gigabyte EA-EP45-DSR3

CPU:  Intel Core2 Quad CPU Q9550 @2.83 GHz

CPU fan: Zalman CNPS9500A

PSU: CoolerMaster Silent Pro 720W (RS720-SPM2D3-US)

RAM:  4G Kingston DDR2      (KVR800D2N5K2/4G  i.e. 2 sticks, each 2G)

Hard Drives:  2 x Western Digital Caviar SE16 Sata (2 drives, 1GB each)

Network Adapters: 2 x Realtek 8111C (10/100/1000 Mbit)

Graphics:  EVGA NVidia GeForce 9800 GTX/9800 GTX+

Sound: Asus Xonar D2X

Blu-ray/DVD-ROM: LG GGC-H20L

Mouse: Logitech G5

Keyboard: Leopold FC500RR/AB

Monitor: SyncMaster 2253BW

UPS: APC BACK-UPS ES750)

Case: Antec P182

OS:  Windows 7 Professional with SP1



BC AdBot (Login to Remove)

 


#2 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 28 January 2014 - 01:46 PM

Not sure if you have tried this: use 1 single stick of ram, try in one dimm, if it does not post move it to the next dimm and repeat through all dimms, if that stick does not work try your other stick through each dimm.

 

MAKE SURE to power off and turn off the psu switch when removing components every time.


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#3 dc3

dc3

    Bleeping Treehugger


  • Members
  • 30,710 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Sierra Foothills of Northern Ca.
  • Local time:06:35 AM

Posted 28 January 2014 - 02:20 PM

Not sure if you have tried this: use 1 single stick of ram, try in one dimm, if it does not post move it to the next dimm and repeat through all dimms, if that stick does not work try your other stick through each dimm.

 

MAKE SURE to power off and turn off the psu switch when removing components every time.

 

I sounds like you are inferring that a DIMM is the socket for a RAM module.  A DIMM means Dual Inline Memory Module.


Family and loved ones will always be a priority in my daily life.  You never know when one will leave you.

 

 

 

 


#4 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 28 January 2014 - 02:28 PM

You are correct, I guess I should say dimm 0 dimm 1 dimm 2 dimm 3 or dimm slot rather, dimm does somewhat refer to the slot but also refer to the module. I just have always called the slots dimms because they usually are referred to dimm 0-3 on most motherboards not slot/socket 0-3 etc.


Edited by zingo156, 28 January 2014 - 02:43 PM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#5 dc3

dc3

    Bleeping Treehugger


  • Members
  • 30,710 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Sierra Foothills of Northern Ca.
  • Local time:06:35 AM

Posted 28 January 2014 - 02:56 PM

Most people just refer to them as slots.  If any further designation is required these slots are usually referred to as slots A2 and A4, B1 and B3, or similar nomenclature.

 

I say tomato, you say tomahto...

 

I suspect that the people at N.I.S.T. just shake their heads and hope it doesn't fall into their purview. :wacko:

 

:offtopic:


Edited by dc3, 28 January 2014 - 02:57 PM.

Family and loved ones will always be a priority in my daily life.  You never know when one will leave you.

 

 

 

 


#6 SAESeeker

SAESeeker
  • Topic Starter

  • Members
  • 7 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Toronto, Canada
  • Local time:09:35 AM

Posted 29 January 2014 - 02:21 PM

I thought it was neat that you knew to ask me to test the memory cards.  When I removed everything and the PC power still cycled I didn't think they would be an issue.  Learn something every day!   First I tested with the bare essentials as described in my first post.  Afterward I installed the graphics card and monitor and tested the cards again.  I ended each trial at the 4-min mark.  The results are ... TA-DA ....

 

Stick A (No graphics card & monitor installed)

 

DDR2_1:  At 0 seconds, I pressed ON/OFF button at front, the Phase LEDs lit up and the fans started, at 16 secs the phase LEDs went off while the fans continued to turn, after 4 minutes of listening to fans whirring away, I pressed the ON/OFF button at the front so the power went out and the fans turned off.  I disconnected power cord, moved the memory stick A to DDR2_2.   

