Electromagnets, is thinner wire better? Quantity?

[kolleamm]

Let's assume you have two electromagnets. One has wire with a 1mm diameter and the other has half that diameter 0.5mm. Assuming they both have the same mass of wire which would be stronger or better?

Also if my goal was to attract an object, would it make more sense to have two thinner electromagnets next to each other rather than a thicker one by itself?

Thanks in advance!

 

[berkeman]

et's assume you have two electromagnets. One has wire with a 1mm diameter and the other has half that diameter 0.5mm. Assuming they both have the same mass of wire which would be stronger or better?

To make the best electromagnet, you will want to match the impedance of the electromagnet coil to the source impedance of your power source. What is the source of the current for this electromagnet? Is the source AC or DC?

When you know what the source impedance is of your power source, you can calculate how many turns of what gauge wire to use to match the source impedance (either at 50/60Hz or at DC). You will be able to use more turns of a larger wire to get the same coil impedance (if DC), but it will be more expensive and larger physically than using smaller wire. It's part of the tradeoffs you have to consider when trying to optimize the design.

What core material will you be using? Can you post a datasheet for the material, or other information (like its magnetic characteristics, μ_R, Saturation Flux, Hysteresis curve, etc.)?

 

[kolleamm]

To make the best electromagnet, you will want to match the impedance of the electromagnet coil to the source impedance of your power source. What is the source of the current for this electromagnet? Is the source AC or DC?

When you know what the source impedance is of your power source, you can calculate how many turns of what gauge wire to use to match the source impedance (either at 50/60Hz or at DC). You will be able to use more turns of a larger wire to get the same coil impedance (if DC), but it will be more expensive and larger physically than using smaller wire. It's part of the tradeoffs you have to consider when trying to optimize the design.

What core material will you be using? Can you post a datasheet for the material, or other information (like its magnetic characteristics, μ_R, Saturation Flux, Hysteresis curve, etc.)?

Hi thanks for your reply, the coil is coreless and it will be using DC current. The diameter is about 1 inch and the thickness about 1/4 inch for the electromagnet.
I have a variable power supply which allows me to change the voltage is necessary. Currently I'm using 32 gauge wire at around 9 volts but was wondering if I could make the magnet more effective than it already is.
This electromagnet will attract and repel a neodymium magnet so I can't use a core. Or maybe I could if the electromagnet makes the core repellent too? I'm not too sure.
Please let me know what you think , thanks again

 

[berkeman]

Can you say more about the application? For an attractive force, can you just use a solenoid actuator of some sort?

https://www.tinkery.com.au/wp-content/uploads/2017/04/dummy-solenoid.jpg

 

[kolleamm]

Can you say more about the application? For an attractive force, can you just use a solenoid actuator of some sort?

https://www.tinkery.com.au/wp-content/uploads/2017/04/dummy-solenoid.jpg

View attachment 232178

The application is supposed to be a force feedback on a glove which will create a force on each finger so it must be compact.

 

[Baluncore]

The weight of wire decides the strength of the force because it is proportional to amps * turns.
Twice the volts = twice the turns = half the wire section = half the current.

What limits the current through the coil?
Is it the wire resistance, or the inductance when driven by an H-bridge?

 

[Rive]

In a coil, there is usually only a fixed space available for wires. If you try to fill that up with different wires and run the maximal current available for the different wires then calculating the already mentioned Amper*turn value then for that fixed space it'll be ~ constant for any wire. So for wire thickness, it is really about matching the impedance with your source.

If you want more pushing/pulling force from the same coil then it'll be about the 'core': the magnetic circuit.

This electromagnet will attract and repel a neodymium magnet so I can't use a core.

That is not necessarily true. Many applications with strong magnets has a 'core' for the coil. It is just that it is not always inside the coil. Just take a look on the actuator used to move harddisk drive heads. There are magnets there, and even if the coil itself is 'empty', the magnetic circuit is carefully closed as much as possible.

 

[kolleamm]

In a coil, there is usually only a fixed space available for wires. If you try to fill that up with different wires and run the maximal current available for the different wires then calculating the already mentioned Amper*turn value then for that fixed space it'll be ~ constant for any wire. So for wire thickness, it is really about matching the impedance with your source.

If you want more pushing/pulling force from the same coil then it'll be about the 'core': the magnetic circuit.

That is not necessarily true. Many applications with strong magnets has a 'core' for the coil. It is just that it is not always inside the coil. Just take a look on the actuator used to move harddisk drive heads. There are magnets there, and even if the coil itself is 'empty', the magnetic circuit is carefully closed as much as possible.

Okay this is making more sense now. If the electromagnet is coreless would the neodymium have a tendency to be attracted to the center of the electromagnet? Or just the surface? Looking at magnetic field diagrams online the magnetic field lines remain constantly spaced in the middle.

