Pump magnetism

kusojiji · 11-09-2003, 10:15 PM

Anyone have problems with having their pump causing computer errors? Reason why I'm asking this is because I had a fan cause problems with my hard drive and I was wondering if a pump would do the same.

I have a FS020 case and had a 120 x 38 fan on the top outlet mounting. When I mounted the hard drives in the drive bay under the power supply, they were almost touching the rotor. I started to get data corruption. When I replaced the fan with a 120 x 25, the problems went away.

If I put the pump near a card or hard drive, I'm wondering what kind of problems I am going to see.

Anyone with similar issues?

Gooserider · 11-10-2003, 12:15 AM

I have no data, but it seems unlikely... Fans are mostly plastic, and presumably don't have much to keep magnetic fields contained inside the rotor. (Although your problem going away with only about a 1/2" (13mm) separation says there isn't much magnetic leakage problem on a fan anyways)

Most pumps will have more metal around the motor body which will tend to shield much of the fields. Plus you will have the pump guts on one end spacing the motor away from anything sensitive. Last the shape of most pumps would make it harder to mount a drive close enough to the pump to cause a problem.

At any rate, I can't think of seeing any threads reporting such problems (and even reports of fan induced problems like you mentioned are extremely rare...)

Gooserider

BrianW · 11-10-2003, 02:32 AM

I have had my pump within 2-4 in of my hd(s) for over 2.5 years with no problems. Eheim-1250 and hydor-l30.

Brian

kusojiji · 11-10-2003, 03:19 AM

Gooserider,

Yes, you are correct about plastic not being able to shield. I was surprised, however that the fan did cause my errors because of the large use of fans and the proximity of my fans to my hard drives in my other case (chieftec).

I don't think that these fish waterpumps are shielded either, though. Some of the larger pumps have a metal can motor while most small ones, like the Eheim for instance, are totally encased in plastic with the electrical stuff encapsulated in some kind of epoxy. I don't know what shielding properties epoxy has.

Would using a compass be a good test for checking the magnetic properties of a pump? I did that to an AC fan and found that it did not affect the compass at all while running.

I agree in that I haven't heard much about these types of problems. I'm posting this question on two other sites as well. The fan was really close to my drives when the errors were occuring. It wasn't only one error and it was happening intermittently.

BrianW,

That's good to hear.

Thanks both.

kusojiji · 11-10-2003, 03:22 AM

BrianW,

Why are you running two radiators? To reduce backpressure? You are sure running a lot of hardware for one waterblock. Your temps must not change much.

TerraMex · 11-10-2003, 01:05 PM

My 5 cents:

Neither a fan nor a pump generate a sufficiently large magnetic to physically affect the platters inside a hard drive.

You also need to take in account that the field drops with the square of the distance. A few cm (or less) are enough to drop the generated field to not very significant levels .

It would need something ALOT stronger than the earth magnets inside the hard drive, if you take in account that those are the ones that generate the levels on the platter, to affect them, because it had to beat the distance, and the shielding, and the error correcting methods of the hard drive and IDE controller.

And even if the generated field would interfere with the electronics on the back side, they are grounded and can deal with it. Besides, the PCB had to be touching the rotor. This was not the case.

(By default, a magnetic field can't be blocked. That's the rule. It can, however, be conducted or diverted. That's what the magnetic shielding in a lot of devices does. It gives the generated field a better medium to propagate than the air and/or the circuits inside the shielding. So it works both ways. The problem usually is to magnetize the material over a long period of use. Anyway.)

I've had water-cooling for a while and never had a problem with hard drives. Neither with corrupted files. And when I had a Fujitsu drive, it had 2 fans on it and never gave me problems.

Honestly, I'm sure (with a 99% sure

) the fan wasn't the source of your problems. It’s simply not logical given the physical facts.

jaydee · 11-10-2003, 01:11 PM

Quote:

Originally posted by TerraMex

Honestly, I'm sure (with a 99% sure ) the fan wasn't the source of your problems. It’s simply not logical given the physical facts.

Not magnetically, but I would suspect faulty wireing. If the fan was plugged into the same molex loop the hard drive was on and had a minor short or something it could cause problems. I killed a hard drive once with a faulty factory wired fan.

TerraMex · 11-10-2003, 03:19 PM

That could be. I was referring to the magnetic field.