DDR2_2, DDR2_3, and DDR2_4 socket testing produced the same results as DDR2_1 in each case.

 

Stick B (No graphics card & monitor installed)

 

Stick B - First Trial Set:

DDR2_1: It worked the same as Stick A.

DDR2_2: After starting up the Phase LEDs never went out until I hit the front ON/OFF button after 4 mins.

DDR2_3: It worked the same as Stick A.

DDR2_4: It also worked the same as Stick A.

 

Stick B - Second Trial Set:

DDR2_1:  Like Stick A

DDR2_2:  Like Stick A

DDR2_3:  Began like Stick A but after 2 minutes it started cycling on and off: 2 secs off, 53 secs on.

DDR2_4:  Began like Stick A but after 2 minutes it started cycling on and off: 2 secs off, 53 secs on.

 

Stick B - Third Trial Set:

DR2_1:  After starting, the power would cycle on and off: 4 sec off, 58 secs on.

DDR2_2:  Like Stick A

DDR2_3:  After starting, the power would cycle on and off: 4 sec off, 58 secs on.

DDR2_4:  Like Stick A

 

Being curious, I decided to install my Graphics Card & Monitor to see what was going on if anything ...

 

Stick A (WITH graphics card & monitor installed)

 

DDR2_1: At 0 secs, I pressed the front ON/OFF button, the phase LEDs lit up and the fans turned on and the screen was black.  At 26 secs, the phase LEDs went out (fans continued) and the Gigabyte Logo screen appeared, quickly followed by the BIOS screens.  At 36 secs, the screen showed a list of "stuff" ending with "Verifying DMI Pool Data ........   Update Success"  And on the line below: "DISK BOOT FAILURE, INSERT SYSTEM DISK AND PRESS ENTER" 

(I don't have the hard drives installed.)

DDR2_2, DDR2_3, and DDR2_4 socket testing produced the same results in each case.

 

Stick B (WITH graphics card & monitor installed)

 

Stick B - First Trial Set (with graphics):

 

DDR2_1:  At 0 secs, I pressed the front ON/OFF button, the phase LEDs lit up and the fans turned on and the screen was black.  At 26 secs, the Gigabyte Logo screen appeared, quickly followed by the BIOS screens.  At 36 secs, the phase LEDs went out (fans continued) and the screen showed a list of "stuff" ending with "Verifying DMI Pool Data ........   "  (did not show the words Update Success) and on the line below: "DISK BOOT FAILURE, INSERT SYSTEM DISK AND PRESS ENTER" 

DDR2_2:  Same as above DDR2_1. 

DDR2_3:  Pressed ON/OFF button to start but the screen never showed any text at all it was black throughout the 4 min test period.  I forgot to write down what happened with the LEDs!

DDR2_4:  At 30 secs, the phase LEDs flashed on & off.  A message about Checksum Defaults Control appeared.  I tried to write down info but it disappeared too quickly.  At 50 secs the Gigabyte logo screen appeared.  At 61 secs, the POST screens appeared .  The words "update success" did not appear..

 

Stick B - Second Trial Set (with graphics)

 

DDR2_1:  Same as DDR2_1 in First set (with graphics).

DDR2_2:  Same as DDR2_4 in First set (with graphics)  ....  checksum message appeared

DDR2_3:  Almost the same as DDR2_1 in First set (with graphics) but this time it showed "Verifying DMI Data Pool .... Update Success"

DDR2_4:  Same as DDR2_3 Second set (with graphics).

 

Conclusion: Memory stick B seems faulty!

 

Do you have any suggestions after having read the above?  I was thinking of installing stick A along with the hard drives and seeing if it would boot up normally.  Although I am not sure this would prove anything else.

 

I hope I have written this correctly.  I tried a preview on my first post and the message went into the ether.  Luckily I had copied it to a Word file just in case.  So I am not going to try preview again!

 

Thanks,

 

R



#7 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 29 January 2014 - 02:56 PM

I have a feeling you have a dead cmos battery by the Checksum Defaults being loaded (this probably won't affect power cycling but worth replacing anyway as it stores the time and settings for bios). You will get this defaults loaded when you unplug power to the system if the cmos battery that retains information is dead.

