I have a project in which I want a very small electromagnet to activate a very small lever for coupling/uncoupling NGauge model trains. I have wound a small coil (with 0.15mm enamelled wire) that is 6.5mm diameter and 9mm long. The power source is a small 80mAh LiPo cell (3.6v) - all controlled by an Attiny45.
It almost works as I want it to but I have been unable to find any advice about the factors that improve or worsen the performance of the electromagnet (apart from the obvious ones like more turns or more amps - I don't have space for more turns or battery power for more amps).
The problem is that it needs to move the lever a certain distance (about 3 to 4 mm) and the magnetic field is not quite strong enough to move the lever when it is 4mm away from the coil but works fine if it is only 2mm away (which is not enough). The lever is made from a piece of 16thou steel guitar string.
The questions that are in my mind include ...
I have a steel core (I will try to get a piece of iron later today - i.e. a nail). Is the magnetic force at the end of the steel/iron core affected by how far the core sticks out of the coil?
Is the magnetic force increased or decreased if there is a "head" on the core - like the head of a nail, for example.
I have taken a small relay apart and they seem to be designed so that the moving piece is magnetically connected to one end of the electromagnet and attracted to the other end. I don't think that would be practical for my project because my lever must be very lightweight.
The websites that Google has turned up all seem to be about very basic electromagnetism demos using a nail or very advanced industrial stuff.
I wonder if any forum reader has any practical advice or a link to anything useful?
Robin2:
I have a project in which I want a very small electromagnet to activate a very small lever for coupling/uncoupling NGauge model trains. I have wound a small coil (with 0.15mm enamelled wire) that is 6.5mm diameter and 9mm long. The power source is a small 80mAh LiPo cell (3.6v) - all controlled by an Attiny45.
It almost works as I want it to but I have been unable to find any advice about the factors that improve or worsen the performance of the electromagnet (apart from the obvious ones like more turns or more amps - I don't have space for more turns or battery power for more amps).
The problem is that it needs to move the lever a certain distance (about 3 to 4 mm) and the magnetic field is not quite strong enough to move the lever when it is 4mm away from the coil but works fine if it is only 2mm away (which is not enough). The lever is made from a piece of 16thou steel guitar string.
The questions that are in my mind include ...
I have a steel core (I will try to get a piece of iron later today - i.e. a nail). Is the magnetic force at the end of the steel/iron core affected by how far the core sticks out of the coil?
Is the magnetic force increased or decreased if there is a "head" on the core - like the head of a nail, for example.
I have taken a small relay apart and they seem to be designed so that the moving piece is magnetically connected to one end of the electromagnet and attracted to the other end. I don't think that would be practical for my project because my lever must be very lightweight.
The websites that Google has turned up all seem to be about very basic electromagnetism demos using a nail or very advanced industrial stuff.
I wonder if any forum reader has any practical advice or a link to anything useful?
...R
The factors are amp-turns and magnetic circuit, that's all. Sounds like you might be
able to improve the magnetic circuit - use a U-shaped core so that both poles point to the
lever. It won't matter where the coil is on the core if its ferromagnetic.
Choosing the right diameter wire is important - you want the coil resistance to match
what the battery can supply. Finer wire will have more turns but much less current
than optimal, thicker wire may overload the battery.
Halving the wire diameter means four times as many turns but 16 times the resistance,
so the amp-turns is down by a factor of 4 (for a given supply voltage). Suggest you
measure the coil's resistance and see if its close to V/I where V=3.7V and I is the maximum
current you can afford to give to the coil when its on (circuit/battery limits, battery life).
Robin2:
I have taken a small relay apart and they seem to be designed so that the moving piece is magnetically connected to one end of the electromagnet and attracted to the other end. I don't think that would be practical for my project because my lever must be very lightweight.
Your electromagnet will be much stronger if you close the magnetic circuit, as in the relay. One possibility may be to bend the nail into a hairpin or horseshoe magnet shape, and wind the coil between the inside and outside of that.
Other than that, it's a matter of using slightly thicker wire to get more ampere-turns for the same voltage, as already explained.
I didn't/don't think I have room for a horseshoe shape.
It occurred to me that it might help if the metal core could be extended closer to the lever but arranged so the lever couldn't touch it. A few minutes fiddling with a paper clip has resulted in a very satisfactory result. The attached diagram (not to scale) should illustrate what I did. The black lines are the paper clip. The blue rectangles are a cross-section of the coil and the red and blue lines are the lever in its normal and activated positions.
