HELP! New E4252 doesn't power up. bad PSU? switch? other?

Lilla · Post by **Lilla** » Wed May 19, 2004 10:55 pm

Hi all. I'm back. In Nov/Dec last year I build a computer using the E4252 case, Seasonic 400W Super Silencer, Panaflo L1A fans front/rear. Turned out great, with the help of Ralf and Mike and others in this great group. Recently, I replaced the 3 Zalman Fan Mate 1's with ATN-TP-02E three channel Fan Controller, after reading the good review by SPCReview on it. Was easy to install, works great. Looks nice, it's beige to match beige computer. I bought it from SVC.com $11.50 + 7 shipping. This computer is working good, no problems with it.

The problem I am having is with a different E4252 case. I decided to ask for help here because the problem might be with the PSU or power switch, or front panel connectors of this case. I'm at a loss as to what the problem could be. I don't know how to troubleshoot this. I'm hoping someone will have some suggestions for me, based upon the following details.

I was (and am still) building a computer for my neice out of left over parts. All went well. I installed the hardware, installed Windows XP, installed Office XP, set everything just so. It worked good. I was done. Time to put the case cover back on. I put it on, then pressed the power switch and and ... and nothing, absolutely no signs of life at all. It was working fine before I put the cover back on. What happened?

I checking and rechecking everything, tried everything. No luck. I concluded that the PSU chose that moment to die. The PSU was old, so it's not surprising that it went out, but it seemed odd that there were no warning signs that it was about to go out. Unfortunately, the PSU was hardwired to the front panel on the case, so both had to be replaced. I ordered an E4252 case with bundled 300W PSU and rear fan (figured it would be good enough for a budget computer, a humble Pentium 233 with 128MB memory) from NewEgg.

The new E4252 case with bundled PSU and rear fan arrived today. To start with I installed just the motherboard, video card, keyboard, hard drive, and turned it on. Nothing, no signs of life at all, same as happened with old case and old PSU. Hummm, made me wonder if the initial problem was with the PSU, or something else. But what could it be?

Anyway, I have a new PSU now, and it doesn't show any signs of life.

I unplug the power cord and rechecked everything.

Yes, the CPU power connector is plugged into the PSU connector.
The 20 pin connector is plugged to the motherboard and is firmly pressed.

I read a post that said you could unplug the hard drive until you got some sign of life from your power supply, so I did that to simplify.

I checked the front panel Power SW connector (that goes from front panel to motherboard). I tried flipping it around, no change. I removed all the other front panel connectors, figuring the Power SW connector is the only essential one. If this isn't true, please let me know. Tried again, no change.

What does this sound like? I mean if the power supply is good, shouldn't I at least see it turn a tiny bit, even if the motherboard / CPU were somehow not good. I mean it was all working great before I put the case cover back on.

I don't know how to troubleshoot this. Anyone got any ideas for me?

Thanks,
Lilla

luminous · Post by **luminous** » Thu May 20, 2004 2:32 am

With things like this, the best thing to do is start simple. Start at the wall socket and work towards the computer. Test everything that you come across and note the last point at which you have power.

It could be something really simple like a blown fuse in the PSU power lead. It could be a dodgy cable or multi block. These things have a limited life and can sometimes work intermittantly if it is a wire that has aged. Make sure you actually test the wall socket, they do break in the end as well.

Where possible always swap out bits for known good working items, eg multi blocks, PSU lead etc. I once had a dodgy PSU lead that would only work if it was bent in a certain way. When the case was off I had the machien pulled a little way forward, it worked. I put the case on and moved the machine and it would just stop.... (took me ages before I swapped out the PSU lead and found the problem)

sthayashi · Post by **sthayashi** » Thu May 20, 2004 5:38 am

There are a couple possibilities.

First, you could have a bad momentary switch. This can be easily diagnosed with another case (you don't even have to disassemble anything). Even better if you have a multimeter and a couple pins.

Second, you could have forgotten RAM. I've encountered some machines that don't power up without RAM.

Third, you could have a bad motherboard. Again, this can quickly be determined by plugging the PSU and case switch into a known good and working motherboard.

