PDA

View Full Version : yet another dead p.c.


fp2nc
10-26-2002, 04:52 PM
Hi everybody I have a dead p.c.-no beeps and wouldn,t boot to boot floppy.I thought the problem was memory related so I pulled the ram and tried to put each stick in one at a time but when I try to do this the MB flexes so much it feels like it might crack. This has on-board video so I've lost my video signal. I think I will pull everything out and put on cardboard to start problem solving. I won't go into system specs yet. I've looked at about 7 pages of this forum and after reading posts that appear related to my problem, I thought I would try the usual suggestions first but I have a few questions as I'm new to this kind of problem.
1) I have a wrist strap for grounding that I clip to the case. I want to install a new CMOS battery but do I need to unplug the p.c. from the wall before I replace it? If I do, what will ground the case for my wrist strap?
2) Although I wasn't getting any beeps at attempted start-up, the lights on the 2 cd drives would light up but not the floppy drive. I have a Dell p.c.-can I switch floppy drives or are Dell's unique?
3) I have looked all over the MB but can't find an identifier. Is there some way I can identify this board?
I have to leave now but will check back tomorrow to(hopefully) read any responses.I'm sure I'll have a lot more questions in the future. Thanx in advance.

Budfred
10-26-2002, 05:39 PM
Welcome to PCGuide!!

Some suggestions:

1. You can ground yourself to any grounded metal surface, so it doesn't have to be the pc case. However, if you have a surge protector with a switch, you can turn off the switch which will protect you from any power getting through and the case will still be grounded.

2. You probably don't have to worry about replacing the floppy, the light only comes on when it is being read and it probably didn't get far enough to do that. If you do find that it is bad, you can probably replace it since it is unlikely Dell would bother with propietary floppy drives.

3. If you know what type of Dell you have, you can probably find the type of mobo on their web site.

Other suggestions:

Swapping the RAM around is a good idea, but usually not the problem when you get no POST. Things that could cause no POST include: dead mobo, dead cpu, dead power supply, bad connections, and other things that don't occur to me right now. I would try disconnecting things in the box before taking everything out. Try a minimal boot with only video, mobo, RAM, keyboard, mouse, and power. If still no post, then take everything out and try outside of case. Keep in mind that if this is a newer Dell, you will be much better off having them deal with it since you could end up voiding a warranty.

Good luck,
Budfred

fp2nc
10-26-2002, 06:03 PM
Hi Budfred Thanks for replying. Although I said was leaving, I'm just now getting ready to walk out the door.
1) Great idea. I will do it.
2)I think I gave the wrong impression-the problem p.c. isn't a Dell,I just happen to have a couple of Dell-P2's laying around.I was thinking about swapping the Dell floppy in to see if it would light up up when I turned it on.I thought since the cd-drives lit up the floppy might also but after reading your post I guess this isn't the case.
3)Since I don't know the brand of the MB I don't know where to look on the web.
My wanting to swap floppy drives was my way of checking the power supply(assuming that the drive is functioning) but I'm wondering if there is any way I can find out how to check the PS with a multi-meter. I don't really know how to use it or what I should be checking for. Is there any site on the web that would be able to help me? I'm wondering what values I should see when I check something.Is there anything I shouldn't check with the meter because of safety concerns?
I'll be back tomorrow. Thanks again.

Budfred
10-26-2002, 06:14 PM
Yep, I did misunderstand. If you can post more info about your computer, we may be able to help you figure out the mobo type. However, if it is an unknown brand, that will be tricky. You can find out on a running PC with software, but on a dead one, you will need to find some distinguishing number or logos to find out what it is.

For the power supply, you can definitely use a multimeter and shouldn't need any precautions other than normal electical precautions. You already know that you are getting some power since the CDs light up. You can also tell by whether or not your fans are running. If your CPU fan is dead, that could be the cause of the whole problem and might simply be a dead fan. I would check for that kind of stuff first. To use a multimeter I would search the archives here for the procedure, but it is likely that someone will be along soon to tell you anyway, so stay tuned...:D

Budfred

Fruss Tray Ted
10-26-2002, 07:08 PM
Please be a little more specific.
You can ground yourself to any grounded metal surface, so it doesn't have to be the pc case.
But it must be something that is connected to the pc AND in succession be connected to each part as well. I mean ironically you wouldn't want to ground yourself to your stove then figure you are OK to work on the pc. The reason for grounding is so that you and the pc (and parts going in/out) are not at a difference in potential voltage (neither you nor the pc has a static charge)

If you BOTH have the same charge that is OK too because there would be no current flow when touched. But we don't need to go into that right now.