 

[Rive]

If the electromagnet is coreless would the neodymium have a tendency to be attracted to the center of the electromagnet?

If no or not enough current flowing then it might stuck to the walls (closest to any ferromagnetic material nearby) instead. You need to have something to keep it in its track through the middle.
Having a permanent magnet in such situation is a bit troublesome.
Hard to give advice without knowing the details. Check on those solenoid actuators @berkeman linked previousy. For examle this. There is a 'core' and a 'guider'. But no permanent magnet.

 

[berkeman]

The application is supposed to be a force feedback on a glove which will create a force on each finger so it must be compact.

Having a permanent magnet in such situation is a bit troublesome.

I can see a number of problems, some pretty significant.

@kolleamm -- can you post a sketch of your initial design idea? (use the UPLOAD button in the Edit window to attach a PDF or JPEG file) It would be good to see how you envision using a permanent magnet and a coil in a force-feedback glove. Presumably the glove should make the user feel like there hand is closing on an object of some size, for VR movement of objects, etc.?

Depending on how you manage the magnetic path around those high-power permanent magnets, you could end up with problems associated with those fields going places that do not react well to such fields. Can you say what a couple of those places might be?

A current design approach for reference: https://cdn.shopify.com/s/files/1/2482/8512/files/TwoHands_800x640.png?v=1514890498

 

[kolleamm]

I can see a number of problems, some pretty significant.

@kolleamm -- can you post a sketch of your initial design idea? (use the UPLOAD button in the Edit window to attach a PDF or JPEG file) It would be good to see how you envision using a permanent magnet and a coil in a force-feedback glove. Presumably the glove should make the user feel like there hand is closing on an object of some size, for VR movement of objects, etc.?

Depending on how you manage the magnetic path around those high-power permanent magnets, you could end up with problems associated with those fields going places that do not react well to such fields. Can you say what a couple of those places might be?

A current design approach for reference: https://cdn.shopify.com/s/files/1/2482/8512/files/TwoHands_800x640.png?v=1514890498

View attachment 232219

Thank for your reply, the application is indeed for Vr. Here is a rough sketch of how it should work. An electromagnet and a magnet is placed on a finger. The forces between the electromagnet and the neodymium magnet causes a strap to either tighten or loosen. That's pretty much it.

 

[jim hardy]

One has wire with a 1mm diameter and the other has half that diameter 0.5mm. Assuming they both have the same mass of wire which would be stronger or better?

by "stronger" i assume you mean the field not the wire ?

Really it depends on how well your coil can dissipate heat.

Magnetic field strength you will find is in proportion to the product of amps and turns.
So there's no obvious inherent advantage to either of your postulated configurations over the other.

The smaller wire has 1/4 the cross sectional area of the larger
so will have 4X the resistance per foot
and you should be able to fit 4X as many turns into the same volume .

Now let's look at some practical arithmetic...
IF
from a given mass of copper you can make L feet of wire with diameter 1mm
THEN
from same mass of wire you can make 4L feet of wire with diameter 0.5 mm

since Power = I²R ;
IF
through wire of 1mm diameter you can pass current I and dissipate power P per foot
THEN
through wire of 0.5mm diameter you can pass current I/2 and still dissipate power P per foot
because I² X R = P = (I/2)² X 4R

Now since you can get 4X as many turns of smaller wire in the same space
and operate them at half the current with same P per foot
it would appear that your product of amps X turns will double
[(amps/2) X (turns X 4) ] divided by [(amps ) X ( turns ) ] = 2
meaning you'd get twice the field strength with the smaller wire.

BUT the gotcha is -
you most likely can't operate them at the same P per foot
because the coil made from smaller wire has 4X as many feet
and at 4X as many ohms per foot the coil made of smaller wire will have total resistance 4 X 4 = 16X as many ohms.
and its power dissipation is limited by its surface area.
Taking this line of thought to the extreme
IF you are limited by power dissipation in the coil
THEN you'll have to use 1/4 as much current to keep power dissipation the same
because (I/4)² X (16R) = I²R
and you'll have the same amp turns for both arrangements
because (I/4) X (turnsX4) = I X turns.
So there's no advantage in that case.

BUT -
IF
you're not limited by heat dissipation in the coil,
because perhaps your coil sees a small duty cycle so there is time for heat to conduct out between cycles
or it's a linear application that spends most of its time at low current
THEN
Yes there's an advantage to using the smaller wire in that you'll need less current to achieve a given field strength.
Just be sure you have enough voltage available in your power supply to push the requisite current through the ohms of your coil when it's at maximum expected temperature.

Make sense?

old jim

 

[kolleamm]

by "stronger" i assume you mean the field not the wire ?

Really it depends on how well your coil can dissipate heat.