Fourth, you could be shorting something from the motherboard to the case. This happened to me on my first system build ever when I mounted the motherboard without standoffs. Needless to say, it didn't turn on and I couldn't understand why (until a friend pointed that out).

Post by **Ralf Hutter** » Thu May 20, 2004 5:55 am

I'd pull the board out of the case, set it on something nonconductive (like the mobo box, not the antistatic mobo bag). Disconnect everything but the CPU+HSF, one stick of memory and the Vidcard, keyboard and mouse and try again. Plug in the 20-pin ATX and aux P4 connectors from the PSU and briefly short across the 2 "power switch" pins and it should start up. If it doesn't even power up you're going to have to start switching out the real basic stuff like mobo and CPU, I'll assume one or both of your current PSUs is functioning. Post back with your results.

BTW - nice to hear from you again, too bad it couldn't be under better circumstances.

Lilla · Post by **Lilla** » Thu May 20, 2004 9:43 am

Thanks everyone for the replies. Thanks for the welcome back Ralf.

Other things I tried that I forgot to mention:
- I tried a different power cord.
- I tried a different outlet.
- I assured that the outlet was working.
- I tried a connector from a different clump of wires from the power supply to connect the CPU.

I want to try the workbench idea.

- I have a power supply I can pull from an old Dell D300 Pentium II (ATX). The motherboard went out on it, but the power supply was still working. It has a 20-pin connector (ATX) on it, and looks like a normal ATX power supply, expect that it has less power connectors than standard, that's Dell saving a nickel. It's not hardwired to the front panel or anything. Anyway, I'll use this power supply unless something thinks it's not a good idea. I don't want to disrupt my good E4252 system if I don't have to. It's used by other family members, and also I just don't want to take chance on messing it up.

- my workbench is an old unpainted wooden door, supported by a pair of wooden bookshelf on either end.

Is it safe to place the motherboard (once i remove it) on this unpainted wooden door workbench surface and connect it to the ATX power supply (pulled from Dell PII) on my workbench?

Thanks Lilla

luminous · Post by **luminous** » Thu May 20, 2004 9:59 am

Dell's PSU is not compatible with mobos that are not made by Dell (afaik). I don't know what the differences are, but it is possible to get an adaptor (again afaik) - although I don't know where from..

There is a mention of this issue on this page (just underneath the first pictures) http://www.quietpc.com/uk/psu.php

Hmmmm...I'm not sure this post has helped you much, other than me saying I don't know

I would check with other ppl b4 you start plugging in the Dell PSU and maybe causing damage to your components.

Lilla · Post by **Lilla** » Thu May 20, 2004 10:17 am

luminous wrote:Dell's PSU is not compatible with mobos that are not made by Dell (afaik). I don't know what the differences are, but it is possible to get an adaptor (again afaik) - although I don't know where from..

There is a mention of this issue on this page (just underneath the first pictures) http://www.quietpc.com/uk/psu.php

Hmmmm...I'm not sure this post has helped you much, other than me saying I don't know I would check with other ppl b4 you start plugging in the Dell PSU and maybe causing damage to your components.

Thank you for posting this. I will investigate further before using.

I won't be able to work on this again until this afternoon. I am leaving now to take my Aunt grocery shopping, she is 86. This is an every Thursday event.

Later,
Lilla

Post by **NeilBlanchard** » Thu May 20, 2004 10:31 am

Lilla:

If you plug in the Del PS to a non-Dell mobo -- you can kiss the mobo goodbye!

The Dell's use a different ATX "standard" (which makes it non-standard, right?) and they swapped some of the power pins on the 20-pin connector around...

luminous · Post by **luminous** » Thu May 20, 2004 10:32 am

Aye, we all grow old

At least when I'm old I will be able to order my food from Teasco.com

In fact, I'm not old and may start ordering from them anyway (some would call me lazy), but its only £5 for delivery

Post by **Ralf Hutter** » Thu May 20, 2004 11:10 am

Lilla wrote: I have a power supply I can pull from an old Dell D300 Pentium II (ATX). The motherboard went out on it, but the power supply was still working. It has a 20-pin connector (ATX) on it, and looks like a normal ATX power supply, expect that it has less power connectors than standard, that's Dell saving a nickel. It's not hardwired to the front panel or anything. Anyway, I'll use this power supply unless something thinks it's not a good idea. I don't want to disrupt my good E4252 system if I don't have to. It's used by other family members, and also I just don't want to take chance on messing it up.

