Solved Can a bad voltage stabilizer crash computer?

[yankleber]

OK, this is what is happening. I have this client with a 6 years old Dell desktop (Pentium dual core with 4GB RAM under an old no SP W7). The computer started to freeze and give BSOD. Sometimes it was BSOD other times it just froze completely including the mouse pointer with no BSOD.

I brought it to my workbench and made a fresh install of Windows 7 SP1. I didn't use the more updated Dell drivers and just relied on the Windows embedded drivers since there were no warnings in the hardware list. I usually do that for mostly of computers (including mine) and never had problems before. Beside that I installed ZoneAlarm, Avast, Java, Office 2010, configured Outlook, and sent computer back to client as everything seemed to be fine after to run the computer for hours. The only thing that I left for client do himself was to install the proprietary navigator of his bank (don't ask!).

The day after the client called saying that computer was freezing and sometimes when it was idle. I picked computer again, put it over my workbench and ran either MemTest86 as well checkdisk/r for a total of 16 hours and didn't find any bad sector or RAM error.

I was a bit suspicious of Office 2010 because it was using KMS activator and ZoneAlarm wasn't liking it. So I uninstalled it and downgraded to Office 2007, deactivated several useless services (software updaters, etc). After that I ran the computer for THREE DAYS and it didn't freeze one single time.

So I sent computer back again to the client and again the day after he called saying that it was freezing. Yesterday I went to client home and right at the doorstep he said to me that it froze a bit before I arrive but along one hour I stayed there the computer didn't freeze.

I will pick his PC again today and I am about to make a whole new fresh install. This time I will put it on a SSD of mine that I will lend to show him how faster the computer can become and at the same time completely discard the possibility of a bad HDD. Also this time I will connect to Dell and download the most updated drivers and also tweak the sleep/wake BIOS features. I know that I should think of update the BIOS itself but I am not too much happy with the idea.

What still bugs me more is that the computer works finely on my workbench but not in client's home so I am not too much hopeful that the actions above will sanitize the problem. Yesterday I noticed that he uses a cheapo voltage stabilizer. Personally I don't rely on them. I use UPS for my own computer but in my workbench I have straight outlets (I actually feel better plugging a computer directly into an outlet than on a cheap voltage stabilizer).

Question is: there is a chance of a bad stabilizer be generating fast peaks or dropouts or whatever able to freeze the computer? Or this is a crazy idea of my imagination?

Thanks!

 

[RolandJS]

"...Question is: there is a chance of a bad stabilizer be generating fast peaks or dropouts or whatever able to freeze the computer?" From what I picked up around The 'Net, absolutely. Is there a way you can ask him to conduct a test by using a different much more newer UPS and have that UPS plugged directly into the nearest wall outlet? And,exactly what is his "power train" from the wall outlet to the computer? If there are too many things plugged into one wall outlet, problems could happen. BTW, I use a business oriented UPS on my computer center, it was not cheap. I just remembered, a computer power supply, if the wattage rating is not equal to or more than the wattage the computer needs, can "under power" the computer.

 

[yankleber]

That's what I imagined — but didn't want to beat the hammer without ask others first. Thanks for helping!

:huh:

I hear ya! I have myself two APC RS-1500. One I use in my home-office for my own computer and other stays at the main hall holding the cable TV receptor, and Internet stuff (modem, router and radio antenna). Here in Brasil a similar model (the RS-1500 is not made anymore) cost around $200 each. It gives me shivers every time I see those cheap lightweight stabilizers. This one my client uses shouldn't cost more than $15. Pure crap.



Anyway, I will conduct the first part of the experiment myself by bringing his stabilizer together with the computer today and see if I can increase the 'freezing rate' of my workbench.



EDIT: This client lives in a small building. I was thinking that perhaps someone in other apartment may have a refrigerator plugged into an outlet in the same power line and when it starts the engine it causes a surge that could freeze his computer. Here in home sometimes when my wife opens the kitchen or bathroom tap (and automatically turns the electric water heater on) my UPS reacts.

