PC wakes up whenever I turn on my fan.

[Count von Schwalbe]

Ummm, I don't know who suggested that. I sure didn't, or wouldn't. I only suggested the fan not be plugged into the UPS.

Did you note that the mains voltage at OP's location is variable and has been ~170v for the last couple of weeks?

@caroline! I would suggest looking into a 120v fan and a limiter circuit, it is easier to reduce voltage than to increase it. You can always move it to a single leg of 240v to get the 120v once the voltage is back up.

Or you could look into a Variac to manually stabilize voltage if long-term brownouts are common. Bear in mind you will need a form of over-voltage protection if the power rises unexpectedly and you are using a step-up transformer.

 

[1freedude]

Did you note that the mains voltage at OP's location is variable and has been ~170v for the last couple of weeks?

@caroline! I would suggest looking into a 120v fan and a limiter circuit, it is easier to reduce voltage than to increase it. You can always move it to a single leg of 240v to get the 120v once the voltage is back up.

Or you could look into a Variac to manually stabilize voltage if long-term brownouts are common. Bear in mind you will need a form of over-voltage protection if the power rises unexpectedly and you are using a step-up transformer.

Also, to you, Count, 240 does not work the same way in other parts of the world. We have -120 - 0 - +120 to create 240vac. The rest of world uses 0 and +240 to create 240vac, and typically 50Hz. So, one leg wont work.

And Bill, i suggested using a resistive load as an experiment only to see if it will counteract the inductive load waking the computer, not to test the UPS.

 

[Bill_Bright]

Did you note that the mains voltage at OP's location is variable and has been ~170v for the last couple of weeks?

Nope - because the OP never said that. The OP did say 180, but doesn't change my advice.

Do not run a 1/2 horsepower fan through the same UPS as the computer equipment.

And Bill, i suggested using a resistive load as an experiment only to see if it will counteract the inductive load waking the computer, not to test the UPS.

I have no problem with putting a load on the system - just not a space heater - and not as the same time with the sensitive computer electronics.

 

[Count von Schwalbe]

Also, to you, Count, 240 does not work the same way in other parts of the world. We have -120 - 0 - +120 to create 240vac. The rest of world uses 0 and +240 to create 240vac, and typically 50Hz. So, one leg wont work.

Yep, forgot about that.

Now my brain is thinking of a way to have a self-regulating variable transformer or rheostat to ensure consistent voltage.

 

[1freedude]

Yep, forgot about that.

Now my brain is thinking of a way to have a self-regulating variable transformer or rheostat to ensure consistent voltage.

But, doesn't 170 look suspiciously like half of 338 ass-u-ming 240vac rms? I wish I knew more about foreign eletric supplies

Edit or 240vac from 170vac rms

 

[Bill_Bright]

Also why the UPS gets so hot, AVR is constantly working to step up my piss mains to something that's actually normal for the components.

Because the AVR is working hard to boost the voltage back up to "normal" potentials. It makes sense it would get hot.

I also note that 180VAC is the bottom threshold, according to the ATX Form Factor standard, for 240VAC mains, that an ATX compliant PSU is "supposed" to tolerate and compensate for.

If your mains are dropping that low, and sitting there for hours or days on end, you are using the wrong tools for the job. Compensating for those extreme circumstances is not the job for an UPS. You need a power conditioner with AVR.

 

[caroline!]

I would not recommend using a space heater! Small ones pull 1000W and none provide any regulation or transient suppression. If I were to use a space heater, I sure would not have any of my sensitive computer or network electronics connected to the UPS at the same time.

Don't worry the small heater is packed in a box, I don't think I'll get to use it this year considering it's 43C rather than 15-20 like it should be lol. And the big one has a plug that doesn't fits the UPS, I'd never plug one into the UPS tho, it's a good way to fry something.

I say, if you do NOT run the fan through the UPS, and the UPS maintains power to your equipment during a power outage, it passed the test.

With the computer at full blast it gives me around 10 minutes to save any work and shut it down.

If you NEED for this computer to be running 24/7 and you have frequent and long duration outages, you need to look into a backup generator. Then the UPS should be used to provide power between the grid outage and until the generator takes over.

Not really for 24/7 usage the UPS is great mainly because of the AVR like you said, and if power goes out I quickly save anything I'm doing and shutdown, not going to browse youtube videos on battery power ^^
We do have a generator already but only for the fridge and lights, PC can run on it but it's not a critical piece of equipment, I'd rather have cold water than AI algos or games. Imagine doing F@H on backup power lol, that'd be serious commitment.