That reminds me , in a lan (a few years back), a friend of mine toasted the whole computer with a faulty PSU, something shorted. Ok not the whole computer, the floppy drive and the CPU fan survived...

kusojiji · 11-10-2003, 06:22 PM

I hear what you're saying, but the only thing that changed was the fan. I have my fans plugged into a nexus controller with a noise filter on the power inlet. My hard drives are plugged into their own wire harness.

As for the proximity of the fan rotor to the hard drive pcb, it was really close. And if magnetic strength does decrease that much with distance, then the extra 13 mm of distance does make quite a difference. I am not experiencing any problems now.

Thanks for all the inputs. I'm beginning to get a better picture now. Keep em coming!

BrianW · 11-10-2003, 07:51 PM

Quote:

Originally posted by kusojiji
BrianW,

Why are you running two radiators? To reduce backpressure? You are sure running a lot of hardware for one waterblock. Your temps must not change much.

Well forever (actually 2.5 years), I just ran one heater core with my DD Maze2. I had planned on adding a GPU and CHipset block. That was the reason for the upgrade to two rads. The addition actually dropped temps an avg of 3-4 c across the board, so it was a good upgrade. Since I had problems with my GPU block for my 9800 ultra, I have had to postpone the addition of the other blocks in the loop. After careful consideration in this system, I have decided to just go two seperate loops with ther own pumps and rad's. The Pro core will be used for my Cascade, and the BI Pro will cool the GPU and NB.

Brian

kusojiji · 11-10-2003, 08:17 PM

Brian,

How does that pro core work for you? I am purchasing one and will be using it with three blocks.

On my other system, I have a giant Camaro heatercore and two AC fans. It is a 3/4 inch system that splits and reduces to 1/2 inch to a mcw5000 and dd geforce block. I used to be able to see the tiny airbubbles as they jetted through the tubing. they were really hauling, so I know that flow was good. The pump was (died) a dolphin 560 gph unit. It was making a racket so I opened it up to find a broken ceramic shaft. Since temps only rose 2 to 2.5 degrees C, I figured I didn't really need that big of a radiator. My AC fans are really throttled down and the temps barely move. Even if I crank up the fans, the temps don't change much.

BrianW · 11-10-2003, 11:59 PM

It works great. I would keep the core. In the space I had for a second, the BI Pro was about only one that would fit. I would rather have another HC any day!

BrianW

Gooserider · 11-11-2003, 08:47 PM

Quote:

kusojiji:
Gooserider, Yes, you are correct about plastic not being able to shield. I was surprised, however that the fan did cause my errors because of the large use of fans and the proximity of my fans to my hard drives in my other case (chieftec).

I agree with the concensus that it seems unlikely that the fan was the actual cause of your problem for the reasons stated. I don't have a clue about what else might have caused it though. However a fan SHOULDN'T be putting out that much of a field. A floppy is considerably less shielded than a hard drive, and it takes a pretty strong field to hose a floppy. Of course it may have been a case of having things in just the right (wrong) position to be a problem...

Quote:

I don't think that these fish waterpumps are shielded either, though. Some of the larger pumps have a metal can motor while most small ones, like the Eheim for instance, are totally encased in plastic with the electrical stuff encapsulated in some kind of epoxy.

There has to be some sort of metal can under the epoxy to make the windings work... Motor designers are usually pretty careful about designing things so that the fields are well contained, not only because of regulatory concerns, but also because stray magnetic fields don't do any useful work... The outer plastic shells are primarily cosmetic and protection against mechanical damage. However they also will act to space the motor guts away from any adjacent stuff.

Quote:

I don't know what shielding properties epoxy has.

No data, but unless it was metal filled, I would expect about the same properties as plastic. After all epoxy IS a form of plastic by some definitions!

Quote:

Would using a compass be a good test for checking the magnetic properties of a pump? I did that to an AC fan and found that it did not affect the compass at all while running.

Possibly. However keep in mind that a compass is just a magnet, and as such it will be impacted by any ferrous metals (steel, etc) along with magnetic fields.
I know that in the marine world, great care is needed when setting up a compass to ensure that it isn't adversely affected by nearby magnetic objects and electrical equipment. Usually special extra magnets and / or steel masses are used to cancel out such influences.