I tried to include an image in the middle of my previous post but it doesn't show up at all in my browser. If anyone else can see it I would appreciate it if you would let me know. (It's the same as the attachment).
An air gap of 3 to 4 mm is a nearly 'infinite' distance for a magnetic field, because the field decreases reciprocal to the square of the distance.
Even using an iron core wouldn't change this, it would just increase the static retention force of the magnet.
You could use a 'pot magnet', or more general a 'charchteristic curve affected magnet' (Attention: Two direct german translations ;-). They are quite difficult to design.
The second alternative is an 'moving coil actuator' (the direct german translation is really funny: diving coil actuator ;-). Most probably you know the principle very well, because it's driving ordinary audio speakers. This would be my recommendation. Maybe you can take one from a bass speaker.
The problem with a moving coil actuator is that it would have to be connected to the lever and the combination would make the lever too heavy (or with too much inertia).
The entire model railway wagon that I am using would probably fit inside a bass speaker coil
I think/thought the purpose of the iron core is to concentrate the magnetic field?
Thanks @jackrae, but what I'm looking for is info on designing electromagnets, not about how to install them. The point motors (I have some) are much too large for my project.
My project (Attiny, IR detector, battery, transistor, resistors and coil) has to fit within a space that is smaller than the point motor.
How about turning the problem on its head. Have a permanent magnet, one of those small strong ones for the coupling and then the electromagnet makes an opposite pole for the uncoupling.
How are you actually turning on the electromagnet coil with the arduino ? Trying to drive it directly from an arduino output ? Using a transistor ?
You need to think about what voltage and current is being actually applied to your coil.
The other thing to consider, if you electromagnet will actually pull 2mm, and you want to pull your lever 4 mm, then if your magnet is strong enough, set up another level to magnify the distance ( and halve the force ) that the magnet will pull.
In post #3 I explained how I solved the problem. However I am still open to other ideas.
@grumpy-mike - I foresee 3 problems with using a permanent magnet. First, it would be very heavy compared to 30mm of 16thou steel guitar string which is what the lever is made from. The extra weight would probably interfere with the coupling process which isn't (and doesn't need to be) automated. Second, I suspect the magnet would be inclined to "stick" in the activated position. Third, there is probably not enough space.
@michinyon I am using a transistor to control the current. I think the resistance of the coil plus a 5R6 resistor is about 10 ohms and with 0.7v(?) drop in the transistor the coil current might be about (3.8-0.7)/10 or 300mA. I don't really want to draw more from the 80mAh battery in case it causes a brown-out for the Attiny. (By the way, the electromagnet is only activated for 1 second at a time).
I don't understand what you have in mind in your final paragraph.
Robin2: @michinyon I am using a transistor to control the current. I think the resistance of the coil plus a 5R6 resistor is about 10 ohms and with 0.7v(?) drop in the transistor the coil current might be about (3.8-0.7)/10 or 300mA. I don't really want to draw more from the 80mAh battery in case it causes a brown-out for the Attiny. (By the way, the electromagnet is only activated for 1 second at a time).
I'm glad you got it working. Two comments:
If you use a suitable transistor (e.g. BC337) with a suitable value of base resistor (e.g. 180 or 220 ohms), then the voltage drop in the transistor will only be 0.1 to 0.2V. OTOH if you use the wrong type of transistor or an unsuitable value of base resistor, then the voltage drop in the transistor may be much higher than 0.7V.
The fact that you are having to use a series resistor to limit the current implies that you would have been better off using more turns of thinner wire, assuming you are using that maximum number of turns of your present wire size that will fit.
@dc42, I agree about using thinner wire. However I tried using some I "borrowed" from a small relay and found it broke too easily. I haven't had that problem with the 0.15mm wire.
Also I had/have a very limited appetite for making several coils to determine how much resistance there would be when I have no method to know in advance how many turns would fit and thus what length of wire and resistance.
I have a 1k base resistor so maybe it would be better with the lower values you suggest. That will have to wait until I make a second example because it would be too much trouble to take it apart.
@Grumpy_Mike, I suspect you have a picture embedded in your post and I see nothing - not even a link to click.
This is not the first time this has happened for me since the website changed and I don't understand why. Could you post the link to the picture so I can try it separately from the forum? (I assume the picture is hosted elsewhere).
Obviously I can't comment on your magnet suggestion