Ok, outside not an option.

100 + should do it.

There was a change. With a 150W bulb the PC sometimes wouldn't wake up, it was seemingly random as I repeated the test several times on a row.

Also, to you, Count, 240 does not work the same way in other parts of the world. We have -120 - 0 - +120 to create 240vac. The rest of world uses 0 and +240 to create 240vac, and typically 50Hz. So, one leg wont work.

And Bill, i suggested using a resistive load as an experiment only to see if it will counteract the inductive load waking the computer, not to test the UPS.

Probe any two live wires and you get ~415V, in theory. Though it's only used for large buildings and industries.
The easy math is V*1.73, so if you do 240*1.73=415.2, you can do the same to get 208V but the american system is obscure knowledge to me.

Because the AVR is working hard to boost the voltage back up to "normal" potentials. It makes sense it would get hot.

I also note that 180VAC is the bottom threshold, according to the ATX Form Factor standard, for 240VAC mains, that an ATX compliant PSU is "supposed" to tolerate and compensate for.

If your mains are dropping that low, and sitting there for hours or days on end, you are using the wrong tools for the job. Compensating for those extreme circumstances is not the job for an UPS. You need a power conditioner with AVR.

The large ones cost both of my kidneys so I'd have better luck hunting for another AVR, they're pretty common because old TVs were too bitchy about input voltage, so if you didn't had exactly 220/240 (depending on the manufacturer) the brightness would drop or it'd refuse to start, same for vintage radios you can only hear static or the AM whistling if their input voltage isn't right.
As far as I know there's been problems with the power since the 1950's, it's fun to look up old newspapers and read that another blackout has left the city in darkness, people complains on the streets and it's literally if it was today's news heh. Old habits die hard.

 

[caroline!]

Update: So basically what I did was swap two wires inside the fan and it stopped happening. For now, been using it for the whole day and it was fine.

Someone must've wired incorrectly at the factory so that the speed knob was actually wired to the neutral wire rather than the live one, morons. Never cease to amaze me.

 

[Bill_Bright]

Wow! What made you check that?

I doubt that is a one-off problem. In other words, if that was a factory mistake, it seems it would have happened to that entire lot/production run of fan, meaning there could be 1000s of incorrectly wired fans out there. Not good. I would write the company.

I would also check the company's website to see if there are any product recalls. Note, being notified of any recalls is the only good reason (IMO) to register new products for warranty purposes.

That said, because this "could" result in a significant safety hazard, and because it "should" have triggered a product recall, I worry that your "fix" just created another safety hazard by simply moving the problem down the line, sorta speak.

I asked multiple times above about the wall outlets being properly wired and grounded, recommending you ensure the wall outlets are indeed, properly grounded. You replied,

You got this one right.

... suggesting to me, the outlet is NOT properly wired or grounded. By swapping the wires in the fan now, you might have introduced another hazard where, if (when?) that fan is plugged into a properly wired outlet, it's wiring (now truly reversed) is dangerous again. That would not be good.

I strongly urge you to check your wall outlets. Every home and every computer user should have access to a AC Outlet Tester to ensure the wall outlet is properly wired and grounded to Earth ground. I recommend one with a GFCI (ground fault circuit interrupt) indicator as it can be used to test bathroom and kitchen outlets (outlets near water) too. These testers can be found for your type and voltage outlet, foreign or domestic, (like this one for the UK, or this one for German outlets) at most home improvement stores, or even the electrical department at Wal-Mart. Use it to test all the outlets in the home and if a fault is shown, have it fixed by a qualified electrician.

Then, if it turns out your outlet was incorrectly wired, be sure to swap back the wires in the fan BEFORE plugging it in again.

 

[1freedude]

Running any cooler?

 

[caroline!]

Wow! What made you check that?

I went to check the capacitor.

That said, because this "could" result in a significant safety hazard, and because it "should" have triggered a product recall, I worry that your "fix" just created another safety hazard by simply moving the problem down the line, sorta speak.

I doubt, I mean there's only two possibilities when it comes to the speed knob and either makes the fan work, only switching the neutral can cause more issues, I've seen it before with flickering lights or neon indicators that dim but never turn off, in fans it causes this problem I assume.

I asked multiple times above about the wall outlets being properly wired and grounded, recommending you ensure the wall outlets are indeed, properly grounded. You replied,
... suggesting to me, the outlet is NOT properly wired or grounded. By swapping the wires in the fan now, you might have introduced another hazard where, if (when?) that fan is plugged into a properly wired outlet, it's wiring (now truly reversed) is dangerous again. That would not be good.

