RE: [SLE] Need help diagnosing hardware problem
On Monday, October 16, 2006 @ 11:30 PM, I wrote: <snip> I'll try unplugging and re-plugging the video card... <snip> Well, it's been a week now and no more video problems. Looks like pulling the video card, vacuuming out the slot, wiping down the connectors on the card, and putting it back in was all that was necessary to fix the problem. I guess a little dust somehow got in there and caused a bad connection. Hard to believe that's all it took to fix it. Anyway, in future if I have any problems along this line and doing that doesn't fix it, I know what to expect a technician to check (power supply, CMOS battery, connections all around, etc.), so I did learn a great deal from this thread for down the line. Thanks to all who provided input on this problem. Greg Wallace
On Monday 23 October 2006 15:58, Greg Wallace wrote:
Fixed a neighbor's computer that lost video by just loosening the slot screw holding the video card down and pushing the card down in the slot. Had to remove the card and adjust its slot cover for a longer term fix. What happens over time when you power a system on and off is that metal connectors and connections expand and contract with the heating and cooling cycles. Eventually they will move. How much and whether they lose connection you don't know. All warnings about static electricity and grounding apply and only attempt this with power the cord removed: Fully removing and reseating cards, cable connections, etc can usually resolve a whole lot of problems. Stan
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Monday 2006-10-23 at 17:02 -0500, Stan Glasoe wrote:
Other people recommend leaving the cord connected, with the switch in the off position: this way the earth wire remains connected. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFPUa3tTMYHG2NR9URAkKjAJ0YBfhudyEvk4YP0x8Xjf7Y9W6ozgCffMXR ueYy5u0SYb/eG7EaKzIXguw= =N8D9 -----END PGP SIGNATURE-----
On Monday 23 October 2006 17:48, Carlos E. R. wrote:
_IF_ there is a power switch _on_ the power supply near the power cord! The main on/off switch on the front of the system doesn't remove all power from the system and not all power supplies have their own on/off switch. With static it is important to have all parts at the same potential not that the static is eliminated. You, the machine and whatever part you are adding/removing/reseating should be at the same static 'level'. That way no discharge occurs. Stan
On Monday 23 October 2006 17:48, Carlos E. R. wrote:
The magic that burns electrostatic sensitive parts is amount of electrostatic charge that will go trough the part driven by potential (voltage) difference between human body and the part. Not every static is damaging, but in the winter our skin is dry enough to load quite a bit. Attached ground to the chassis will not help if operator forget to use antistatic strap attached to the metal part of the case, but then the for the safety reasons cord must be unplugged from the wall outlet, because you are accessing malfunctioning device where you have no idea what is wrong. That means you have to find a ground that is for sure ground, and not "it should be ground", because "it could be line voltage" as well. Second reason not to use this method is construction of todays computers, where "switch" is just a signal key that tells motherboard to turn power supply on. So there is no real separation of electric power (110 VAC) like it was some 10 years ago. Thanks to Stan to remind us on this. Groundign everything and using antistatic straps is valid in industry where worker/operator has to touch many objects and in the same time has to focus on job, not on ESD (ElectroStatic Discharge) problems. On a single device it will work without straps, but with unplugged power cord, as long as repairer has in mind that discharging is not one time process, but it has to be done after each contact with something outside the computer case. Did I mentioned, that disconnected cord helps a lot against bad luck. I'm not that superstitious, but by now it helped better than a rabbit foot. -- Regards, Rajko M.
On Mon, 2006-10-23 at 20:52 -0500, Rajko M wrote:
230 V~ 50Hz Where do they have 110 V~ ???
-- /Peo -- Registered Linux User #432116, get counted at http://counter.li.org -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
On Tue, 2006-10-24 at 00:28 -0800, John Andersen wrote:
Mostly to the local pub. :-)
-- /Cheers Peo -- Registered Linux User #432116, get counted at http://counter.li.org -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
Peo Nilsson wrote:
Where do they have 110 V~ ???