I wouldn't do it, for the reasons mentioned in the couple of posts above. You'll likely fry your mobo.

Lilla wrote: my workbench is an old unpainted wooden door, supported by a pair of wooden bookshelf on either end.

Is it safe to place the motherboard (once i remove it) on this unpainted wooden door workbench surface.
Thanks Lilla

Yes, that'll work perfect.

Try the new PSU out of your latest case. You can check to see if it works w/o plugging it into the mobo. Basically, with the PSU plugged into the mains and the switch turned on, you short out the one green wire on the ATX connector to any of the ground wires (black)on the ATX connector and your PSU will start (fan(s) will turn). At least you can tell that your PSU is alive this way. Supposedly the safest way to do this is to have a load on the 5V rail so I plug a case fan into one of the molexes before I start up the PSU.

luminous · Post by **luminous** » Thu May 20, 2004 11:48 am

Ralf makes a good point here. Some PSU's must have a load connected to them if they are turned on. The better PSU's can cope without a load, the others can blow a fuse or worse. Its not a good move to connect important things when you are testing (ie a HDD with data) so fans are a good alternative (as mentioned).

vortex222 · Post by **vortex222** » Thu May 20, 2004 12:32 pm

i have a motherboard sitting in my parts bin that wont fire up a psu, because it has a broken trace from its front panel connector. if u got a new psu and case, that is likly not the problem.

another thing to consider with the evercase, is they have hard mounted motherboard standoffs, make sure that there isnt an extra standoff where the motherboard has no hole for it, and also make sure the mother board has no solderpoints that could come in contact with the roundish motherboard standoffs.

if that is the case, the motherboard may already be dead.

Post by **MikeC** » Thu May 20, 2004 2:43 pm

There's lots of good info here already... but here's one thing no one has suggested. Chances are, the board/cpu/ram -- is dead. Most likely the board, that's my bet. Could even be the CMOS battery.

Anyway, you can easily check if the PSUs are alive. All you need is an ordinary paper clip. Break it in half -- so you have 2 U shaped pieces of wire. Take the smaller of the two, and stick one end into the green wire lead and the other end into any black wire lead on the 20-pin connector of the PSU. This will jump start the PSU.

The PSU should be unplgged from the when when you put in the jumper and probably should be unplugged from any components to be perfectly safe.

PCGUY · Post by **PCGUY** » Thu May 20, 2004 3:25 pm

I would suggest what MikeC had mentioned for testing the PSU.

Also try to do what is known as minimum boot as suggested by Ralf - Here is an excellent article on the subject - http://www.motherboards.org/forums/view ... hp?t=39873

Good Luck!

Lilla · Post by **Lilla** » Thu May 20, 2004 6:05 pm

Thanks everyone! I appreciate the help.

Question:
All is unplugged from the PSU in the new E4252 case, and the motherboard is sitting on the plain wood workbench with Video Card connected. Do I need to remove the PSU from the case, or can I leave it in the case while I plug it into the motherboard that is sitting on the workbench?

luminous & Neil, thanks for advise not to connect mobo to Dell PSU. I taped a note to the unit so I won't forget.

sthayashi & vortex222, thanks for sharing thoughts about standoffs. Fortunately, I had read that unused standoffs could be a problem, so I put a plastic insert (looks like a plastic thumb tack) into each of the two unused holes. The plastic inserts were in the parts bag that shipped with the E4252 case. I took a wild guess that this was what they were for.

Ralf Hutter wrote:Try the new PSU out of your latest case. You can check to see if it works w/o plugging it into the mobo. Basically, with the PSU plugged into the mains and the switch turned on, you short out the one green wire on the ATX connector to any of the ground wires (black)on the ATX connector and your PSU will start (fan(s) will turn). At least you can tell that your PSU is alive this way. Supposedly the safest way to do this is to have a load on the 5V rail so I plug a case fan into one of the molexes before I start up the PSU.

