No POST after case change

[Accessless]

I finally decided to replace my rickety old case today and surprise surprise something went wrong. So to be clear no other hardware changes have taken place & the computer was working fine this morning.

Symptoms are a failure to POST, beep or respond to the power button to turn off. I've gone through the standard problem solving check list:

1) Unplug everything except bare minimum: 1 stick RAM, Power, keyboard, graphics device.
2) Alternated different RAM in different slots.
3) Swapped PSU for a known good one.
4) Swapped graphics card for a known good one
5) Reset the BIOS
6) Threaten with violence.

Oh and the monitor works fine as well. I'm thinking that my motherboard did not survive the transplant.

[DaveLembke]

Last time I had a client with a similar issue it was that they added too many stand offs in a case that allowed you to add the bronze stand offs in the needed locations and one of them was in a location that is not a mount point, but just happened to be where IC legs were sticking out from, sending the signal to chasis ground.

Hoping this is not the cause of your failure as for they blew the motherboard by this oops.

The other time a person had an issue like this was because they were attempting to use a flat headed screw driver to eject the RAM out of the wings and the screw driver slipped off of the wing and struck the surface of the motherboard and travelled about 1/8" inch in a scratch motion. They cut across about 9 hairline traces that went from the CPU and RAM. I ended up helping them save this board by using an exacto to scratch the mask off of the traces to expose the copper and then with a soldering iron and solder added solder bridges at the gouge and reconnected the RAM traces to the CPU. 

So if a screwdriver jumped anywheres inside when working look at the motherboard surface..


Other causes for this can be the following:

- Motherboard was flexed during the process causing an open to occur.
- Screw somehow loose is lodged and shorting component(s).
- Screws tightened down with too much force and cracked/fractured the motherboard material which broke traces internally.
- New cases sometimes have rivot material or metal filings that fall into contact with board.
- Jumper(s) fell off in process such as BIOS Jumper etc.
- Other devices that the motherboard connect to were damaged and so you need to see if you get different results with minimal boot on integrated video etc vs video card etc.
- Static discharge through the process by not following proper ESD safe procedures
- Molex Power connectors not inserted properly or open connection between male/female pins
- 12v ( 4 pin ) molex power connection is not plugged in near CPU ( This has been found more than once!!!! )
- power switch and LED wiring to front panel wired wrong such as to RESET vs the POWER BUTTON so its actually holding the reset into a high vs low state ( and yes I have seen this before in my travels assisting people for 25+ years ) This was on a old 486 33Mhz years ago with button that was not a momentary switch as I believe 100% of modern computers are momentary switches so getting power switch and reset backwards just means that the reset buttom powers and the power button resets.
- Part on motherboard or other device damaged in the move.
----------------------------------

I think that covers all... but feel free to add any that I may have missed for him to try.

In this situation you could pull the motherboard, PSU out and try to boot it outside of the case on a tabletop with a tabletop surface that is dry and non conductive and if it boots then, work your way backwards to the case to find the cause of the failed boot. ( verify that component legs on the board are not comming into contact with the case and shorting out as for many cases now ship without the bronze stand offs, but instead are a press formed platform that are drilled and tapped and sometimes the dome of the raised mount point can conflict with IC legs etc. This can spell disaster for the board depending on which leg is sent to ground.

[Accessless]

When I say bare minimum that included switching on via screw driver with the front panel disconnected and the case disconnected entirely. I'll be honest I was using the CPU heat sink as a handle to move it about (It felt more fragile than what I'm used to). Then there is the RAM that's not on a supported part of the board so there was flex there...

I miss my old Pentium II boards that just got thrown in cupboards and yet still somehow worked afterwards.

[DaveLembke]

If you had a POST card to plug into a PCI slot that might point you in the right direction at this point.

I have an older version of one of these with a LED readout that you can see where it is hanging. Mine is called a POST FLIP card because it has ISA on one edge and PCI on the other so you can plug it into either. But its like 13 years old and I should probably get a newer one, but rarely need it. I paid like $49.99 for it and used it maybe 3x in 13 years.

http://www.newegg.com/Product/Product.aspx?Item=9SIA2C51051131

[Accessless]

Unfortunately I'm not American so newegg products aren't any good to me.

It appears that the motherboard is still under warranty, if it counts that I probably broke it. The question is do I want to put up with the hassle of returning it for ?weeks? or just splash out the extra £55 for a new one.

[DaveLembke]

I have always used my warranty when available. Never had to send a motherboard back before, but have had to send Hard Drives ( Seagate x 3 )and RAM ( 1 x Kingston + 1 x Crucial ) back and it has been no hassle replacements.

Seagate was great for all 3 drives, and of the 3 drives, they shipped me back a 160GB IDE replacement to the 120GB IDE drive that died as well as a 500GB Maxtor drive that was not covered by Maxtors 3 year warranty and became a paperweight instantly was covered by warranty when Seagate bought out Maxtor and rolled their 5 year warranty over to the Maxtor product line in which the drive was 4.5 years old, and Seagate gave me a RMA approval on this drive and I got shipped back a healthy Seagate branded 500GB SATA II drive for just the cost of shipping $ 8.00 (USD) and they gave me a 3 year warranty on this replacement drive that started from the delivery date of the drive to me. This 3 year warranty is just about up and the drive that is a work horse for video editing and compression is still healthy 

[Accessless]

One last after thought, what would the typical response from the motherboard be if it thought that the CPU fan wasn't working properly/at all?

[camerongray]

Have you tried running it out of the case? I'd predict that something is shorting the board out.