 

I would suspect you may have a problem with stick B with the power cylces from your second trial set with it... If you can get the computer to post with stick b you can run Memtest86 and see if there are any errors.

 

Also what happens now when you install both sticks of ram at the same time? If it continues to power cycle unplug, pull stick B and see if it stops. It is strange that the computer posted some of the time with stick B and other times it was delayed or did not work.

 

Overall, it would seem stick B has some issue especially because stick A had no problems and results which stayed the same throughout the tests.

 

EDIT: I would still run Memtest86 on stick A to verify no errors on that stick as well as Memtest86 on stick B. I would do each separately.


Edited by zingo156, 29 January 2014 - 04:48 PM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#8 SAESeeker

SAESeeker
  • Topic Starter

  • Members
  • 7 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Toronto, Canada
  • Local time:09:35 AM

Posted 29 January 2014 - 03:07 PM

Ever buy a gizmo with a battery and find out that the replacement battery costs more than the gizmo originally did??  Someone is making a lot of money on batteries!!  I replaced the CMOS battery 2 weeks ago with a brand new one.  Nonetheless, I shall invest in a new one later today and try your suggestion with it installed.



#9 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 29 January 2014 - 04:43 PM

I forgot to mention, when you run memtest86 allow it to do 3 passes, if you have no errors then in theory the stick should be ok.


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#10 SAESeeker

SAESeeker
  • Topic Starter

  • Members
  • 7 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Toronto, Canada
  • Local time:09:35 AM

Posted 01 February 2014 - 03:34 PM

A) Memory Testing

 

Following Zingo 156's suggestions, I downloaded MemTest86 v 5.01 to a bootable CD. I added my keyboard, mouse and 2 hard drives to my PC setup.  Upon starting, I entered the BIOS and changed it to "optimized settings".  I tried to run 7 passes for each session which took slightly more than 4 hours per session.

 

Stick A in DDR2_3 (4 passes - a kid wanted to go to sleep in the room where the PC is perched so I relented on my 7 pass ambition!) - Result: Zero errors.

 

Stick B in DDR2_1, _2,_3, and _4 (7 passes each) - Result: Zero errors.

 

 

B) Zingo156 had asked what happens when I put both sticks of RAM back in together. 

 

Sticks A & B:

 

After the memory tests, I installed Stick A and Stick B in DDR2_2 & DDR2_4, and booted the PC (sans MemTest86 CD of course!).  The power didn't cycle on and off.  Apart from the fans staring to whir, nothing seemed to happen, the screen was black, so I turned the PSU off. 

 

I turned the computer back on.  It went through the boot sequence to DMI Pool Data verification then the Windows Error Recovery (WER) screen appeared.  I launched Startup Repair.  It indicated that it was checking the hard disk for errors.  Eventually a message appeared saying "Startup cannot repair the computer automatically".

Problem Details were:

  • Event Name = StartupRepairOffline;
  • Sign 03 = Unknown;
  • Sign 04 = 21199545;
  • Sign 05 = AutoFailover;
  • Sign 06 = 2; and
  • Sign 07 = No Root Cause.

The Diagnostics and Repair Details were as follows:

  • System Directory = \Device\Harddisk 0;
  • Windows Dir = D:\Windows;
  • AutoChk Run = 0;
  • # of Root Causes = 1
  • The numerous tests, analyses, and diagnoses were completed successfully with 0x0 error codes. 
  • Root Cause: "Unspecified changes to system configuration might have caused the problem". 

Stick A Only:

 

With Stick A in DDR2__1, _2, and _3:  For each socket, all the fans started to whir but nothing at all appeared on the screen, entirely black, then after waiting about 2 mins, I shut down and started up a second time.  It booted to the WER screen.  Results were the same as A&B together. 

 

Stick A in socket DDR2­_4:  It booted up first time to the WER screen.  Results were the same as A&B together. 

 

Stick B Only:

 

I tried Stick B alone in each socket.  DDR2_1 booted directly to WER and DDR2_2 needed a second boot but in both cases, the results were the same as A&B together. 