North America. Here we have a 240/120V distribution system. You'd use 120 for small appliances, lighting etc. and 240 for heavy appliances such as stoves etc. The power feed to homes is 240V with a grounded center tap or neutral, so there's 120V either side of neutral.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Tuesday 2006-10-24 at 07:37 -0400, James Knott wrote:
Dual phase? Curious. Here (Spain) we have a three phase system. Voltage is 230 (was 220) live to neutral, with 230*sqroot(3) = 400 Volts from phase to phase. We have 230 for homes, and 400 (3 phases) for industries and heavy loads like lifts or air condiotining. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFPhRftTMYHG2NR9URAlPJAJ0WWM13qKME5c8Qb0bvFqyMglFDUgCfYWdl vPbMN+XcQVymzUh6ZK3XYDA= =CLDf -----END PGP SIGNATURE-----
Carlos E. R. wrote:
Hope this ends the interesting OT discussion :) <http://en.wikipedia.org/wiki/Domestic_AC_power_plugs_and_sockets#World_maps_by_plug.2Fsocket_and_voltage.2Ffrequency> Cheers, Dave
Dave Howorth wrote:
Perhaps not that OT as you would imagine :-) . Let me start with this: I have a darkroom set up and my friend asked if he could borrow for a few weeks my enlarger control unit (which determines the exposure time for enlargements and then switched off the lamp after the calculated period of seconds). I had to resort to the manual method of doing test strips to find the correct exposure time. I soon discovered that I was wasting paper because what the test strip showed as the right time for an enlargement produced a useless actual print. Now, I am not *that* bad in the darkroom, and therefore started to do some testing. I discovered that the power supply--we have 230v/50hz--was fluctuating very badly in the downward direction at certain times of the day and night and NOT--which really surprised me--during the normal times when households are cooking, showering, or watching TV. I am talking here even after midnight and in the early hours of the morning when people are asleep. Anyway, the point here is that the power supply drops well beyond the accepted 10% limit at which damage to electronic gizmos is (deemed) probable. (I even called our electricity provider; an inspector came out, stuck a gizmo on the power line for a week- and confirmed my findings. A transformer some 500 meters up the line was eventually replaced.) From the computer point of view: how many people just buy a box at a shop with a prepackaged $15 power supply and then merrily run their computer for hours? Does the PSU regulate the voltage sufficiently to provide the safety needed for such a delicate and complex bit of equipment? OK, over to you for an interesting discussion :-) . Cheers. -- "Our enemies are innovative and resourceful, and so are we. They never stop thinking about new ways to harm our country and our people, and neither do we." George W Bush 5 August 2004.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Wednesday 2006-10-25 at 00:34 +1000, Basil Chupin wrote:
They should, but they might not. When I design a psu for my use, the safety margins are as big as I can, although there are limits on the overvoltage the transformer can handle without burning. When a manufacturer designs a PSU with price in mind... the margins may be much narrower, down to the "legal" limit. In theory, if the overvoltage is too big to handle, they should just switch off, but they might not have that protection; in that case, they should die on the post, while on duty. Then again, that self destruction should not destroy the electronics downstream, but that, again, might not happen. Me, I don't design for such an overvoltage, I'm afraid: mine would die. It simply should not happen, so we don't design for it normally. In case of undervoltage... I'm not sure what would happen with switched-mode power supplies used by computers. In the best case the output voltage would go down, or maybe it wouldn't cope with the current needed. And... ;-) http://en.wikipedia.org/wiki/Power_supply - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFPlpUtTMYHG2NR9URAvtuAJ9s6lv1Rt7eRn9U6JxHiknSzMC0bQCfYBDi nn29RG7qMCbffU5o6TTsnPs= =ZepQ -----END PGP SIGNATURE-----
On Tuesday 24 October 2006 08:42, Dave Howorth wrote:
Hope this ends the interesting OT discussion :)
What is OT is relative. We talk about computer hardware etc, and that is close to topic as it can be. You can't run Linux if your hardware doesn't work. Than, if we take that it is OT, we don't disturb normal conversation as this is some 200 mails deep from current new mails. Even I missed some interesting conversation yesterday and found thread again today :-) BTW, it is time to look how to filter some threads in separate folder, so that I can find them again, easy. -- Regards, Rajko M.
On Wednesday 25 October 2006 16:48, Rajko M wrote:
You can't run Linux if your hardware doesn't work.
And you can't run linux if your ex-girlfriend ran off with your computer. Should we then widen the scope to include relationship problems? Where does it end? -- _____________________________________ John Andersen
Carlos E. R. wrote:
We also have 3 phase with 120V to neutral and 208 between phases. However, to get the 120V the 240V load side of the transformer has a center tap, that is grounded. Either side to neutral is 120V. This has the advantage where current in one side balances current in the other, reducing the current in the neutral, thus reducing resistance losses in the wiring.
