Oil filter magnets..

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Evadman

Administrator Emeritus<br>Elite Member
Feb 18, 2001
30,990
5
81
I just want confirmation that my idea [to use Paramagnetism to filter non-magnetic metals from engine oil] could theoretically could work.

Hold on, let me get my slide rule from the highly technical forum.

ROM estimate is that to deflect a piece of titanium the weight of a grain of sand (20mg) moving at 10 meters per second a distance of 5mm though a vacuum (not 30 weight so we can ignore some of the variables at least) it would take:

B= F/(A*S*.005/.0005s)/6.8x10^5
B = F/(A*S*10)/6.8x10^5
B=(.00002*(10/.005) * 1)/0.000068
B = 5880

A magnetic field of roughly 5880 tesla. The supercooled magnets of the LHC have a nominal 0.54 tesla and a maximum of 8.3 tesla each (can spike to that power twice per day) and there are 1624 of them at 27 tons each. So, assuming that you ran the LHC at half power so you could get perhaps 8 events per day instead of 2 (should be the square of the power theoretically so half power = 4 times more events). LHC at full power is 180mw, so half would be 90MW. So you would be able to deflect a piece of titanium 20mg in size into your trap approximately 8 times per day if you used a total of 90mw*24= 2160MWH per 24 hour period. Assuming you bought the power wholesale direct from a power plant, you could probably get it for around 4 cents per KWH so the total cost would be 2160000*.04 = $86,400 per day for power.

The dynamic viscosity of motor oil at operating temp (lets say 100c) is 0.010. that puts the force required to move that bit of titanium though 30 weight 100c motor oil at:

F =10m/s * 5^2 (0.010/5mm)
F = 10 * 25 * (0.010/.005)
F = 500 newton s

1 newton is actually pretty light and equates to lifting a kilogram per second^2. But toss this into where it needs to be in the equation above (replace F) and you end up with 2,940,000 tesla. So the actual power required for 8 events per day would be 2940000B / 5880B * 90MW = 45000MW or 45GW. The largest nuclear power plant in the world is Bruce at 3.2GW with 8 reactors.

So, theoretically, if you built 14 of the largest nuclear power plants in the world and ran them at nameplate capacity to power 708,500 superconducting magnets surrounding your oil filter, you would be able to remove 8 flakes of titanium from your engine oil per day. Your electricity bill would also be $520 per second. Total weight of the magnets would be 19,129,500 tons. For scale, the empire state building is 365000 tons. The magnets would weigh as much as 52 empire state buildings or 182 Nimitz class supercarriers. The entire US Navy's displacement is 'only' 3,121,014 tons. From the same link, you can see that the US Navy is bigger than the next 13 navies combined.

TLDR - Using 112 nuclear reactors and a weight of magnets roughly equal to the weight of every military and non-military ship on the planet surrounding your oil filter, it is theoretically possible to remove 8 titanium contaminant particles per day from your engine oil using only the power of Paramagnetism at a cost of 520 dollars per second (not including fixed costs for magnets, supercooling, and other infrastructure)
 
Last edited:

olds

Elite Member
Mar 3, 2000
50,127
781
126
Hold on, let me get my slide rule from the highly technical forum.

ROM estimate is that to deflect a piece of titanium the weight of a grain of sand (20mg) moving at 10 meters per second a distance of 5mm though a vacuum (not 30 weight so we can ignore some of the variables at least) it would take:

B= F/(A*S*.005/.0005s)/6.8x10^5
B = F/(A*S*10)/6.8x10^5
B=(.00002*(10/.005) * 1)/0.000068
B = 5880

A magnetic field of roughly 5880 tesla. The supercooled magnets of the LHC have a nominal 0.54 tesla and a maximum of 8.3 tesla each (can spike to that power twice per day) and there are 1624 of them at 27 tons each. So, assuming that you ran the LHC at half power so you could get perhaps 8 events per day instead of 2 (should be the square of the power theoretically so half power = 4 times more events). LHC at full power is 180mw, so half would be 90MW. So you would be able to deflect a piece of titanium 20mg in size into your trap approximately 8 times per day if you used a total of 90mw*24= 2160MWH per 24 hour period. Assuming you bought the power wholesale direct from a power plant, you could probably get it for around 4 cents per KWH so the total cost would be 2160000*.04 = $86,400 per day for power.

