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

need help with new rig - ASrock z170 Extreme7+ power button glitch


  • Please log in to reply
15 replies to this topic

#1 Mechcondrid

Mechcondrid

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 11 September 2015 - 10:35 PM

ok so i just put together a new mid tower rig and it has a bit of a quirk that i cant seem to pin down, the MB will not power on the PSU unless i hit BOTH the power and reset button at the same time; if i just hit the power it cycles on for a split second then shuts down.

if i hit and hold reset and then hit power then immediately release both it boots fine and has no issues (besides the no ehci issue of the 170 chipset but thats more an annoyance than anything)

so i've tried swapping pins for the case buttons and it does the same thing either way.

now there are two front panel pin groups one says the usual (pwr, rst, hdd_led, pwr_led) the other says speaker and rst or rstled (cant tell looks like the silkscreen didn't hold well)

havent tried those as obviously not sure what that second set is for beyond the board speaker, and id rather not damage my new $300 board not even a day after i got it in.

 

any help would be greatly appreciated.



BC AdBot (Login to Remove)

 


#2 ranchhand_

ranchhand_

  • Members
  • 1,623 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Midwest
  • Local time:06:50 PM

Posted 12 September 2015 - 12:20 PM

The fact that your new unit runs perfectly after boot proves that your components and configurations are good.

>  Power down and disconnect the power cable to the PS. leave all other cables connected, including the video cable.

> Open the side cover and trace the tiny cable(s) that run from the power and reset switches to where you connected them to the mainboard. Unplug them. Leave all other cables connected.

> Reconnect the power cable.

> Take a flat-bladed screwdriver, and cross the two tiny connectors for the power-on switch. The unit should power up and boot properly.  If it does, you have a bad power switch, OR possibly a bad reset switch, OR both.

I have had both bad power switches cause this, and even the reset switch on my box went bad and the unit did crazy things until I tested as above and the reset switch started shorting. I disconnected it and my unit is running fine to this day.

Since the box is new and probably still under warranty, I suggest contacting either the retail store or the manufacturer before your new-component warranty expires, then you are up a creek. If you purchased from Newegg, you have 30 days to return the box and they will send you another. After that you must work through the manufacturer, and good luck with that.

IF the power/reset switches are replaceable, having them send you new ones may be easier than dismantling your box and sending it back.


Help Requests: If there is no reply after 3 days I remove the thread from my answer list. For further help PM me.


#3 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 13 September 2015 - 10:52 AM

ok i''m ashamed i didn't think of that; considering untill about 6 months ago i was helping to manage an enterprise network (with one other person) with ~350 pieces of networked equipment.........

but yes it was a short, but not due to anything being broken; more due to either bad engineering or bad QA or probably both.

the power and reset switch are just simple SMD microswitches and the two pins on the opposite side of the connections (which on the SMD versions are connected to the switch contacts) were exposed and not snipped off as they should have been and were being shorted by the two screws that hold the module together (the bad engineering part) so i just put a small strip of electrical tape on the bottom of the switch and that fixed it.

 

although now i'm having a separate issue with "memory management" error BSODs (i've actually never encountered that type of BSOD so no real idea on what exactly is causing that)

if i should make a new topic just let me know, but if you could give some insight on that i'd be greatfull.



#4 jonuk76

jonuk76

  • Members
  • 2,167 posts
  • ONLINE
  •  
  • Gender:Male
  • Location:Wales, UK
  • Local time:01:50 AM

Posted 13 September 2015 - 01:01 PM

Having just gone through weeks of troubleshooting BSOD's on my main PC following an upgrade install of Windows 10, and certainly some of them were "memory management_1a" BSOD's, I would say they are NOT necessarily related to bad hardware, or incorrect memory timings, but they can be.  They may equally be caused by device drivers.  In my case, I seem to have cured it (touch wood) by re-installing Windows 10.

 

But you can perform some checks anyway.  Memtest86+ is run from a bootable CD and it's recommended to run at least 3 complete passes of it.  Prime95 is useful for CPU and memory stability testing.

 

BluescreenView by Nirsoft may be helpful in highlighting any drivers implicated in any of the BSOD's.  Probably most helpful if one in particular is consistently causing them.

 

Make sure you are using the latest drivers for everything.  Also I would suggest uninstalling unnecessary utilities (particularly overclocking utilities) provided by the motherboard manufacturer, as they sometimes cause problems.  

 

If you continue having crashes then there are some experts in the Windows Crashes and BSOD section of the forum, which has some specific instructions to carry out before making a new thread on there.


7sbvuf-6.png


#5 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 13 September 2015 - 05:12 PM

well just for bleeps and giggles turned on the desktop at 8 am this morning (win 7 btw, i'm not touching 10 till they put the first service pack out; especially after the string of boot loop issues it's had) it is now 6:04pm and it only just now had a memory management BSOD (it does not say the 1a part on mine btw, dont know what that means if anything) but only after i started my copy of batman arkham knight (which ran pretty much perfectly up until the game froze and had to hard cut the system because alt tab and ctrl+alt+del were not responding) and i had my sandisk 32gb flash drive plugged in as a key-fob keyring access; and it got the memory management error shortly after coming to the windows login screen.

so i'm maybe thinking the keyring program that comes with the mobo is bugging something out?

as i didn't have it plugged in all day as i had the flash drive with me and i was out and about.

and i do overclock it a bit but i never use in-windows apps to do that, way to many issues. i just alter multipliers directly in the uefi bios, i do use programs to monitor the temp and clock speeds though (cpufan mainly) but not with any programs that can alter any of those settings.

i'll try that bluescreenview though, wish i had heard about that before; could have been a lot of help in the last job instead of re-imaging workstations by hand for 2 hours a piece (decentralised workstations FTL)



#6 ranchhand_

ranchhand_

  • Members
  • 1,623 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Midwest
  • Local time:06:50 PM

Posted 13 September 2015 - 10:49 PM

Run Memtest 86+ as jonuk suggested; in addition, I suggest you run it on one stick at a time; I have had the experience of running it on multiple sticks and it messaged me as  "no errors".  Since I still had the same problems, I ran it single stick at a time and one of them was bad. That stopped my problems. Use the Prebuilt and ISO header and burn it to a disc.