Our electrical system is hard to explain. To begin with, there's no ground as you know it in other countries, the only similar thing I've seen is what the Japanese use, ground or earthing exists but only for specific appliances like fridges, electric ovens, washing machines, "box" air conditioners, things with metal housings you can touch and might shock you. Here's a Japanese outlet, I think something similar was featured in a Linus Tech Tips videos where they test a Japanese PSU but I don't remember the name now (also the channel got nuked lol)

Our system works the same way but the wire requires -in theory- a round terminal, though you can coil it around the nut and it'll work the same. Here's one I'm not using and the plug that fits it.

These sockets aren't everywhere so the ones in my room are regular as there's no need for anything to be earthed, you'd assume the fan might need ground as it's metallic but I mean, I'm not the ones making the rules lol, perhaps if it was a fixed wall mounted fan but then nobody would touch it anyway... and the thing with the fan is that's imported from China, and comes with the polarised flat blades plug, so you have to use an adapter, or in my case I've got one of those "universal" power strips that can take a bunch of different plugs including the one for the fan. The "long" blade in the US plug should be N, and it is, it was the fan that was wired incorrectly inside. I could've just replaced the cord for the standard one but then I wouldn't know which pin connects to the neutral wire.

I strongly urge you to check your wall outlets. Every home and every computer user should have access to a AC Outlet Tester to ensure the wall outlet is properly wired and grounded to Earth ground. I recommend one with a GFCI (ground fault circuit interrupt) indicator as it can be used to test bathroom and kitchen outlets (outlets near water) too. These testers can be found for your type and voltage outlet, foreign or domestic, (like this one for the UK, or this one for German outlets) at most home improvement stores, or even the electrical department at Wal-Mart. Use it to test all the outlets in the home and if a fault is shown, have it fixed by a qualified electrician.

I've made my own tester with the LEDs and all but as you can see it's sort of useless when it comes to ground, it only indicates me which of the contacts is live, works for that.

Running any cooler?

No but at least the computer doesn't wakes up anymore.

 

[LabRat 891]

EMI from fan / poor EMC on cable or PC chassis.
-Try an AM radio or not-connected speakers w/ the volume up; you'll hear the 'EM pop' in the radio and speakers, from the fan's coils being energized/deenergized.

If your case is open, put the panel back on.
I was crashing my system about every 3rd-4th 'sitting down', just sitting in my office chair (this winter). Putting the (non-windowed) side panel back on fixed the problem.

Alternatively, I've also seen similar happen due to extremely poor quality HDMI cables 'inducing' a fault from just walking by the machine on a carpeted floor. Any long and exceptionally sensitive (and poorly shielded) cable could be a potential source.

 

[caroline!]

EMI from fan / poor EMC on cable or PC chassis.
-Try an AM radio or not-connected speakers w/ the volume up; you'll hear the 'EM pop' in the radio and speakers, from the fan's coils being energized/deenergized.

If your case is open, put the panel back on.
I was crashing my system about every 3rd-4th 'sitting down', just sitting in my office chair (this winter). Putting the (non-windowed) side panel back on fixed the problem.

Alternatively, I've also seen similar happen due to extremely poor quality HDMI cables 'inducing' a fault from just walking by the machine on a carpeted floor. Any long and exceptionally sensitive (and poorly shielded) cable could be a potential source.

My current speakers don't have any pop noise but I tried with a cheap mono speaker and it does it.
Case is closed for airflow reasons, the cards get hot but the fans are installed in such a way the heat hits the side panel and moves up towards the top fans, avoiding the CPU cooler almost completely, if I remove the panel temps rise.

I dunno about video cables, I'm using a HDMI to VGA converter.

 

[ShiBDiB]

These sockets aren't everywhere so the ones in my room are regular as there's no need for anything to be earthed

The UPS needs to be grounded or you might as well throw it in the garbage.

 

[Bill_Bright]

The UPS needs to be grounded or you might as well throw it in the garbage.

No it doesn't need to be grounded but the excessive surge and spike protection feature works much better when excessive anomalies can be shunted to Earth ground.

And just because an outlet only has two wires does not mean it is not grounded. But a 3rd wire to ground is considered safer.

Edit comment: fixed typo.

 

[caroline!]

The UPS needs to be grounded or you might as well throw it in the garbage.