Also most recommendations are to unplug the pc from the wall before doing any work on it. Sometimes I do othertimes not. As long as a switch is turned off at the plug end of the pc cords (monitor, tower, printer etc.)

If either of you want to learn to use multitesters here's a good start:

http://www.7volts.com/atx2.jpg

Budfred
10-26-2002, 07:43 PM
Ted,

I have never seen that recommendation before, every reference I have ever seen says to connect to any grounded metal and says nothing about needing to match potential voltage. Do you have a reference where I can read more about this or does anyone else know anything about this??

Also, if you do have to ground to the same device you are working on, how do you do this with the device unplugged.

Budfred:confused:

Fruss Tray Ted
10-27-2002, 02:28 AM
OK,
I will try to explain a lttle...

Per my pic, If I were to be connected to pin #1 with a wrist strap (+3.3vdc) and I touched (or worked on) pin #2 (+3.3vdc again) there is zero difference in 'potential' or voltage (If you checked with a meter there would be zero volts between the two). Therefore no current would be caused and no damage could happen if work was to be performed between the 2 'ports'. Also: why power wires are used by nesting birds as long as they do not offer a 'difference' in 'potential' by providing a path to 'ground' (physically) there is no damaging current flow. This is why they are so far above 'ground' AND apart from each other.

As to my recommendation, when "HASN'T" the advice been to unplug the electrical device? Pc's, blenders, appliances?

I think the confusion here is the difference between 'earthen ground' and 'grounding' one (self or device)to another. Two entirely different horses.

You'd think after a some courses and 30+ years to hash it over, I'd be able to explain it better, but I keep tryin'...

Basic electricity 101

coils fields magnetism dissipation inductance impedances, don't ask me, I suck at HTML!!! :D

try again...

Thunderstorms: The friction of the air and whatever it is carrying causes a charge. this is one of various things called a 'potential' or voltage. (When your hair is on end, there's a potential between you and 'ground'. As soon as you touch something, the dissipation of voltage results in an amount of current that sometimes items like RAM cannot survive as programmed. Voltage alone does not do anything, nor does it kill you. It is current alone that does any AND ALL the damage. That is caused when one: the sudden change of volts between two objects coming into contact with each other come into contact and 'ground' each other to themselves.

As you pat your cat (friction) and you and the fur gets magnetized, you are building a charge (a potential or voltage difference between the outside of the cat's fur (and you) and the cat's 'ground' or self. When you touch the cat's nose you are equalizing yourself to the cat's (inner) state of charge. BANG! Current discharge until the voltages equalize. You are then grounded to the cat (nose). If you hold the cat's paw as you pat it, the buildup is less, due to the continual low level dissipation of built up charge (which is the same effect as a wrist/ground strap).

(I STILL suck at HTML) :(

TheSkippyKing
10-27-2002, 03:10 AM
After reading this "zap your comp" tread I wandered into a theoretical question. If your comp is unplugged & you do a Michael Jackson moonwalk thingie across the carpet and then touch the frame of the comp , does the resulting static discharge effect the PC's components?

Budfred
10-27-2002, 04:12 AM
Ted,

I read your post several times and I still don't get it. I understand how static charge builds up and I understand that you need to be grounded to get electrocuted. However, my understanding of grounding for ESD is that you dissipate any electrical charge as it builds up by being grounded to any grounded metal, so it wasn't necessary to be grounded to the particular computer you are working on. And again, everything else I have ever seen about this says the same thing; ground yourself by touching the case or preferably wear a grounding strap connected to grounded metal. If I understand you correctly, you are saying that being grounded to anything other than the box you are working on may cause ESD and since it is best to unplug the box you are working on, there is really no way to get grounded then. This seems paradoxical and I really would like to know if this is the way it works.

Budfred:confused: :confused: :confused:

david eaton
10-27-2002, 07:02 AM
Ideally, both you and the case should be connected to ground, through the correct lead. Picture below.

These leads contain a resistor to limit the possible discharge current to a safe value. even better, cover the workbench with a conductive anti static mat connected to ground, with a wrist strap connected to the mat. Kits containing cords, wrist straps, mats and a power plug with snap connector that will accept only a correct lead.

Pictures from Maplin Electronics.

Fruss Tray Ted
10-27-2002, 11:32 AM
Michael Jackson moonwalk thingie across the carpet and then touch the frame of the comp , does the resulting static discharge effect the PC's components?
If the pc is NOT grounded to earth, quite possibly. If it IS grounded to earth then since current chooses the path of least resistance, no. It will pass the current through the case to earth.


