Water Pumps and Magnetic fields.

[woodenman80]

Iam writing this to just give you possibly another easy to implement cheap protectivity. Iam no electrician or rocket techman but in my recently constructed water cooling setup ive noticed alot of electromagnetic output of the pump, my ehiem 1250 pump in this setup sits behind well close ish to the monitor about 9 inches away to be precise when i first wopped the setup to go mode my monitor was flickering severly i deduced that is was the magnetic field produced by the motor i was correct move the monitor away it all goes. Then i was thinking my god people put pumps like this in there pcs and dosnt magnetic thingamyjig damage there pc components, but then reading up some one elses debate on the forum this shouldnt happen due to the pump producing a fixed field etc.. but i still had a flickery monitor so i have made myself a simple shield which costed about 20 pence ish... using yes cooking foil, i decided that this thin sheet of flexible metal is just what i needed, so i produce a small wrap around jacket for my pump and voila!!!! no monitor flickering. Id just thought id stick this on a forum it may be complete bore to read but i thought it was a good idea. any reasons why not please pipe up :)...

Woodenman

Duron 750 oc to a mere 1000
512mb 133 cas2 crucial ram
geforce ddr
40.9 gb fujitsu hd
SBlive 5.1

Completely water cooled !0! fans (Passive cooling only)
20C idle
24C full load (well 3hrs of quake 3)

[The Overclocker]

yeh, anything with a magnet in can produce a magnetic feild, i have a huge problem with my wafedales which are right next to my monitor (i sloved it with 2 big sheets of steel which i grounded)

[bsblues]

i am thinking a lead lined box conected to ground if ever put a pump in my case. I would like my 1's and 0's to stay 1's and 0's

[azhari]

What you are talking about is what we call a Faraday cage. Essentially, it converts the magnetic field variation into a current inside the metal, which then dies down due to resistance (so the metal warms up very slightly).

In plain english, this means that you want a lining of copper not lead. If you were trying to shield yourself from gamma radiation, then lead would be better ;) So if you have a component that creates a variable magnetic field, such as a fan or a water pump, then wrapping copper around it would help a lot.

I work at a nuclear physics lab where we use highly sensitive detectors. Needless to say, we need to cut down the noise from magnetic fields and such. We actullay wrap things such as preamplifiers in aluminum foil. Just like a turkey. It works great.

[skrill]

Interesting

Could this be what caused the broken stripe in my RAID array?

So if I wrap my Eheim 1048 in Aluminum foil -- reflective side inwards I assue -- this would offer some protection?

[azhari]

I can't say for sure if that is what killed your stripping. After all, HD's have a metal case that reduces/blocks external magnetic fields. But nothing is impossible :D

Also, it doesn't matter which way the aluminum is facing. It's not reflecting the magnetic field, rather, it's converting the energy from the magnetic flux into heat.

[bsblues]

:) thanks for the info

copper it is then

[Christoph]

Let me see...
So far my cooling system will use
a shower head
6" PVC pipe
an aquaruim water pump
a water block
an anitbiotic
clear plastic tubing
aluminum foil


This is getting ridiculous

[Colin]

Neither aluminum or lead are permeable metals. They will not do much if anything for shielding magnetic fields. Take a look at this thread. (http://forums.overclockers.ws/vb/showthread.php?s=&threadid=12693)

[azhari]

It is true that aluminum is not the best for shielding from magnetic "field". There is a shield called mu-metal for that, and it is kinda expensive. But this is for a static (not changing magnetic field). Motors work by switching magnetic fields, therefore, causing a magnetic "flux". Any conductive metal is good for reducing this flux, and aluminum being so cheap and easy to come by is good for this use.

Just remember, there is a difference between a magnetic "field" and a "flux". You can ask any engineer, but as a PhD in nuclear physics, I know a thing or two about these things ;)

OK. I just checked the frequency range of the magnetic flux for a motor driving a pump. Unfortunately aluminum foil is NOT thick enough (in geek terms the "skin depth" is not enough). Best option would be, again, the mu-metal. But a cheaper option would be a steel box. It has a smaller skin depth, so don't need too thick a piece.