Ralf, the CPU fan has a 4-pin molex plug. That's the only fan with molex plug I have. I don't have a 3-pin to molex converter needed to connect a case fan. Should i plug it in? Mike's way sounds simpler, he says unplug all components, and do the jumpering while the power is off.

MikeC wrote:Chances are, the board/cpu/ram -- is dead. Most likely the board, that's my bet. Could even be the CMOS battery.
Anyway, you can easily check if the PSUs are alive. All you need is an ordinary paper clip. Break it in half -- so you have 2 U shaped pieces of wire. Take the smaller of the two, and stick one end into the green wire lead and the other end into any black wire lead on the 20-pin connector of the PSU. This will jump start the PSU.

The PSU should be unplgged from the when when you put in the jumper and probably should be unplugged from any components to be perfectly safe.

Mike, I didn't know a dead CMOS battery could cause the system to act dead. I thought it would just stop keeping time, and perhaps loose it's BIOS settings. I have an extra battery so I can try replacing it when I am testing again. Thanks for giving me that idea to work with.

OK, I cut a paper clip as you said and I see that it is a perfect fit for jumpering two pins of the 20-pin ATX, but I have no green wire. The closest would be a blue wire?

The wires, as documented on the side of the power supply, are:

Ever 300W Power Supply, Model: MPT-301
Red +5 30A
Yellow +12V 15A
White -5V .05A
Blue -12V .8A
Brown +5VSB 3A <=== see motherboard note below
Orange P.G.
Black Grd.
Gray PS-ON

The ASUS TX-97E motherboard manual says this.
IMPORTANT: Make sure that the ATX power supply can take at least 10aAmp load on the 5Volt standby lead (5VSB). You may experience difficult in powering on your system without this.

This PSU supplies more than what is required, correct? If not, please advise.

Lilla

sthayashi · Post by **sthayashi** » Thu May 20, 2004 7:54 pm

What's usually the green wire is the 4th pin down on the right side, assuming you're looking STRAIGHT at the pins with the connector clip on the right. Yours is the first I've heard of that didn't use green for PS-ON (which is labeled gray on yours).

On mine, Gray is PWR_ON.

Post by **MikeC** » Thu May 20, 2004 8:25 pm

Gray PS-ON

Sounds right. 4th pin sounds right too. I have seen some grey ON wires instead of green.

Pate · Post by **Pate** » Thu May 20, 2004 9:57 pm

Have you tried booting without the video card? If not, that might be worth a try.

Sorry if you have already tried this..

Pate

Lilla · Post by **Lilla** » Fri May 21, 2004 8:17 am

Progress report. GOOD NEWS. From the workbench, I am able to power the CPU, a fan, and a hard drive. It's a start. Details below.

First, for any other newbies who might stumble upon this thread. I first tried to connect the PSU cables to the motherboard lying on the workbench while the PSU was still in the case. The ATX cable was too short to reach. The other PSU cables I needed were long enough, but the ATX cable was too short. Duh! Newbies!

Here's what I did, and I got power...

Remove motherboard form computer case, put it on workbench
Removed PSU from computer case, put it on workbench
Connected PSU 20-pin ATX connector to motherboard
Connected PSU 4-pin molex connector to CPU fan
Connected PSU 4-pin molex connector to old hard drive *
Connected PSU 3-pin connector to old case fan *
Replaced CMOS battery, just in case (per MikeC), since I had a spare
no-video card connected

* I figured this would add some load for the Power Supply in case that would be helpful. Ralf and others said that some lower-quality PSUs could be damaged if started without a load. This PSU came bunded with PSU & fan for only $15 more than bare case, so can't expect too much. I decided to play it safe and supply a load so as not to damage the PSU if it happens to require a load.

I turned the mains power switch on the back of the power supply to the ON position.

Now for the big moment... I used a small, flat bladed screw driver and short together the two pins on the motherboard that would normally receive the on/off switch from the front panel of the case.

When I first touched the screw driver to the two pins nothing happened, but I decided to hold it there for a bit (maybe 2+ seconds) and to my surprise, just when I was ready to conclude that it wasn't working, everything started up and ... the CPU fan turned, the case fan turned, the hard drive started up.