 

[Megahertz07]

PSU will work well from -15% to + 15% nominal voltage. New ones works automatically from 100 to 240 Vac. Unless he lives on a rural area, don't use any voltage stabilizer. Use a only a voltage arrest.

There was a Win update that screwed Windows Updates.
Try to set Windows update to disabled.
Reboot and see if it works normally.

 

[yankleber]

PSU will work well from -15% to + 15% nominal voltage. New ones works automatically from 100 to 240 Vac. Unless he lives on a rural area, don't use any voltage stabilizer. Use a only a voltage arrest.

Go try to convince a brazilian computer user about this!

There was a Win update that screwed Windows Updates.
Try to set Windows update to disabled. Reboot and see if it works normally.

Windows update is disabled. I set it by default in all my installs...

 

[Megahertz07]

Those automatic voltage stabilizer uses a multi tap transformer. If the voltage is low, it selects a higher tap. If the voltage is high it selects a lower tap.
The problem is that the switching is done by a relay. If the voltage is low and uses a high tap, when it changes to normal, it takes some time to the tap to switch back to a lower tap, resulting in a high voltage on the output.
By the way, I'm Brazilian, live in S. Paulo and as an electronic engineer, don't recommend them.

A good UPS, like those from APC (BZ1200-BR) is the right choice. No-Breaks APC | Solo Network | Compre com os Especialistas...

 

[yankleber]

Those automatic voltage stabilizer uses a multi tap transformer. If the voltage is low, it selects a higher tap. If the voltage is high it selects a lower tap.
The problem is that the switching is done by a relay. If the voltage is low and uses a high tap, when it changes to normal, it takes some time to the tap to switch back to a lower tap, resulting in a high voltage on the output.
By the way, I'm Brazilian, live in S. Paulo and as an electronic engineer, don't recommend them.

Yeah, I hate them. There is over 20 years that I don't use it myself. But being a brazilian like me you know how this belief is deepness stuck in our culture. Yesterday I stayed with my client and mentioned that the problem could be the stabilizer and that he should get rid of it ASAP. You should have seen the way he looked at me...

A good UPS, like those from APC (BZ1200-BR) is the right choice.

I love this make. I have two RS1500 that I purchased over 10 years ago and they are still like new. I was thinking about to recommend my client the APC BZ600 that is half the capacity of the one you recommended because he has only a computer with monitor so probably it will be enough. Besides I don't think that he will want to spend over $150 on an UPS. First things first is to convince him about the issue.

 

[RolandJS]

I find it best to not call anything of the clients' "crap" "garbage" etc. because all that does is give 'em a door to "I'm offended". IMO, it's better to state the advantages, features, benefits, etc. of what clients should be using, give 'em a chance to agree - eventually and do what was suggested.

 

[yankleber]

I find it best to not call anything of the clients' "crap" "garbage" etc. because all that does is give 'em a door to "I'm offended". IMO, it's better to state the advantages, features, benefits, etc. of what clients should be using, give 'em a chance to agree - eventually and do what was suggested.

Fortunately I didn't do that! I just am feeling that he has some resistance to believe that the stabilizer may be the cause of the problem. He looked at me as if I was some kind of lunatic inventing the thing. The very first argument was "I have using it since the computer was new and I never had a problem before". Man how I hate such situations!

 

[RolandJS]

"I have using it since the computer was new and I never had a problem before." and its cousin "We have always done this/that way before." -- has been the downfall of many cooperate users and home users -- I have no idea on how to convert such thinking. Yankleber, you're a trooper for trying to help this client

 

[yankleber]

OK, seems that the problem is solved (so far). The client agreed to bypass the stabilizer and this morning when I stayed at his house to help him to install a couple security modules for Internet banking he confirmed me that the computer never froze again since 4 days ago when I delivered it.

Everything signs that the problem was really being caused by a strained PS semi-damaged by the cursed voltage relay switcher. Now I am going to prescript him a good surge board or a no-break.