No it doesn't need to be grounded but the excessive surge and spike protection feature works much better when excessive anomalies can be shunted to Earth ground.

And just because an outlet only has two wires does not mean it is not grounded. But a 3rd wire to ground is considered safer.

Edit comment: fixed typo.

I know it's better but rn I don't have 30+ meters of wire to run from the kitchen to my room. And installing the earthed wall socket would leave me without a socket for the night light ^^

And indeed the N wire technically goes to ground already.

 

[LabRat 891]

HDMI to VGA converter.

Whatever situation necessitates this, I empathize.
In the past, I've had no choice but to use the VGA input on a budget HDTV. I didn't even realize how awful of an experience it was until I eventually got a decent LCD Monitor.

Going digital to analog almost universally looks awful (excepting profe$$ional adapter$) and usually incurs input latency.
It's not impossible a cheap and poorly shielded VGA cable or extremely poorly designed adapter is the cause. However, unless there's an easy way to swap/test this, I'd leave it towards the bottom of the list on troubleshooting.

Even if the VGA thing is somehow related, the direction I've seen generally in the thread on 'improving electrical earthing' might resolve the issue.
Tip: You can take any IEC 'kettle cord', break off the Hots/Neutrals, and use it as a 'pluggable earthing cable'. I strip off the other end and use the Green ground wire as a earthing lead for ESD straps and PDU grounds.


ACHTUNG! Never. EVER. trust the neutral as an Earth/Ground. At least in North American use, there is always a non-0 chance of a fault allowing potentially lethal current to become momentarily or continuously present on the Neutral.
There are good reasons that most regulators demand Neutrals only be bonded to earth at a single point. (Even if it complicates ElectroMagneticCompatibility in the 21st century)
While less of a safety issue, you will always see harmonics, inducted voltage, and RFI on Neutrals. They're long antennas with a single shared earth. Plus, they're strung alongside loaded Hot conductors.


While it might be unsightly, you could 'plug into a ground' somewhere else and (re)use any (copper) wire as an extension to where you need it. This isn't an NEC-approved or max-EMC solution (at all) but, I would (personally) be trying manually earthing things in my room until I caused a (diagnostically relevant) change / stopped the issue with the fan. After that, then I'd be trying to figure out a practical solution.

edit: had an idea, but dunno if my concept of the physics of it are even accurate.
What about a ferrite choke/bead on the fan's power cord? Might be able to salvage one from an old cable.
I'd even try different configurations/coilings of the fan's cable, to see if it changes the symptoms. The point here, is that the power cable for the fan is an antenna as much as it's a power cable; making it a worse antenna could help. Heck, got any metal gas/plumbing pipes handy? Try running the fan's cable thru or around the pipe. lol.

The idea with the 'oddball' diagnostics is more to make a change to the symptoms, than to directly solve it.
Ex. I thought grounding a metal desk that was ESDing a lot would 'help'. No, it made the ESD worse, but the change pointed me towards the actual cause: An air cleaner w/ a HV element.

Please note: I am neither an electrician or electrical engineer, nor HAM / radio tech. I'm just an enthusiastically curious techie that has ran into more than his fair share of 'spooky electrical problems'.

 

[caroline!]

Whatever situation necessitates this, I empathize.
In the past, I've had no choice but to use the VGA input on a budget HDTV. I didn't even realize how awful of an experience it was until I eventually got a decent LCD Monitor.

Going digital to analog almost universally looks awful (excepting profe$$ional adapter$) and usually incurs input latency.
It's not impossible a cheap and poorly shielded VGA cable or extremely poorly designed adapter is the cause. However, unless there's an easy way to swap/test this, I'd leave it towards the bottom of the list on troubleshooting.

Even if the VGA thing is somehow related, the direction I've seen generally in the thread on 'improving electrical earthing' might resolve the issue.
Tip: You can take any IEC 'kettle cord', break off the Hots/Neutrals, and use it as a 'pluggable earthing cable'. I strip off the other end and use the Green ground wire as a earthing lead for ESD straps and PDU grounds.

I'm using a somewhat good converter (not profe$$ional tho) with cables that *look* fine on the outside, HDMI cable I got it from a STB kit because my TV doesn't uses HDMI anyway, and the USB looks fine, not ultra cheap but midrange I suppose, it's an USB after all. VGA is from an old Compaq screen and the only one I could fine that actually makes use of pin #9 (EDID), most new ones skip it, cable is thick so it has proper shielding for sure.

I mean, the fan works fine now and the PC has stopped waking up so swapping the motor wires did something for sure.