FYI, since this worked I won't need to use Mike's homemade paper clip jumper to test this power supply. But, I will keep track of that method for future reference.

QUESTION:
I didn't know how to stop it so I just turned it off using the power switch on the back of the PSU. It that the best way to stop it all? Or was I suppose to touch the screw driver blade to the pins again to stop it?

Next I will try adding the video card and see if that works. But, I'd sort of like to know if I am stopping things correctly first.

Lilla

luminous · Post by **luminous** » Fri May 21, 2004 8:49 am

When you shorted the pins with your screwdriver you turned the machine on just as if the power switch had been pressed. As the machine was started in the normal manner, turning it off is also in the normal manner ie:-

If you are fiddling with components/in the BIOS so that the machine has not booted into an OS then the power switch on the front can be used. The switch on the PSU is also OK.

If you have connected sufficient components to the machine so that it actually boots into Windows then it is best to click start, then shutdown etc.

sthayashi · Post by **sthayashi** » Fri May 21, 2004 9:08 am

Lilla wrote:I didn't know how to stop it so I just turned it off using the power switch on the back of the PSU. It that the best way to stop it all? Or was I suppose to touch the screw driver blade to the pins again to stop it?

In an ideal situation, yes you should have touched the screwdrive blade back to the pins, since that's the 'proper' way to shut down a computer normally when all else fails. But truth be told, I don't think it makes a difference whatsoever. Many years ago, we shut down our computers by flipping the switch on a surge-protector, never had a problem due to it, and it's essentially the same as pulling the plug.

It seems very weird that it took so long to start up when you shorted the pins. There's something mighty suspicious about that.

Lilla · Post by **Lilla** » Fri May 21, 2004 9:11 am

Progress Report. Added in video card and monitor and it works.

A message on the monitor says "CMOS checksum error - Defaults loaded". I didn't clear CMOS via motherboard switch, but I did replace the battery. Perhaps the old battery needed to be replaced and caused this error? Perhaps replacing the battery (regardless of whether it needed to be replaced or not) caused this error? Perhaps because components are not as expected caused this error? Fortunately when I optimized the BIOS settings I documented all BIOS settings.

luminous wrote:If you are fiddling with components/in the BIOS so that the machine has not booted into an OS then the power switch on the front can be used. The switch on the PSU is also OK.

Thanks for confirming that the switch on the PSU is OK to power down with after starting by shorting power on pins on the motherboard. I am still on the workbench. Computer case is not in the picture yet. The hard drive does not have an OS installed. It was recently reformatted, so I am not booting into an OS.

Lilla

Post by **NeilBlanchard** » Fri May 21, 2004 9:25 am

Hello Lilla:

Replacing the battery *will* clear the BIOS!

Glad to hear that there is progress and positive results...

Lilla · Post by **Lilla** » Fri May 21, 2004 10:51 am

Followup and confession.

FYI. I switched the CMOS battery back to the original and set some BIOS settings and it is holding those settings so original CMOS battery was OK.

NeilBlanchard wrote:Hello Lilla:
Replacing the battery *will* clear the BIOS! Glad to hear that there is progress and positive results...

Thanks for the support.

OK, time for true confessions. I found the root problem. Me! You can add this one to the list of things to check "when the computer won't power up after connecting motherboard to front panel of computer case" list.

The old case had a 3-pin Power-On connector. The new case has a 2-pin power-on connector. The manual brackets 3-pins for Power-On switch, but upon closer inspection I see that it shows that only two out of the three pins are used. With the new case, I MUST skip the first pin since it provides a 2-pin connector.

I had checked the manual any number of times, but this was so subtle that I missed it. Once it was on the workbench I was looking closely at the diagram of the pins in the manual to see which pins to bridge with the screw driver and there it was - the light bulb in my head went off. When the motherboard was in the case I was connecting the front panel power switch connector to the wrong two out of three pins. If you could see the diagram you would understand how easy it was to miss this.

Know I'm feeling confident that the initial problem (old case and old power supply) was indeed a bad power supply. So, I'm on the right path now with new E4252 case and power supply.

I almost glad it happened, otherwise I would not have learned now to do the bit of workbench testing that I did. This is good stuff to know.