However I would not be worrying about it to much.

Quote:

I agree in that I haven't heard much about these types of problems. I'm posting this question on two other sites as well. The fan was really close to my drives when the errors were occuring. It wasn't only one error and it was happening intermittently.

Again, I don't know what caused your problem, but I really doubt it was the fan.

Gooserider

SexyMF · 11-12-2003, 11:31 AM

Pumps are electrically noisy. This can be a source of problems.

11-09-2003, 10:15 PM	#1
kusojiji Cooling Neophyte Join Date: Nov 2003 Location: Japan Posts: 86	Pump magnetism Anyone have problems with having their pump causing computer errors? Reason why I'm asking this is because I had a fan cause problems with my hard drive and I was wondering if a pump would do the same. I have a FS020 case and had a 120 x 38 fan on the top outlet mounting. When I mounted the hard drives in the drive bay under the power supply, they were almost touching the rotor. I started to get data corruption. When I replaced the fan with a 120 x 25, the problems went away. If I put the pump near a card or hard drive, I'm wondering what kind of problems I am going to see. Anyone with similar issues?

11-10-2003, 12:15 AM	#2
Gooserider Cooling Savant Join Date: Jun 2003 Location: North Billerica, MA, USA Posts: 451	I have no data, but it seems unlikely... Fans are mostly plastic, and presumably don't have much to keep magnetic fields contained inside the rotor. (Although your problem going away with only about a 1/2" (13mm) separation says there isn't much magnetic leakage problem on a fan anyways) Most pumps will have more metal around the motor body which will tend to shield much of the fields. Plus you will have the pump guts on one end spacing the motor away from anything sensitive. Last the shape of most pumps would make it harder to mount a drive close enough to the pump to cause a problem. At any rate, I can't think of seeing any threads reporting such problems (and even reports of fan induced problems like you mentioned are extremely rare...) Gooserider __________________ Designing system, will have Tyan S2468UGN Dual Athlon MOBO, SCSI HDDS, other goodies. Will run LINUX only. Want to have silent running, minimal fans, and water cooled. Probably not OC'c

11-10-2003, 02:32 AM	#3
BrianW Cooling Savant Join Date: Jun 2001 Location: Desert City in California Posts: 631	I have had my pump within 2-4 in of my hd(s) for over 2.5 years with no problems. Eheim-1250 and hydor-l30. Brian __________________ Water Cooled Inwin Q500 (Dual Rads: Rad1 = DTEK Pro Core \| Rad2 = Blick Ice Estreme, Hydor L30, Dangerden Maze2, Bay Res Typhoon Reservoir, 1/2 " DD Tygon Thick Wall Hose). Flow: Res, Pump, CPU watervlock, Y into both rads, both rads into res independently. Athlon XP 1800+ (@ 1731 - 150mhz fsb.), on a Asus A7N266-c, and a Radeon 9000 *waiting for RMA'd Saphire 9800 ultra from Newegg)

11-10-2003, 03:19 AM	#4
kusojiji Cooling Neophyte Join Date: Nov 2003 Location: Japan Posts: 86	Gooserider, Yes, you are correct about plastic not being able to shield. I was surprised, however that the fan did cause my errors because of the large use of fans and the proximity of my fans to my hard drives in my other case (chieftec). I don't think that these fish waterpumps are shielded either, though. Some of the larger pumps have a metal can motor while most small ones, like the Eheim for instance, are totally encased in plastic with the electrical stuff encapsulated in some kind of epoxy. I don't know what shielding properties epoxy has. Would using a compass be a good test for checking the magnetic properties of a pump? I did that to an AC fan and found that it did not affect the compass at all while running. I agree in that I haven't heard much about these types of problems. I'm posting this question on two other sites as well. The fan was really close to my drives when the errors were occuring. It wasn't only one error and it was happening intermittently. BrianW, That's good to hear. Thanks both. Last edited by kusojiji; 11-10-2003 at 03:26 AM.