As for grounding the UPS I'd have to run a ridiculously long wire and if I do that I'd rather run it through the conduits like usual. I do have wire available because I installed a solar panel and a few lights but it's too thick, it'd be a waste to use it for that purpose if it even fits the conduits, they're 1" wide but there are places where multiple wires run together, usually for lights and random derivations that must go through there.

Heck, got any metal gas/plumbing pipes handy? Try running the fan's cable thru or around the pipe. lol.

Steam radiator but on the opposite side of the room lol

 

[Bill_Bright]

Heck, got any metal gas/plumbing pipes handy?

"Gas"!

Sorry dude, but that is some of the most dangerous, reckless advice ever given!

Sure, bond something that has the potential to produce sparks to a gas pipe that has the potential to leak and blow up the entire neighborhood.

As far as plumbing pipes, in most cases these days, even that will not do. First, it never was just any "plumbing" pipes. Drainage pipes are no good for several reasons. Many (such as the main stack) are made of cast iron (a relatively poor conductor as metals go). Some older drainage pipes are even made of tar infused paper products. So you might have a metal pipe in the house that meets up to a cast iron or paper fiber (non-conductive, ungrounded) pipe as it leaves the house. Plus, since drainage pipes are not carrying pressurized water, connections are not soldered, electrical continuity is iffy, at best.

Hot water pipes should not be used because they don't go into the ground - thus are not properly grounded.

The main, cold water service pipes in the old days were commonly used because they did come out of the ground. And connections throughout the facility were soldered from end to end for good electrical continuity. But for decades now, PVC pipes have been used for new builds AND replacement plumbing in remodeling projects. This means remodeled bathrooms, laundry rooms and kitchens were commonly remodeled using PVC pipes. So even in older homes where copper cold water pipes were used, unless tested, there is no assurance those pipes will provide a good path to Earth ground from end to end.

So no. The best solution is to run a solid copper ground wire directly to Earth ground either through the service panel, or directly out to a proper ground stake.

Never, as in NEVER EVER to a gas pipe.

BTW, if you were to use a gas pipe, first, it would never pass permit inspection. And if you failed to get the required permits and blew up the neighborhood, chances are you would live the rest of your life behind bars (assuming you were not home at the time of the explosion).

 

[caroline!]

"Gas"!

Sorry dude, but that is some of the most dangerous, reckless advice ever given!

(I only have this meme saved in ukrainian but everyone gets it)

As far as plumbing pipes, in most cases these days, even that will not do. First, it never was just any "plumbing" pipes. Drainage pipes are no good for several reasons. Many (such as the main stack) are made of cast iron (a relatively poor conductor as metals go). Some older drainage pipes are even made of tar infused paper products. So you might have a metal pipe in the house that meets up to a cast iron or paper fiber (non-conductive, ungrounded) pipe as it leaves the house. Plus, since drainage pipes are not carrying pressurized water, connections are not soldered, electrical continuity is iffy, at best.

I've heard of the plumbing pipe grounding before, it's mainly suggested by old school electricians, I remember our known electrician who always came to fix something suggested grounding the electric fridge (there's the other that runs on kerosene) by running a wire to the pipes under the sink, we never did it tho and it was only until last year I installed the grounded wall socket and wire. Hot water and steam pipes are copper, and the rest must be iron or lead given their appearance, I think they're covered in tar on the outside though, that and over a century of grime made them black.

Never, as in NEVER EVER to a gas pipe.

BTW, if you were to use a gas pipe, first, it would never pass permit inspection. And if you failed to get the required permits and blew up the neighborhood, chances are you would live the rest of your life behind bars (assuming you were not home at the time of the explosion).

Gas pipes aren't really visible, they're underground, only the feed for the boiler is visible and painted in bright red so you know it's gas, and the stove is connected via a flexible metal hose covered in a high-temp resistant insulation which I believe is non-conductive. Not like I'd run a wire behind an oven, I'm dumb but not *that* dumb.

 

[Bill_Bright]

Gas pipes aren't really visible, they're underground, only the feed for the boiler is visible and painted in bright red so you know it's gas, and the stove is connected via a flexible metal hose covered in a high-temp resistant insulation which I believe is non-conductive. Not like I'd run a wire behind an oven, I'm dumb but not *that* dumb.

Well, you don't tie a ground to something that is painted since the paint itself will offer some resistance - never good in a ground circuit.

My gas line enters the side of the house then, via "black iron" pipes run about 25 feet through the main level floor/basement ceiling joists to the furnace room. They then are exposed when they drop down to the furnace and water heater.