Thanks to everyone in this thread for sharing your experience and advise and for helping me troubleshoot this problem.

There's a next part to this install, but I'll make that in a separate post in this same thread.

Lilla

Post by **Ralf Hutter** » Fri May 21, 2004 11:13 am

Every build I make, I initially make with the board out of the case with a minimal hardware setup, including starting the board by shorting the pins with a screw driver. Initial setup is CPU, HSF, 1 stick of memory, vidcard, keyboard and mouse. I boot into the BIOS, do a little configuring and generally make sure everything works OK. Then I put in all the memory I'm going to use and hook up a floppy drive so I can run Memtest86 to make sure the memory subsystem is working correctly. Only then do I hook up the HDD and install an OS. At that point I'll install the hardware into the case and hook up whatever peripherals I'm going to use. This assures me that the base components are working fine. It doesn't take much extra time to do this and potentially saves hours of headaches down the road.

Lilla · Post by **Lilla** » Fri May 21, 2004 12:01 pm

Ok, moving on... I think I'm going to have to mod the new E4252 case to fit this old motherboard. The motherboard is ASUS TX97-E.

I've got a general idea for accomplishing it but perhaps there's a better way. Decided to see if perhaps someonee has experience or better ideas on how to go about this, or just a vote of confidence that my plans are reasonable.

Here's the problem.

The side of the E4252 case that the motherboard attaches too has a section that is raised about 5/8" and is angled. The 5.25" drive cage attaches to this area. The motherboard fits in a potato chip fashion because one corner of the motherboard is held higher than the standoffs because it comes to rest on the corner of the corner of the 5/8" raised angled section.

Anyway, my idea is to drill holes close together in the corner of the 5/8" raised section to create a slot that the corner of the motherboard can fit into so it can lie flat - instead of like a potato chip.

Wish I had a digital camera so I could show a picture.

Anyway, I hope I've described it adequately.

I'm going to ask my Dad to do the drilling.

Any ideas or thoughts on this before we perform the operation?

Thanks,
Lilla

luminous · Post by **luminous** » Fri May 21, 2004 12:42 pm

Wooohoooo !!! So often these threads end up with a dead computer being confirmed. Glad to see that yours is all fine and well

I've no thoughts on the drilling bit, as a picture paints a thousand words - any you haven't used 1000 so I don't have a picture

Seriously, I'm not able to follow the last bit. Its been far too long since I fiddled with old machines

Lilla · Post by **Lilla** » Fri May 21, 2004 3:08 pm

Follow up: The surgery is over and the motherboard now fits nicely in the E4252 case.

I made a card board template of the motherboard to use instead of the motherboard to determine what metal to cut away.

Dad had the right metal saw to make this cut, it was a circular saw. It worked great. Then he filed the jagged edges so it won't cut fingers. He did a nice job, the hole even looks nice.

To prevent metal dust damage I first covered the front panel components, removed fan. PSU was already removed. Removed both sides (of course) and front panel. Covered the top to prevent it from getting scratched.

Afterwards, I gave it a good vacumning to remove metal dust particles. Then I put the motherboard in place to see for sure. Yep! it fits perfectly now.

But, the funs not over. The back I/O template doesn't match this old motherboard so next step is to make one for it. Back to the drawing board.

Lilla

Post by **NeilBlanchard** » Fri May 21, 2004 6:41 pm

Hello Lilla:

Lilla wrote:But, the funs not over. The back I/O template doesn't match this old motherboard so next step is to make one for it. Back to the drawing
board.

Can't you use the one from the old case? Or was it not an ATX case?

Post by **Ralf Hutter** » Sat May 22, 2004 5:30 am

Lilla wrote:But, the funs not over. The back I/O template doesn't match this old motherboard so next step is to make one for it. Back to the drawing board.

Lilla

Great idea about cutting away the angled section of the case to give yourself clearance!

If you're looking for I/O panels, you can get them here, although it may also be worth your while to visit some "mom-n-pop" computer stores and ask if they've got any laying around.

Isn't that TX-97 an old AT/ATX "Franken-board"? A lot of those have no standardized ATX type of connectors on them so you really may have to carve up something on your own. Have fun!