Budfred
I'm not trying to contradict nor confuse you, but am just using this as an opportunity to discuss electricity possibly for those readers that may gain a little from it. Yes you are right when you state if you and the pc are grounded (to earth), all charge gets dissipated. But once you unplug or unsolder anything from 'earth' ground, this is where I meant that it is better to be directly connected to the pc or ultimately, as David said, attached to a grounding 'mat' which everything is placed into physical contact with. Also with the pc unplugged, Murphy's law and the power switch cannot come into play.

Back to the issue:
What I am saying is more like charactersitics of electrical energy and not just "grounding techniques" although it is all part of it.

English being what it is and words having multiple meanings add to the confusion at times. Ground in one sense means earth/dirt/rock/planet. In another it merely means a connection to another object. Where it is most common to have the negative side of an electrical circuit as 'common' or 'ground' it is/was not always so.

Does anyone remmember from the sixties or so, I think it was Chrysler's that had (btw left hand threaded lugnuts & studs on the left side of the vehicle) but more importantly positively grounded wiring systems. If you needed to jump start the engine because of a dead battery, the red lead of the cable needed to be attached to the chassis and the black (negative) to the 'hot' terminal of the battery. If for some reason, the cars bumpers came into contact, BOTH vehicles batteries would be shorted in this situation. BANG!

If you notice in the ATX2 pic most voltage points are positive. Therefore the negative lead of a tester is attached to ground to take the reading. BUT! If you want to take a reading of one of the 'negative legs' of the power supply, your positive lead of the tester then needs to be attached to the ground (or chassis or what WAS negative) and the negative lead to the point where the voltage is on the connector/component.

So just because the lead is black or is designated negative it is not always 'ground'. Best bet is to attach directly or use a mat.

Here (http://www.7volts.com/atxnotes.htm) is some good reading for anyone interested.

TheSkippyKing
10-27-2002, 12:34 PM
Thanks for the reply...AND the link to this ,
While we're at it, you can power up an ATX power supply by shorting the green wire with any black ground wire.

Always wondered by what "magic" the pwr swtch contacts turned on the power supply , but never had the ambition to actually follow the traces to their destination.
There goes another coupla grey cells used up for "storage"

fp2nc
10-27-2002, 06:06 PM
Thank you guys for the info on grounding. I'm still not sure what to do regarding protecting the p.c. from static electricity. I have a wrist strap with a clip on the end. Will this suffice?
Regarding the MB identification, here's what I found-Amibios ez73433,Sis 5595, Sis 530-pc100, 1 isa slot, 2 pci slots, 3 memory slots and the board is gold in color. The proc is a K6-2. PS and proc fans both work.When turning switch on, the lights on the two cd drives light but not the floppy light and there are no audible beeps.
My biggest problem right now(I think) is trying to get a memory stick installed in slot #1 so I can get the on-board video working again. The latch on each end of the slot start to close on the stick but don't latch all the way. I can push the latches into place but is the stick really seated in the slot? It feels like the mb is flexing too much.Any ideas on what I should do will be appreciated.

Fruss Tray Ted
10-27-2002, 07:14 PM
I'm still not sure what to do regarding protecting the p.c. from static electricity
Is your hair standing on end? No? Stop worrying about it and just connect the strap to the case. It'll be fine. When there's no sensitive or valuable data on my pc's and I start tinkering, all I do is rest my wrists on the case as I work. I have not fried anything yet. Most of this stuff is for obsessed insurance that nothing can go wrong. Now around here in the dead of winter when I pull my stocking cap off to find my hair standing up like Phyllis Diller(sp?) :eek: I won't be digging in my heels to deeply into the insides of a pc. At least not until a shower and other static neutralizing steps are taken.

To install your memory:
If pulling the motherboard is not an option, you could try several layers of corrugated cardboard or rubber type material cut to fit behind the motherboard and in front of the back plate. Then remove before plugging anything back in electrically. Just don't use anythging that will conduct electricity.