The dynamic viscosity of motor oil at operating temp (lets say 100c) is 0.010. that puts the force required to move that bit of titanium though 30 weight 100c motor oil at:

F =10m/s * 5^2 (0.010/5mm)
F = 10 * 25 * (0.010/.005)
F = 500 newton s

1 newton is actually pretty light and equates to lifting a kilogram per second^2. But toss this into where it needs to be in the equation above (replace F) and you end up with 2,940,000 tesla. So the actual power required for 8 events per day would be 2940000B / 5880B * 90MW = 45000MW or 45GW. The largest nuclear power plant in the world is Bruce at 3.2GW with 8 reactors.

So, theoretically, if you built 14 of the largest nuclear power plants in the world and ran them at nameplate capacity to power 708,500 superconducting magnets surrounding your oil filter, you would be able to remove 8 flakes of titanium from your engine oil per day. Your electricity bill would also be $520 per second. Total weight of the magnets would be 19,129,500 tons. For scale, the empire state building is 365000 tons. The magnets would weigh as much as 52 empire state buildings or 182 Nimitz class supercarriers. The entire US Navy's displacement is 'only' 3,121,014 tons. From the same link, you can see that the US Navy is bigger than the next 13 navies combined.

TLDR - Using 112 nuclear reactors and a weight of magnets roughly equal to the weight of every military and non-military ship on the planet surrounding your oil filter, it is theoretically possible to remove 8 titanium contaminant particles per day from your engine oil using only the power of Paramagnetism at a cost of 520 dollars per second (not including fixed costs for magnets, supercooling, and other infrastructure)

So where can I get this magnet?
 

tortillasoup

Golden Member
Jan 12, 2011
1,977
4
81
Hold on, let me get my slide rule from the highly technical forum.

ROM estimate is that to deflect a piece of titanium the weight of a grain of sand (20mg) moving at 10 meters per second a distance of 5mm though a vacuum (not 30 weight so we can ignore some of the variables at least) it would take:

B= F/(A*S*.005/.0005s)/6.8x10^5
B = F/(A*S*10)/6.8x10^5
B=(.00002*(10/.005) * 1)/0.000068
B = 5880

A magnetic field of roughly 5880 tesla. The supercooled magnets of the LHC have a nominal 0.54 tesla and a maximum of 8.3 tesla each (can spike to that power twice per day) and there are 1624 of them at 27 tons each. So, assuming that you ran the LHC at half power so you could get perhaps 8 events per day instead of 2 (should be the square of the power theoretically so half power = 4 times more events). LHC at full power is 180mw, so half would be 90MW. So you would be able to deflect a piece of titanium 20mg in size into your trap approximately 8 times per day if you used a total of 90mw*24= 2160MWH per 24 hour period. Assuming you bought the power wholesale direct from a power plant, you could probably get it for around 4 cents per KWH so the total cost would be 2160000*.04 = $86,400 per day for power.

The dynamic viscosity of motor oil at operating temp (lets say 100c) is 0.010. that puts the force required to move that bit of titanium though 30 weight 100c motor oil at:

F =10m/s * 5^2 (0.010/5mm)
F = 10 * 25 * (0.010/.005)
F = 500 newton s

1 newton is actually pretty light and equates to lifting a kilogram per second^2. But toss this into where it needs to be in the equation above (replace F) and you end up with 2,940,000 tesla. So the actual power required for 8 events per day would be 2940000B / 5880B * 90MW = 45000MW or 45GW. The largest nuclear power plant in the world is Bruce at 3.2GW with 8 reactors.

So, theoretically, if you built 14 of the largest nuclear power plants in the world and ran them at nameplate capacity to power 708,500 superconducting magnets surrounding your oil filter, you would be able to remove 8 flakes of titanium from your engine oil per day. Your electricity bill would also be $520 per second. Total weight of the magnets would be 19,129,500 tons. For scale, the empire state building is 365000 tons. The magnets would weigh as much as 52 empire state buildings or 182 Nimitz class supercarriers. The entire US Navy's displacement is 'only' 3,121,014 tons. From the same link, you can see that the US Navy is bigger than the next 13 navies combined.

