My computer was working fine. I sut it down to install a hard drive heat sink/fan. After reconnecting the hrad drive power connectors the computer will only turn on for a second before it shuts down. This is a Compaq computer I got from a friend. It has been running great until now. I've disconnected a piece of hardware at a time, but still nothing. What am I doing wrong?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Power Supply shuts down after I plug it in
- Thread starter vanlee
- Start date
It sounds like the CPU fan sensing on the motherboard is not detecting the new fan so is shutting down, kind of a fail safe mechanism so the CPU doesn't over heat.
It could be that the new fan is a larger/lower spinning type that the motherboard is struggling to pick up OR you have plugged the new fan into the wrong fan header on the motherboard.
Martin
Replying helps further our knowledge, without comment leaves us wondering.
It could be that the new fan is a larger/lower spinning type that the motherboard is struggling to pick up OR you have plugged the new fan into the wrong fan header on the motherboard.
Martin
Replying helps further our knowledge, without comment leaves us wondering.
- Thread starter
- #3
i think you are right in that something is causing the power supply fail safe to shut it down. but, i installed a hard drive fan, not a CPU fan and I've since disconnected it. I've disconnected everything, but then the power supply won't even start for a second. I read that you have to have at least 1 thing plugged into the power supply or it will short out.
franklin97355
Technical User
It's a good idea to have a load on a switching power supply. I would disconnect the new hardware, make sure nothing else has come loose and try again.
Sorry, didn't see the hard drive bit, thought you were talking CPU :>)
Try resetting the Cmos and just generally reseating everything including the main power connector to the board.
If you still have no joy I would be starting to suspect a PSU problem.
Martin
Replying helps further our knowledge, without comment leaves us wondering.
Try resetting the Cmos and just generally reseating everything including the main power connector to the board.
If you still have no joy I would be starting to suspect a PSU problem.
Martin
Replying helps further our knowledge, without comment leaves us wondering.
Well, a switch mode PSU won't actually short out w/o load.
Try disconnecting all power connectors, then connecting only the HD (not the new HD fan) and see if it starts spinning. If it does, then connect one power connector at a time -you could be facing a PSU overload.
If the HD does not spin, then the PSU could be dead! And that's a pain -if it's been overloaded, then there is no better solution than getting a new one.
Try running your new HD fan from a battery to make sure that you won't blow another PSU §;O)
Good Luck
Jakob
Try disconnecting all power connectors, then connecting only the HD (not the new HD fan) and see if it starts spinning. If it does, then connect one power connector at a time -you could be facing a PSU overload.
If the HD does not spin, then the PSU could be dead! And that's a pain -if it's been overloaded, then there is no better solution than getting a new one.
Try running your new HD fan from a battery to make sure that you won't blow another PSU §;O)
Good Luck
Jakob
dkdude,
A switching power supply (easily the most commonly found) will not short out, it just will die very quickly without a load on it.
I think the essence of your point was that there is something that is creating a non-power good situation between the processor and the PS.
I have never timed it, but a switching power supply in a no-load condition would likely last less than a minute. (I am guessing under 30 seconds). Ed Fair I suspect has better experience with this than I. I have never attempted a no-load, even with many mistakes, to see (discover) what could happen. But a new power supply would likely be the result.
But it sounds to me like this thread starter needs to follow your advice: motherboard, one drive. Are the fans spinning? If not, replace the power supply.
A switching power supply (easily the most commonly found) will not short out, it just will die very quickly without a load on it.
I think the essence of your point was that there is something that is creating a non-power good situation between the processor and the PS.
I have never timed it, but a switching power supply in a no-load condition would likely last less than a minute. (I am guessing under 30 seconds). Ed Fair I suspect has better experience with this than I. I have never attempted a no-load, even with many mistakes, to see (discover) what could happen. But a new power supply would likely be the result.
But it sounds to me like this thread starter needs to follow your advice: motherboard, one drive. Are the fans spinning? If not, replace the power supply.
Don't know, I've never let one come fully up without load. Too many bad experiences in the old days.
And with ATX I normally put the M/B in place so all the sections have some load.
The immediate shutdown is a crowbar shutdown, generally from excessive current draw in one section. And should have cleared when the new fan was removed. But sometimes when they crowbar they commit harikari.
Ed Fair
Give the wrong symptoms, get the wrong solutions.
And with ATX I normally put the M/B in place so all the sections have some load.
The immediate shutdown is a crowbar shutdown, generally from excessive current draw in one section. And should have cleared when the new fan was removed. But sometimes when they crowbar they commit harikari.