Also with your fans running but no post, could you take out everything except the one stick of memory, your keyboard and your harddrive. Try to boot with that. Post back results. Try without the harddrive too if no go and try to get into BIOS.

fp2nc
10-27-2002, 08:42 PM
Thanks Ted I feel better about the static concern. I was trying to figure out what to stuff under the MB but hadn't thought of cardboard-will give it a try.I'll try your other suggestions tonite and post back tomorrow.Thanks for the help.

fp2nc
10-27-2002, 11:36 PM
Just re-read your last post. Do you mean I should pull the ribbon cables from everything except the harddrive? Should I also pull the cable on the on-board video?

integral
10-30-2002, 01:54 AM
If you notice in the ATX2 pic most voltage points are positive. Therefore the negative lead of a tester is attached to ground to take the reading. BUT! If you want to take a reading of one of the 'negative legs' of the power supply, your positive lead of the tester then needs to be attached to the ground (or chassis or what WAS negative) and the negative lead to the point where the voltage is on the connector/component

Showing your age here, this is a must for old school analog meters (like mine) but most modern meters are digital, you can connect the leads any way you wish, the display will show negative or postitive.

Fruss Tray Ted
10-30-2002, 07:53 AM
Integral,
I know that digital meters can read either way but analog meters are still available. The explanation is to give an idea of how current flows. But that doesn't change the fact that I'm OLD! Now where's that cane I was fixin' to HIT you with? LOL! :D (BTW I have both)

fp2nc,
Sorry, I just noticed a reply by you that I missed, yes, disconnect ALL your drives, (CD-ROM, DVD, floppy, extra cards remove them,). What do you mean the cable to the onboard video? I have NONE with my onboard vid/sound board. Leave the cable to your monitor alone and leave your keyboard attached. A PS-2 mouse is OK but if optical, remove it.

If you can get to BIOS with just your monitor, mobo and PS, then add one component at a time (powered down of course) until it won't. That is if you get to a BIOS screen at all.

Do you have another pc available to check anything with? Your monitor for starters?

Grumper
01-27-2003, 09:43 PM
The way to solve alllllllllllllll of your problems is run a single #12 copper wire from any water pipe [like a washing machine does] or matalic electicial box or ground wire fron an outlet or your main panel.
and attach it to the frame of computer. This way you can pull the computer plug and go to work---I would still use a wrist strap.

mjc
01-27-2003, 11:27 PM
Easier way, use a power strip with a light, plug the machine into it and then plug it into a grounded outlet. Kill the switch (light out), the computer will still be grounded and no power.

Grumper
01-28-2003, 06:21 AM
Now thats a great way to do the same job------ I was reading through the thread and the back and fourth was "interesting" my reply [water pipe potential] was sort of tongue in cheek.

Sylvander
01-28-2003, 07:14 AM
Hello fp2nc.

May I butt in?
Here’s what my diagnostic charts say about your problem.
They may confirm some of what you’ve been told already.

START UP TEST
1. Is the PC dead? No.
2. Did POST appear to run? No.
3. Were there any audible beeps? No.
4. The Power-On Self-Test failed to complete. [NO POST]

NO POST
1. Are the PSU voltages correct? Check them if you can. If they are not you may have a faulty PSU. Assuming they ARE ok then:
2. Disconnect all peripherals and system adaptor cards except the keyboard and video card/monitor.
3. Does the POST now run? If it doesn’t then your system board [or something on it (video card, CPU, memory)] is faulty. If it does run then:
4. Re-connect the floppy disk drive and run advanced diagnostic software from a floppy disk. Get “TuffTest-Lit” from http://www.tufftest.com/free.htm, or better if you can. This is simple and basic.
5. Does the system board pass? If not then it is faulty. If it does pass then:
6. Switch of and re-connect ONE peripheral/adaptor card.
7. Does POST now run? If it doesn’t this last device connected is faulty and should be disconnected. If it does run the device should be tested using the diagnostics and if it fails is faulty and should be disconnected. If it passes it's ok and should be included.
8. Repeat from step 6 until all devices are tested, all faulty are excluded and all working included, then re-test the system.

If you would like a copy of these charts, then send me a personal message giving an e-mail address I could send them to.

I liked what MJC said and was wondering when someone was going to say it.
I've seen contradictory instructions in books about disconnecting from the mains or not, with suggestions to remain connected but SWITCH OFF as one. His was even better.
Seems to me that to be truly grounded [that is to be at ZERO POTENTIAL by virtue of being connected to planet earth's soil] both the PC and the user should be at zero potential relative to earth.
You wouldn't want the user and the PC to be at an equal potential but different to earth, with the possibility that they both short to earth with a bit of a bang.
[Whem you touch an earthed desk lamp]

Budfred
01-28-2003, 11:40 AM
Has anyone noticed that the previous last post on this thread was October 30, 2002????

I suspect the issue is either resolved or dismissed by now.

Budfred