TLDR - Using 112 nuclear reactors and a weight of magnets roughly equal to the weight of every military and non-military ship on the planet surrounding your oil filter, it is theoretically possible to remove 8 titanium contaminant particles per day from your engine oil using only the power of Paramagnetism at a cost of 520 dollars per second (not including fixed costs for magnets, supercooling, and other infrastructure)

Pardon my ignorance but 1. We're not talking about titanium but aluminum instead and 2. How do scrap yards that harvest metals from vehicles separate the different metals. I thought they used an electromagnet to do the separation of the different metals. Like when scrapping a car and they're chopping it up into small pieces, the copper wires in the electrical harnesses which are surrounded by rubber/plastic (wire), how do they separate that from the copper wire core on the inside?
 

Evadman

Administrator Emeritus<br>Elite Member
Feb 18, 2001
30,990
5
81
Pardon my ignorance but 1. We're not talking about titanium but aluminum instead
You linked to the article yourself, so do the math yourself for aluminum (hint: it changes 'not feasible' to 'not feasible'.)

How do scrap yards that harvest metals from vehicles separate the different metals. I thought they used an electromagnet to do the separation of the different metals.

You thought the single stream recycling truck that comes to your driveway has pixies in it that sprinkle fairy dust on metal and it instantly goes 'Poof' and sorts into buckets? No. It goes on a conveyor and lots of people sort it by hand. There are a few exceptions such as aluminum cans (which can be thrown around by eddy currents) and ferrous metals with can be sorted magnetically; but that doesn't include what you think it does. I already did the math on why what you think doesn't actually happen. I understand the numbers and words were big and that may have confused you so I have attempted to compensate by typing this post much slower than the last one.

Like when scrapping a car and they're chopping it up into small pieces, the copper wires in the electrical harnesses which are surrounded by rubber/plastic (wire), how do they separate that from the copper wire core on the inside?

Insulation is removed chemically or by that process we all know and love; fire. I am not your search engine; stop being lazy and solve your own ignorance.
 

foghorn67

Lifer
Jan 3, 2006
11,883
63
91
You linked to the article yourself, so do the math yourself for aluminum (hint: it changes 'not feasible' to 'not feasible'.)



You thought the single stream recycling truck that comes to your driveway has pixies in it that sprinkle fairy dust on metal and it instantly goes 'Poof' and sorts into buckets? No. It goes on a conveyor and lots of people sort it by hand. There are a few exceptions such as aluminum cans (which can be thrown around by eddy currents) and ferrous metals with can be sorted magnetically; but that doesn't include what you think it does. I already did the math on why what you think doesn't actually happen. I understand the numbers and words were big and that may have confused you so I have attempted to compensate by typing this post much slower than the last one.



Insulation is removed chemically or by that process we all know and love; fire. I am not your search engine; stop being lazy and solve your own ignorance.

Nominated for metallurgical pwnage of the year.
 

tortillasoup

Golden Member
Jan 12, 2011
1,977
4
81
You linked to the article yourself, so do the math yourself for aluminum (hint: it changes 'not feasible' to 'not feasible'.)



You thought the single stream recycling truck that comes to your driveway has pixies in it that sprinkle fairy dust on metal and it instantly goes 'Poof' and sorts into buckets? No. It goes on a conveyor and lots of people sort it by hand. There are a few exceptions such as aluminum cans (which can be thrown around by eddy currents) and ferrous metals with can be sorted magnetically; but that doesn't include what you think it does. I already did the math on why what you think doesn't actually happen. I understand the numbers and words were big and that may have confused you so I have attempted to compensate by typing this post much slower than the last one.



Insulation is removed chemically or by that process we all know and love; fire. I am not your search engine; stop being lazy and solve your own ignorance.

I thought they can't burn off the insulation due to environmental concerns. As for sorting non ferrous metals, why makes you think that eddy currents are limited to aluminum cans and certain other unmentioned metals when from what I understand, eddy currents can affect a whole host of metals. Where can I find a comprehensive list of metals that are affected and not affected by Eddy currents?
 

nerp

Diamond Member
Dec 31, 2005
9,865
105
106
Can't burn off? Have you ever been to a metals plant? There's quite a bit of heat and fire going on there.