Ed Fair
Give the wrong symptoms, get the wrong solutions.
I'm suggesting to test one component at a time to see if/what/when the PSU fails. Might be a faulty something or a PSU overload.
The fan in the PSU will always act as a load, so I wouldn't worry about blowing out the PSU. As an engineer I can't see why it should happen anyway -we do it at work all the time -but that's a whoooole different story, heh §;O)
vanlee,
Hope you got some troubleshooting ideas and get the thing working again §;O)
Cheers,
Jakob
The fan in the PSU will always act as a load, so I wouldn't worry about blowing out the PSU. As an engineer I can't see why it should happen anyway -we do it at work all the time -but that's a whoooole different story, heh §;O)
vanlee,
Hope you got some troubleshooting ideas and get the thing working again §;O)
Cheers,
Jakob
Jakob,
Hard to generalize. Like as was suggested above, never dared to try it without a reasonable load on a switching supply.
In the "old days" a switching PS with just its own fan as a load would be toast, my guess within 30 seconds. I know for the serious they still sell little boxes for USA $4 to allow you to put a load on modern switching supplies to test their leads without damage to the PS itself.
Again, the PS fan is not sufficient from the warnings I have read for even ATX switching PS units. To run the PS without load other than its fan is almost certainly fatal.
If you have better advice, I would appreciate hearing it. I always ensure a load before testing the power leads with a VM, and I know from reading the Forums here and elsewhere that the PS without load seems even with modern ATX supplies not to work very well, or at all, from amatuer diagnostics of a "No POST" by pulling all the power supply leads to the system.
Hard to generalize. Like as was suggested above, never dared to try it without a reasonable load on a switching supply.
In the "old days" a switching PS with just its own fan as a load would be toast, my guess within 30 seconds. I know for the serious they still sell little boxes for USA $4 to allow you to put a load on modern switching supplies to test their leads without damage to the PS itself.
Again, the PS fan is not sufficient from the warnings I have read for even ATX switching PS units. To run the PS without load other than its fan is almost certainly fatal.
If you have better advice, I would appreciate hearing it. I always ensure a load before testing the power leads with a VM, and I know from reading the Forums here and elsewhere that the PS without load seems even with modern ATX supplies not to work very well, or at all, from amatuer diagnostics of a "No POST" by pulling all the power supply leads to the system.
None of the suppliers put a minimum load requirement. Probably because everybody will have a M/B and hard drive which will meet minimums.
IBM supplied a load resistor on some ATs. Seems like the PS didn't want to regulate if +12 was less than 1A or so. It has been a while, guess I'll pull the manual to refresh the memory.
Ed Fair
Give the wrong symptoms, get the wrong solutions.
IBM supplied a load resistor on some ATs. Seems like the PS didn't want to regulate if +12 was less than 1A or so. It has been a while, guess I'll pull the manual to refresh the memory.
Ed Fair
Give the wrong symptoms, get the wrong solutions.
Okay folk, let's move focus from the initial problem:
1. Connect MB
2. Test - power on then off
3. Connect CPU fan
4. Test - power on then off
5. Connect GPU fan (if any)
6. Test - power on then off
7. Connect HD
8. Test - boot to OS - shut down
9. Connect CDROM/DVD
10. Test - boot to OS - shut down
11. Connect CDR/W
12. Test - boot to OS then burn - shut down
13. Connect HD fan
14. Test - boot to OS and do benchmark on HD...
Cheers,
Jakob
1. Connect MB
2. Test - power on then off
3. Connect CPU fan
4. Test - power on then off
5. Connect GPU fan (if any)
6. Test - power on then off
7. Connect HD
8. Test - boot to OS - shut down
9. Connect CDROM/DVD
10. Test - boot to OS - shut down
11. Connect CDR/W
12. Test - boot to OS then burn - shut down
13. Connect HD fan
14. Test - boot to OS and do benchmark on HD...
Cheers,
Jakob
- Thread starter
- #15
Thanks for all the great suggestions! I finally found the problem. When I mounted the heat sink/fan assembly to the hard drive and installed it in the drive bay, the heat si was just touching the small board housing the on/off switch. That must have been causing the problem because when I removed the drive from the bay it worked. Just for the record, I hate Compaq chassis! They are the worst freakin design! In order to mount the hard drive so that the heat sink would not touch the board, i have to snap off the 2 metal plates that supply support to the drive. I don't see why these are even needed when the drive is screwed into the side rails.
Thanks again for all the help.
Thanks again for all the help.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