Moving the board by the heatsink shouldn't be an issue - In fact, that is one of the recommended ways to hold a board when putting it into the case as the cooler is always very solidly attached and is not sensitive to static.

[Accessless]

It's on a table now.

[Salmon Trout]

One last after thought, what would the typical response from the motherboard be if it thought that the CPU fan wasn't working properly/at all?

Can you see if it is rotating? Some motherboards simply won't boot if there is a zero RPM signal from the fan, or it is disconnected, others will and you will get a pretty short period of operation before the CPU goes into thermal shutdown.

Do we know what make and model this board is?

[Accessless]

The fan appears to be running fine. I'm just trying to rule out everything.

Its the computer in my profile Asus M5A97 LE R2.0. Sorry, always assume that it's my computer in question unless I otherwise say so, because I always will say so.

[Salmon Trout]

assume that it's my computer in question unless I otherwise say so

I hardly ever remember to look at peoples specs; they don't interest me very much.

[Accessless]

I don't usually look either. For example I have no idea what your listed specs are. I just use it to save time typing it all out every time. At the very least it makes it easy to copy and paste into a discussion.

[camerongray]

Again, have you tried running it out of the case?  It is very probable that the board is shorting on something in the case - Very easy mistake to make.

[Accessless]

Have you tried running it out of the case? I'd predict that something is shorting the board out.

Moving the board by the heatsink shouldn't be an issue - In fact, that is one of the recommended ways to hold a board when putting it into the case as the cooler is always very solidly attached and is not sensitive to static.

It's on a table now.

Again, have you tried running it out of the case?  It is very probable that the board is shorting on something in the case - Very easy mistake to make.

Yep still on a table, still not working. Also I would like to add that the motherboard was and always has been mounted on those hexagonal screw thingies. (sorry that got a bit technicle :p)

[Accessless]

I'm going to put together an old build whilst my motherboard has been sent off but I want to be sure that I don't create another paper weight by using a botched case.

Can anyone think of any electrical checks to perform on the case?

USB ports showed no crossed connections. So that's one down.

[DaveLembke]

If its pressed metal it will have round domes that are drilled and tapped, otherwise it will have the brass stand offs that are hexagonal. I prefer the hexagonal stand offs vs the pressed metal as for depending on the case, sometimes the domes are in conflict if IC legs. A long time ago I had to snip the legs of components on the bottom of a motherboard and also add electrical tape to the dome so that there was an insulator between the board bottom and the chasis metal. But with more and more surface mount parts this is usually a problem of the past, and modern boards are designed usually with a good area around the mounting holes to avoid a short to ground problem.

If its the hexagonal brass stand off type, verify that each one is exactly in reference to a mounting hole and none are screwed in... in a location where there is no mounting hole. This will cause the motherboard surface to short too ground which will cause problems.

[Accessless]

I tested with a multimeter. I thought that I got some strange readings but apparently they were well within tollerance. My old build is quite happly running in the new case for now.

May as well share the info about testing. If you connect everything up (save the power cord) and apply a multimeter to the ground pin on the mains input of your power supply, and then sequentially through all the connections on the ATX 24 pin plug, then provided you don't get a reading bellow 50 ohm all is well (excluding the black cables (ground) obviously).



I may never understand why in England we changed our power cable colours to the european ones. Red, Black, Green to Brown, Blue, Yellow & Green respectivly. Especially when in any application not covered by the same regulations uses the old colours. This does bring up a point of trivia though who's wiring system are computer/electronic systems based off?

[Salmon Trout]

I may never understand why in England we changed our power cable colours to the european ones.

It changed in Wales, Scotland and Northern Ireland as well. The UK colour scheme for 240 V flexible cords, namely live: brown, neutral: blue, earth: green (green/yellow since 2006) is not specifically "European", it's also used in Australia, South Africa, Hong Kong, Singapore, Malaysia and Saudi Arabia etc as well as the EU and is based on the international standard IEC 60446. It changed in Britain more than 40 years ago, so any equipment still using the old (red=live, black=neutral, green=earth) scheme must belong in a museum and may be unsafe because insulation that old may well have deteriorated. One reason for the change had to do with the danger posed if someone who was red/green colour blind wired up a plug. The earth (safety) conductor which connects to any exposed, touchable metal in an appliance is intended to serve as protection if a live conductor somehow contacts the metalwork. In that situation there would be a short and a fuse would blow or a breaker would trip. If the earth and live conductors were swapped in the plug, the exposed metalwork would immediately become live when the appliance was plugged in. The fuse would not blow, and there would be no sign that anything was wrong, and the applicance would probably still work. This would, of course be very dangerous. Of course nowadays we have RLCBs and moulded on plugs on many appliances. These, and the wiring colour change, are two reasons why there are fewer domestic electrocutions than decades ago.

[DaveLembke]

I have been finding these 2 types for years, but its only discovered when I cut the computer power cable to use the male plug and 3ft to 5ft of cable to replace a damaged power cord for some other device that I find the euro colors for the 3 vs the white/black/green or red/black/green when I cut through the black, grey, or beige outter jacket.

I also found out something pretty dangerous with computer power cables and that is that if you ever get one for a inkjet printer that is the type that plugs into the external power supply. Never use these as a replacement for a computer power cable.

It was discovered when a cord started to smell a short time after power up like hot plastic and was pretty hot at the plug end that instead of 16GA or 18 GA wire as is in most computer power cables, the power cables for the printer an older HP Officejet ink jet all-in-one actually had like 22GA wires at the core of the outter jacket and the computer that was drawing around 400 watts of power was overloading this light wire and so its a fire hazard if anyone decides to use a long gone printer power cable for their computer instead.