Help Requests: If there is no reply after 3 days I remove the thread from my answer list. For further help PM me.


#7 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 17 September 2015 - 09:43 AM

ok ran memtest and got errors on one run on test 7, but i was looking in the menu at the time so i think that bugged it.

i ran it 3 more times (and didnt touch it) and got no errors, also i only have 1 8gb stick in atm; i also cant seem to get the system to generate a minidump ? it will bluescreen, say its dumping and then i reboot but the minidump folder is completely empty? i've never really delved into anything to do with minidumps before, my usual course would have been a re-image of the drive if i couldnt find the culprit in the software installed.

 

on a side note i have had the system running for about 2 days straight and have been using it, even ran some pretty intensive games multiple times (dying light, witcher 3, elite dangerous) and only got a memory management error last night (or morning i guess) at 1am when i was just using chrome immediately after i closed the browser windows.

if i didnt know better i'd think it was something to do with program garbage collection on the ram...



#8 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 17 September 2015 - 12:50 PM

ok i think i found the culprit (though there is still no minidump file anywhere) the asrock utility USBkey has a scanner app that is called IOmonitor.exe and (i'm assuming) since it allows you to use any usb device as keyfob login access (so not just flash drives) i think its causing some kind of error.

i, just for bleeps and giggles opened up task manager and forced the iomonitor.exe process to close and got an instant "memory management" BSOD (and of course no minidump) so i have now uninstalled that utility and and going to try continuing normal use of the system, so far it has not BSOD'd going on 3+ hours.

here's to hoping!



#9 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 17 September 2015 - 02:39 PM

ok so no dice it just did it again and still no minidump, so i'm at a loss.

i re-ran memtest and it still shows no errors on multiple runs

i'm not sure what else i can try here...



#10 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 18 September 2015 - 09:33 PM

ok i'm pretty certain i nailed it, and it was a fix just as stupid as the power button one....

the motherboard comes set as default, that if a stick of ram supports the XMP SPD extension to automatically use it as the default profile; my one stick here DID support it and that profile called for a ramp up of the memory speed to the equivalent of a DDR4-2800 stick (the stick is stock at 2133) so it was overclocking the one ram stick by something like ~400Mhz as well as altering the CAS timings as well as the cycle time (tRAS).

after going in and disabling the XMP setting in bios i've not had a single BSOD since yesterday afternoon.

 

yay for stupid mistakes i guess? (or overlooking them rather...)



#11 jonuk76

jonuk76

  • Members
  • 2,167 posts
  • ONLINE
  •  
  • Gender:Male
  • Location:Wales, UK
  • Local time:01:50 AM

Posted 18 September 2015 - 09:47 PM

Isn't the memory matching then?  If the same type you'd expect all of the DIMMs to support XMP.


7sbvuf-6.png


#12 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 19 September 2015 - 12:24 AM

if you mean matching the XMP profile, then yes it was; or at least it was trying to as far as i can tell.

i think the different timings were causing the memory error BSODs, i mean 400Mhz on a basic budget (i guess? was about $35 - $40 for the one stick) ddr4 stick with no heat spreader to speak of might be a bit much, at least without messing with voltages and i dont feel confident enough in that area to tinker with in a new system.

but i've never really got into OCing any ram much before, at least other than Vram. i have always just OCd the cpu and sometimes the gpu and just bought higher grade ram.

i dont know if this is the case or even possible but i would assume the combination of the XMP profile and the Mobos default so called "auto" setting for the frequency and timing settings could conflict in some cases? would have been better for asrock to set the default to not overclock anything instead of setting everything to "auto" straight out of the box.

 

but ya had no further BSODs still and it's now 1:20am here so i think its safe to say that was the cause.


Edited by Mechcondrid, 19 September 2015 - 12:26 AM.


#13 ranchhand_

ranchhand_

  • Members
  • 1,623 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Midwest
  • Local time:06:50 PM

Posted 19 September 2015 - 07:11 AM

 

different timings were causing the memory error BSODs,

Thanks for following up; this is interesting; I have used different memory timing sticks in older units that come in with no problems. (I know it is not the best practice for general use, but these are old units used by folks for general purposes, nothing stressful). I guess each mainboard is slightly different in requirements. Good thing to remember for the future. Thanks again.


Help Requests: If there is no reply after 3 days I remove the thread from my answer list. For further help PM me.


#14 Mechcondrid

Mechcondrid
  • Topic Starter

  • Members
  • 14 posts
  • OFFLINE
  •  
  • Gender:Male
  • Location:Connecticut, USA
  • Local time:07:50 PM

Posted 19 September 2015 - 07:34 AM

well i know that you can use alternate timings to improve response times but i think in this case it was a combination of the ~400Mhz frequency overclock and the timing change.



#15 Captain_Chicken

Captain_Chicken

  • BC Advisor
  • 1,354 posts
  • OFFLINE
  •  
  • Gender:Male
  • Local time:08:50 PM

Posted 27 September 2015 - 08:42 PM

I would not recommend prime95. The latest version can stress components to much and cause damage.

Computer Collection:

Spoiler

Spoiler

Spoiler

Spoiler




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users