11-10-2003, 01:05 PM	#6
TerraMex Cooling Savant Join Date: Nov 2002 Location: Portugal, Europe Posts: 870	My 5 cents: Neither a fan nor a pump generate a sufficiently large magnetic to physically affect the platters inside a hard drive. You also need to take in account that the field drops with the square of the distance. A few cm (or less) are enough to drop the generated field to not very significant levels . It would need something ALOT stronger than the earth magnets inside the hard drive, if you take in account that those are the ones that generate the levels on the platter, to affect them, because it had to beat the distance, and the shielding, and the error correcting methods of the hard drive and IDE controller. And even if the generated field would interfere with the electronics on the back side, they are grounded and can deal with it. Besides, the PCB had to be touching the rotor. This was not the case. (By default, a magnetic field can't be blocked. That's the rule. It can, however, be conducted or diverted. That's what the magnetic shielding in a lot of devices does. It gives the generated field a better medium to propagate than the air and/or the circuits inside the shielding. So it works both ways. The problem usually is to magnetize the material over a long period of use. Anyway.) I've had water-cooling for a while and never had a problem with hard drives. Neither with corrupted files. And when I had a Fujitsu drive, it had 2 fans on it and never gave me problems. Honestly, I'm sure (with a 99% sure ) the fan wasn't the source of your problems. It’s simply not logical given the physical facts. __________________ "we need more cowbell."

11-10-2003, 03:22 AM	#5
kusojiji Cooling Neophyte Join Date: Nov 2003 Location: Japan Posts: 86	BrianW, Why are you running two radiators? To reduce backpressure? You are sure running a lot of hardware for one waterblock. Your temps must not change much.

11-10-2003, 03:19 PM	#8
TerraMex Cooling Savant Join Date: Nov 2002 Location: Portugal, Europe Posts: 870	That could be. I was referring to the magnetic field. That reminds me , in a lan (a few years back), a friend of mine toasted the whole computer with a faulty PSU, something shorted. Ok not the whole computer, the floppy drive and the CPU fan survived... __________________ "we need more cowbell."

11-10-2003, 06:22 PM	#9
kusojiji Cooling Neophyte Join Date: Nov 2003 Location: Japan Posts: 86	I hear what you're saying, but the only thing that changed was the fan. I have my fans plugged into a nexus controller with a noise filter on the power inlet. My hard drives are plugged into their own wire harness. As for the proximity of the fan rotor to the hard drive pcb, it was really close. And if magnetic strength does decrease that much with distance, then the extra 13 mm of distance does make quite a difference. I am not experiencing any problems now. Thanks for all the inputs. I'm beginning to get a better picture now. Keep em coming!

11-10-2003, 08:17 PM	#11
kusojiji Cooling Neophyte Join Date: Nov 2003 Location: Japan Posts: 86	Brian, How does that pro core work for you? I am purchasing one and will be using it with three blocks. On my other system, I have a giant Camaro heatercore and two AC fans. It is a 3/4 inch system that splits and reduces to 1/2 inch to a mcw5000 and dd geforce block. I used to be able to see the tiny airbubbles as they jetted through the tubing. they were really hauling, so I know that flow was good. The pump was (died) a dolphin 560 gph unit. It was making a racket so I opened it up to find a broken ceramic shaft. Since temps only rose 2 to 2.5 degrees C, I figured I didn't really need that big of a radiator. My AC fans are really throttled down and the temps barely move. Even if I crank up the fans, the temps don't change much.

11-10-2003, 11:59 PM	#12
BrianW Cooling Savant Join Date: Jun 2001 Location: Desert City in California Posts: 631	It works great. I would keep the core. In the space I had for a second, the BI Pro was about only one that would fit. I would rather have another HC any day! BrianW __________________ Water Cooled Inwin Q500 (Dual Rads: Rad1 = DTEK Pro Core \| Rad2 = Blick Ice Estreme, Hydor L30, Dangerden Maze2, Bay Res Typhoon Reservoir, 1/2 " DD Tygon Thick Wall Hose). Flow: Res, Pump, CPU watervlock, Y into both rads, both rads into res independently. Athlon XP 1800+ (@ 1731 - 150mhz fsb.), on a Asus A7N266-c, and a Radeon 9000 *waiting for RMA'd Saphire 9800 ultra from Newegg)

11-12-2003, 11:31 AM	#14
SexyMF Cooling Neophyte Join Date: Jul 2003 Location: NZ Posts: 26	Pumps are electrically noisy. This can be a source of problems. __________________ A7N8X Deluxe AMD 2600+ 512MB Corsair XMS DDR333

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)