On Tue, 2006-10-24 at 07:37 -0400, James Knott wrote:
Hmm. Ok, now I know. Thanks for the info! -- /Peo -- Registered Linux User #432116, get counted at http://counter.li.org -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
Peo Nilsson wrote:
[clipped more] Torroid Corp. has a (somewhat) simplified chart of voltages: http://www.toroid.com/custom_transformers/technical_bulletin_2.htm http://www.powerstream.com/cv.htm lists Country Voltages and Plug Styles. http://users.pandora.be/worldstandards/electricity.htm is a more graphical/map presentation. Wikipedia history is good reading (if more than wanted). http://en.wikipedia.org/wiki/Electricity_distribution North American centric with worldwide comparisons. Single and three phase stuff even. It is difficult to find out about the many years of generating systems inching upward from 105V, through 110V, past 115V and now firmly at 120-volts. Seldom does anyone say plainly to the consumer that a 120-volt standard allows for losses inside a building, so that 115-volt equipment is meant to be used on 120-volt systems. Mostly 110 is a memory artifact from bygone years.
On Tuesday 24 October 2006 18:48, Stanley Long wrote: .....
Mostly 110 is a memory artifact from bygone years.
Yes. It is 120 V nominal, but I still call it 110, or 220 for Europe. Old guys can't always forget fast enough :-) -- Regards, Rajko M.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Monday 2006-10-23 at 20:52 -0500, Rajko M wrote:
I know. But in my case the operator is probably at ground potential: humidity of 80%, artificial stone floor.
For that reason, good antistatic straps have a 1..5 meg resistor in series ;-)
True. But good poower supplies also have a "real" switch at the rear. That's the one I meant. You can also switch it off at the main connection bar.
It's easy to forget to "touch metal" periodically and often.
Just walking a few meters on a carpeted room does it.
Did I mentioned, that disconnected cord helps a lot against bad luck. I'm not that superstitious, but by now it helped better than a rabbit foot.
X-) - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFPe0WtTMYHG2NR9URAt1UAJ9LJrKW6Bt9GCFaWZcdhGv1yDg54ACfVd2j A+kVCpScQ8kfztELn2c3hjs= =wr/j -----END PGP SIGNATURE-----
On 06/10/23 15:58 (GMT-0500) Greg Wallace apparently typed:
If you really think about it, it shouldn't be hard to believe. The main components inside a PC are covered with tiny components crowded closely together. It shouldn't take much imagination to see a spec of litter landing in exactly some wrong place so as to bleed off a signal from one component to some other. Unless you have the PC in a clean room, or its air inputs filtered, or are using entirely passive cooling, it's constantly pumping dust through. Given enough time, something will eventually land in a bad place. -- "The Lord is my strength and my shield; my heart trusts in him, and I am helped." Psalm 28:7 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://mrmazda.no-ip.com/
On Monday 23 October 2006 15:58, Greg Wallace wrote:
Greg, when I worked for AT&T in their switch centers, where the switch was an entire floor of 11 ft equipment bays full of plugin boards, we had all manner of weird problems. Seems like there is a fungus, or was that a bacterium, that likes to live on the gold plating of the connectors and its byproducts formed an atoms-thick insulation on the contacts. Bell Labs developed a spray cleaner for the connectors. Just unplug the board, spray the connector, plug it back in and the ghost would disappear. I seem to remember that the spray was found to be carcinogenic so they came up with a different formula, something with Freon TF, probably not ozone friendly, so maybe they are using something different now. Point is, it ain't dust or corrosion getting into the slot on your m.b. I have used 91% alcohol and, in extreme cases, a soft pencil eraser to clean contacts. All with antistatic precautions, obviously. Simply moving the board in the slot should clean enough surface area to fix the connection, but the problem will return. Then there is the problem of crystalline metallic whisker growth that bridges (shorts out) connections, but that's another story. Fred
On Monday 23 October 2006 15:58, Greg Wallace wrote:
Fixed a neighbor's computer that lost video by just loosening the slot screw holding the video card down and pushing the card down in the slot. Had to remove the card and adjust its slot cover for a longer term fix. What happens over time when you power a system on and off is that metal connectors and connections expand and contract with the heating and cooling cycles. Eventually they will move. How much and whether they lose connection you don't know. All warnings about static electricity and grounding apply and only attempt this with power the cord removed: Fully removing and reseating cards, cable connections, etc can usually resolve a whole lot of problems. Stan