[Colin]

While I don't have a PhD, I do have a telephone. I spent some time on the phone with an engineer who specializes in magnetic shielding for aircraft and aerospace. The application notes are in the thread I linked too earlier and others I have responded to on this forum. It's important to note that if you use a cold rolled steel shield, there should be a gap between the shield and the pump. If you need a Mu metal shield over the steel shield, one of the thicker foils would be best, again with a gap. If necessary, a second thinner Mu metal shield can also be added. I have found it's much easier to reorient the pump or move the computers location relative to graphics pads and monitors.

Regarding data loss, the 100 milligaus field from a magdrive pump is not enough to worry about. Place a floppy on top of your pump for a few days. You won't find any data corruption.

[Christoph]

How far from am Ehiem 120 pump should my monitor be to avoid that nasty flux? I don't want to buy Mu metal, I've spent enough already ('course I suppose that most people don't like spending money like that either). If that's the way I have to go, are there some good Mu metal foil links?

[Crash893]

why woud you need to ground the box?

[Christoph]

I think that when a box stops flux, it transforms the flux into current, therefore if the box isn't grounded, it'll eventually build up enough charge to shock whatever's nearest to it, and around computers, most people don't like that.

[azhari]

The last time I needed mu-metal (which was years ago) I ordered it from a company called Hamamatsu. I don't know if they carry it any more, I'll have to check on that on Monday when I get to work. Then I can check with our purchasing agent to see who is our latest supplier of mu-metal. But I can tell you that its going to be very expensive (I mean hunderds). So you may want to just move the pump away from any sensitive equipment.

One trick that some of the technicians at our lab use is to take a cardboard box, wrap about 10 or 20 coils of copper wire around it and mount it on the monitor. Then they put some current through the wire (so you'd need a power supply). This is used for all monitors that are within 40 meter of the cyclotron (its a magnetic accelerator). It helps to lessen the warping effect on the monitors, but doesn't quite fix it. But then again, a cyclotron works with some extremely high magnetic fields. I don't quite know how much current they put through it, but given time I can sit down and calculate it. Or I could just ask them on Monday :)

To be honest, if you can't go with a steel box, then the cheapest solution would be to move the pump away. Other methods are quite expensive.

P.S. I never meant to imply that I know more than others just because I have a PhD, all I meant was that MY suggestion was made with some knowledge and that I wasn't making guesses. So I meant no disrespect to you Colin ;)

[Christoph]

Originally posted by azhari
But I can tell you that its going to be very expensive (I mean hunderds). So you may want to just move the pump away from any sensitive equipment.


If it's in the hundreds, I'll just find an alternative. I don't know how many of you all are made of money, but discussing the use of Mu metal shiedling at that price seems nutty. To each his own, I guess (no PhD :D ).

[Colin]

I never meant to imply that I know more than others just because I have a PhD, all I meant was that MY suggestion was made with some knowledge and that I wasn't making guesses. So I meant no disrespect to you Colin ;) [/B]

LOL, this is an overclocking forum. :D There are no absolutes. To set the record straight, despite my engineering background, I meant no disrespect either. In my experience, the best solution is to relocate the pump or other equipment.

[azhari]

Although mu-metal is expensive, so are the Vapochill cases ($600-$700), yet some OCers buy them (not me :D ). If you have the money then go for it. I agree moving thigns away from the pump is a good idea, but then its a matter of convenience vs. money. No matter how you cut it, there are tradeoffs.

I'm in the process of building a water cooling system myself and I'll probably end up encasing the pump in a steel box. Its pretty cheap and convenient at the same time (for me anyways).

[jhebel1]

On the topic of RF interference, I have a problem. Ever since I have converted my case to a water cooled rig I have been having problems with my CDRW drive. Also, I had an issue with my computer bluescreening on bootup when I closed my case's doors. I had to use electrical tape around the doors in order to get the computer to boot with them closed!

I have a Danner250 pump installed into an addtronics case (butterfly metal doors)

My CDRW drive is a Ricoh 12x, and since I have installed the pump I can only sucessfully burn at 2X!

What can I do to fix this problem???

Is my HD cable picking up some interference?

Please help me? :-(

[Boywonder]

Would Nilomag alloy 77 work? Its got a permeability rating of 60,000 (300,000 max). I don't know how much it costs yet. Also, where could someone get mu-metal? There's no way I would pay for it, but a friend of mine is interested. Thanks.

Edit:
Oops, I forgot the link to alloy 77: http://www.matweb.com/search/SpecificMaterial.asp?bassnum=NIND25