 

With Stick B in DDR2_ 3 and _4, the PC booted to a black screen (after the Device listing screen) then after booting a second time, it went to the WER screen.  The results were the same as A&B together except that "Root cause found:  Startup Repair has tried several times but still cannot determine the cause of the problem."  Obviously it now considers me a pest!



#11 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 03 February 2014 - 08:26 AM

The first step is to figure out what is causing the power cycles, my guess is that windows may have problems related to unplanned shut downs etc. In your last post it sounded like the computer made it passed the post screen with both sticks, I would run a memtest with both sticks installed.

 

Also on your motherboard look for bulging or blown capacitors because it seems you have some random no post issues, it could be a harddrive but for now look over your motherboard: https://www.google.com/#q=bulging+blown+capacitors+on+motherboard


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#12 SAESeeker

SAESeeker
  • Topic Starter

  • Members
  • 7 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Toronto, Canada
  • Local time:09:35 AM

Posted 04 February 2014 - 01:54 PM

I tested both memory sticks (in DDR2_2 & _4) with MemTest and there were no errors after 5 passes.

 

I tested each HD with WD's Data Lifeguard Tools (extensive test) and both passed.

 

The power is not cycling on and off anymore. 

 

When I booted up this morning, the PC went through POST into Windows Error Recovery screen. 

 

After a few hours, I turned on the PC and nothing happened - no POST, screen black. 

 

After unplugging and plugging in the PC power, I started up again and this time it went through POST to the Windows Error Recovery screen.  I launched the recovery option.   The Problem Signature was almost the same as the 1 Feb except that now Signature 04 = 1.  The Diagnostics and Repair Details were exactly the same as on 1 Feb with the root cause also being "Unspecified changes to system configuration might have caused the problem".



#13 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 05 February 2014 - 08:44 AM

This: "After a few hours, I turned on the PC and nothing happened - no POST, screen black." Sounds like a classic NO POST situation. Something hardware related is still causing an issue. I do suspect you have an issue with windows or possibly even a hard drive, if you can run a chkdsk this might clear up a windows issue.

 

Did you check for any blown capacitors on the mainboard? https://www.google.com/#q=bulging+blown+capacitors+on+motherboard

 

Because it is an intermittent no post issue sometimes bad capacitors can cause this.

 

Also I would recommend running a test on your hard drive(s) a bad hard drive can cause a no post, actually any hardware at all connected can cause a no post situation. The test that I use is MHDD: http://hddguru.com/software/2005.10.02-MHDD/

 

It is pretty tricky to use at first so read through whatever information they have. You will need to change the sata settings in your bios to IDE OR compatibility, it is likely set to AHCI currently. Write down whatever the default setting is before you change it as you will need to switch it back. Booting to windows after changing this setting will likely cause a bsod that is why you will need to change it back to whatever it was.


If I am helping you with a problem and I have not responded within 48 hours please send me a PM.

#14 SAESeeker

SAESeeker
  • Topic Starter

  • Members
  • 7 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Toronto, Canada
  • Local time:09:35 AM

Posted 05 February 2014 - 10:18 AM

I forgot to mention it in my previous post but I had checked for capacitors leaking, bulging, blown.  I didn't see anything on the mother board or the graphics card (although most of the graphics card was covered).

 

I did a hard drive test but used different software.  Since my HDs are Western Digital, I used their software "Data Lifeguard Tool" to run the "extensive" test on each drive.  They passed.  Do you think running MHDD would be better?



#15 zingo156

zingo156

  • BC Advisor
  • 3,345 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:35 AM

Posted 05 February 2014 - 10:36 AM

No the WD tool should be fine as well. I guess the only way to rule out hardware causing a no post is to slowly remove 1 component at a time and see if it posts consistently. It can be anything, add in pci cards, dvd/cd drives, etc...

 

This is the first step before trying to fix the windows boot problem.


Edited by zingo156, 05 February 2014 - 10:42 AM.

If I am helping you with a problem and I have not responded within 48 hours please send me a PM.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users