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Monday 2006-10-23 at 17:02 -0500, Stan Glasoe wrote:
Other people recommend leaving the cord connected, with the switch in the off position: this way the earth wire remains connected. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFPUa3tTMYHG2NR9URAkKjAJ0YBfhudyEvk4YP0x8Xjf7Y9W6ozgCffMXR ueYy5u0SYb/eG7EaKzIXguw= =N8D9 -----END PGP SIGNATURE-----
On Monday 23 October 2006 17:48, Carlos E. R. wrote:
_IF_ there is a power switch _on_ the power supply near the power cord! The main on/off switch on the front of the system doesn't remove all power from the system and not all power supplies have their own on/off switch. With static it is important to have all parts at the same potential not that the static is eliminated. You, the machine and whatever part you are adding/removing/reseating should be at the same static 'level'. That way no discharge occurs. Stan
On Monday 23 October 2006 17:48, Carlos E. R. wrote:
The magic that burns electrostatic sensitive parts is amount of electrostatic charge that will go trough the part driven by potential (voltage) difference between human body and the part. Not every static is damaging, but in the winter our skin is dry enough to load quite a bit. Attached ground to the chassis will not help if operator forget to use antistatic strap attached to the metal part of the case, but then the for the safety reasons cord must be unplugged from the wall outlet, because you are accessing malfunctioning device where you have no idea what is wrong. That means you have to find a ground that is for sure ground, and not "it should be ground", because "it could be line voltage" as well. Second reason not to use this method is construction of todays computers, where "switch" is just a signal key that tells motherboard to turn power supply on. So there is no real separation of electric power (110 VAC) like it was some 10 years ago. Thanks to Stan to remind us on this. Groundign everything and using antistatic straps is valid in industry where worker/operator has to touch many objects and in the same time has to focus on job, not on ESD (ElectroStatic Discharge) problems. On a single device it will work without straps, but with unplugged power cord, as long as repairer has in mind that discharging is not one time process, but it has to be done after each contact with something outside the computer case. Did I mentioned, that disconnected cord helps a lot against bad luck. I'm not that superstitious, but by now it helped better than a rabbit foot. -- Regards, Rajko M.
On Mon, 2006-10-23 at 20:52 -0500, Rajko M wrote:
230 V~ 50Hz Where do they have 110 V~ ???
-- /Peo -- Registered Linux User #432116, get counted at http://counter.li.org -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
On Tue, 2006-10-24 at 00:28 -0800, John Andersen wrote:
Mostly to the local pub. :-)
-- /Cheers Peo -- Registered Linux User #432116, get counted at http://counter.li.org -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
Peo Nilsson wrote:
Where do they have 110 V~ ???
North America. Here we have a 240/120V distribution system. You'd use 120 for small appliances, lighting etc. and 240 for heavy appliances such as stoves etc. The power feed to homes is 240V with a grounded center tap or neutral, so there's 120V either side of neutral.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Tuesday 2006-10-24 at 07:37 -0400, James Knott wrote:
Dual phase? Curious. Here (Spain) we have a three phase system. Voltage is 230 (was 220) live to neutral, with 230*sqroot(3) = 400 Volts from phase to phase. We have 230 for homes, and 400 (3 phases) for industries and heavy loads like lifts or air condiotining. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFFPhRftTMYHG2NR9URAlPJAJ0WWM13qKME5c8Qb0bvFqyMglFDUgCfYWdl vPbMN+XcQVymzUh6ZK3XYDA= =CLDf -----END PGP SIGNATURE-----
Carlos E. R. wrote:
Hope this ends the interesting OT discussion :) <http://en.wikipedia.org/wiki/Domestic_AC_power_plugs_and_sockets#World_maps_by_plug.2Fsocket_and_voltage.2Ffrequency> Cheers, Dave
participants (13)
-
Basil Chupin -
Carlos E. R. -
Dave Howorth -
Felix Miata -
Greg Wallace -
James Knott -
John Andersen -
Peo Nilsson -
Rajko M -
Randall R Schulz -
Stan Glasoe -
Stanley Long -
Stevens