Magnetic field strength you will find is in proportion to the product of amps and turns.
So there's no obvious inherent advantage to either of your postulated configurations over the other.

The smaller wire has 1/4 the cross sectional area of the larger
so will have 4X the resistance per foot
and you should be able to fit 4X as many turns into the same volume .

Now let's look at some practical arithmetic...
IF
from a given mass of copper you can make L feet of wire with diameter 1mm
THEN
from same mass of wire you can make 4L feet of wire with diameter 0.5 mm

since Power = I²R ;
IF
through wire of 1mm diameter you can pass current I and dissipate power P per foot
THEN
through wire of 0.5mm diameter you can pass current I/2 and still dissipate power P per foot
because I² X R = P = (I/2)² X 4R

Now since you can get 4X as many turns of smaller wire in the same space
and operate them at half the current with same P per foot
it would appear that your product of amps X turns will double
[(amps/2) X (turns X 4) ] divided by [(amps ) X ( turns ) ] = 2
meaning you'd get twice the field strength with the smaller wire.

BUT the gotcha is -
you most likely can't operate them at the same P per foot
because the coil made from smaller wire has 4X as many feet
and at 4X as many ohms per foot the coil made of smaller wire will have total resistance 4 X 4 = 16X as many ohms.
and its power dissipation is limited by its surface area.
Taking this line of thought to the extreme
IF you are limited by power dissipation in the coil
THEN you'll have to use 1/4 as much current to keep power dissipation the same
because (I/4)² X (16R) = I²R
and you'll have the same amp turns for both arrangements
because (I/4) X (turnsX4) = I X turns.
So there's no advantage in that case.

BUT -
IF
you're not limited by heat dissipation in the coil,
because perhaps your coil sees a small duty cycle so there is time for heat to conduct out between cycles
or it's a linear application that spends most of its time at low current
THEN
Yes there's an advantage to using the smaller wire in that you'll need less current to achieve a given field strength.
Just be sure you have enough voltage available in your power supply to push the requisite current through the ohms of your coil when it's at maximum expected temperature.

Make sense?

old jim

So basically your saying that if I can somehow manage the heat then thinner wire would be the way to go?

 

[jim hardy]

So basically your saying that if I can somehow manage the heat then thinner wire would be the way to go?

Yes.

Sorry it was so wordy -
While you didn't ask specifically "why", i thought from the tone of your question you'd appreciate the train of thought.


Rule of thumb in my day for power dissipation in small coils was approximately 1 watt per square inch of surface area.

 

[Rive]

Here is a rough sketch of how it should work.

I don't think that this application really requires a permanent magnet. Your problem is, that in this configuration you need a pushing force, right? Then just transform it to pulling force.

(Sorry, my drawing skills are... )

IF ... you're not limited by heat dissipation in the coil

Well, maybe for this one application he is...

 

[Nik_2213]

Rather than mount so much high-momentum hardware on your hands, could you mimic the 'tendon' approach and put the actuators on your fore-arms ??

You may be able to use lightweight pneumatics instead of bicycle cables...

Also, you may be able to use tiny linear actuators, motors or valves, instead of hand-made electromagnets...

 

[anorlunda]

A google search for "muscle actuator" returned a number of hits, including this one.
https://wyss.harvard.edu/actuators-inspired-by-muscle/

The picture shows the device activated/non-activated.

 

[jim hardy]

including this one

Wow ! I'm amazed not so much that they did it as they thought of it.
Face it I'm obsolete ,
But then,
Michelangelo studied natural mechanisms to build his machines, too.

 

[kolleamm]

These are all really brilliant ideas all of you have contributed! Thank you so much. I wish my email would have notified me of the replies. I will rethink this project, I'm thinking the forearm actuator idea was good since the fingers have very little space. I also thought maybe the use of air through tubes to exert a force might be a good idea.

 

[256bits]

A google search for "muscle actuator" returned a number of hits, including this one.
https://wyss.harvard.edu/actuators-inspired-by-muscle/

The picture shows the device activated/non-activated.

Interesting device. The muscles I knew about were the pressurized type
I wonder how the poor robot would be able to activate his muscles in space.
At Earth surface the strength seems to be limited by atmospheric pressure.

 

[Rive]

I will rethink this project

One small thing: for any VR application you should keep in mind the response time. I don't know how are you planning to sense the movements of your hand, but the feedback has to be in sync with the movement and the visual too. Maybe worth checking if anybody made studies about the acceptable delay?

 

[kolleamm]

One small thing: for any VR application you should keep in mind the response time. I don't know how are you planning to sense the movements of your hand, but the feedback has to be in sync with the movement and the visual too. Maybe worth checking if anybody made studies about the acceptable delay?

Your right about that, there's truly a lot of options at this point, I'm still somewhat leaning towards electromagnets since they may not necessarily take up too much space.