BTW, in the US, copper pipes are not allow for natural gas. Iron pipes are typically used. Natural gas will corrode copper and eventually cause a leak. Not good.

 

[caroline!]

Well, you don't tie a ground to something that is painted since the paint itself will offer some resistance - never good in a ground circuit.

My gas line enters the side of the house then, via "black iron" pipes run about 25 feet through the main level floor/basement ceiling joists to the furnace room. They then are exposed when they drop down to the furnace and water heater.

BTW, in the US, copper pipes are not allow for natural gas. Iron pipes are typically used. Natural gas will corrode copper and eventually cause a leak. Not good.

Don't think gas pipes are copper though, I think they're iron, and the flexible hoses are braided corrugated steel with an insulation on top, and what looks like a red heatshrink on top as a visual indicator, they're flexible but not to the point you can make a 90° bends.

New gas pipes for newly built houses are hard plastic (PVC?) so that'd rule out any possible conductivity.
When it comes to water plastic pipes have an advantage and it's thermal insulation, metal pipes act like a heatsink and can condensate during cold days = mold in your walls, and the water doesn't tastes like metal I suppose.
But plastic can't be used for steam though, I think it'd burst the joints or valves. When I tried out the radiators all of the pipes were rattling and the valves whistling until pressure was equal for all of them. I was told by the plumber to test them at least once a year to make sure all of the supply and exhaust valves work. It's an old system but still keeps up, I've cleaned all of the rads with a power washer once.

 

[LabRat 891]

Calm down there DW.
First and foremost, if you weren't getting the implication that the pipe should be "spare, unused, NOT installed", then you probably regularly see some pretty crazy and stupid things (to even go that route in thinking.)

Then again:
Perhaps, it's just an Americanism to have materials on hand from home improvement, DIY projects, (and salvage). I know I have a 5/8" ID ~36" Copper Pipe in my garage, at least 1 piece of 1/2" EMC conduit, and I think I even found a left-over ~24" 5/8" or 3/4" ID Cast Iron pipe, left by the previous owners. (Etc.)

In this scenario, cast iron would be a great choice, it's often used for (electro)magnetic shielding, and (AFAIK) was used in transformers and electromagnets before 'magnetic steel laminations' became common (we're talking 2ish centuries ago).
The worst thing that'll happen from wrapping a (safely and sufficiently insulated) wire around a pipe, is you'll be adding a magnetic load to the circuit (which, is part of the idea).
At the frequencies(48-62hz), voltages 80-270VAC and amperages (<5A) of a home fan, there should be no concern of overheating/overcurrent. You're basicallly creating a 'inductive filter'; not terribly dissimilar to what ferrite bead/chokes do. As with most-anything, given extremes and extenuating circumstance it could become unsafe.

Honestly, even a PEX, PVC, ABS, or CPVC pipe used as 'conduit' with some aluminum foil/tape on it can act as a shield. -Depends on what you're shielding from. -the suggestion came from a place of 'do diagnostic things, work from there'.

 

[Wirko]

There was a change. With a 150W bulb the PC sometimes wouldn't wake up, it was seemingly random as I repeated the test several times on a row.

If you're doing any tests with incandescent bulbs, don't forget that their inrush current is also very high.

I just checked a plain (not halogen) 230V/100W bulb and its resistance is 37.0 ohms when cold. Low enough to cause a short dip in voltage, which in turn triggers the (too sensitive) PSU in the PC. Halogen bulbs are even worse because there's a larger difference between cold and working temperatures.

 

[Count von Schwalbe]

In this scenario, cast iron would be a great choice, it's often used for (electro)magnetic shielding, and (AFAIK) was used in transformers and electromagnets before 'magnetic steel laminations' became common (we're talking 2ish centuries ago).
The worst thing that'll happen from wrapping a (safely and sufficiently insulated) wire around a pipe, is you'll be adding a magnetic load to the circuit (which, is part of the idea).
At the frequencies(48-62hz), voltages 80-270VAC and amperages (<5A) of a home fan, there should be no concern of overheating/overcurrent. You're basicallly creating a 'inductive filter'; not terribly dissimilar to what ferrite bead/chokes do. As with most-anything, given extremes and extenuating circumstance it could become unsafe.

With AC, a choke will introduce voltage drop. A perfect choke would allow no current past, but even a wire around an iron pipe can cut down the voltage pretty well.

More tiny chokes is better for EMF.