Re: DIY electromagnet under £100: Disappointing results

On Oct 2, 2:34 pm, John Larkin
<jjlar...@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
On Fri, 02 Oct 2009 13:07:51 -0500, John Fields





<jfie...@xxxxxxxxxxxxxxxxxxxxx> wrote:
On Fri, 2 Oct 2009 07:38:49 -0700 (PDT), Jessie
<jessie.taylo...@xxxxxxxxxxxxxx> wrote:

Hi everyone,

I took part in a competition of my college's electronics club to build
an electromagnet under £100, with the strongest and cheapest one
winning.

Approach #1:
I bought a 500W ATX power supply and used the 12V 35A lead to energise
an electromagnet. I used a full 100m length of 0.8mm thick enameled
copper wire - that length had a resistance of about 5 ohms. The wire
reel in the as-bought state conveniently had both leads exposed.

I put a 10mm steel hex bolt through the reel's centre hole as the core
material. The electromagnet was feeble and was no where even close
enough to carrying its own weight :(

Approach #2:
I also tried attaching a 6kV Cockroft-Walton voltage multiplier (as in
instructables.com) to the same 100m 0.8mm thick enameled copper wire
reel, but although the current was presumably very high it was of such
short duration that some 5mm chrome steel bearings I had nearby didn't
even budge :(

So I'm back to square 1 and am open to suggestions on how to create
the strongest possible electromagnet under a tight budget.

---
For a solenoid, its internal magnetic field is described by:

         µ N I
    B = -------
           l

where µ is the permeability of the core,
     N is the number of turns surrounding the core,  
     I is the current in the coil, and
     l is the lenghth of the core

so, you can see that B will increase if the permeability of the core
increases, the number of turns increases, the current increases, or the
length of the core decreases.

To get the maximum strength, then, you want to wind a short solenoid
with a lot of low resistance wire because, as the resistance increases
the current will decrease for a given supply voltage.

Also, you'd like the external field to add to the internal field, so
you'd want the solenoid to be encased by a high-permeabilty core.

To do that you'd want the core to be shaped like a bundt cake pan, with
the coil nested inside of it, like this side view: (View in Courier)

.        +---+         +---+         +---+
.        |   |ooooooooo|   |ooooooooo|   |
.        |   |ooooooooo|   |ooooooooo|   |
.        |   |ooooooooo|   |ooooooooo|   |
.        |   |ooooooooo|   |ooooooooo|   |
.        |   +---------+   +---------+   |
.        |                               |
.        +-------------------------------+    

That way, the part of the magnetic field which would be lost will be
captured in the walls of the "core" and will add to the pull of the
center leg.  

That's how junkyard electromagnets are made, BTW.
---

Incidentally, am I correct in concluding that the length of the wire
is not relevant to the magnet's strength as the resistance is
proportional to the length but the strength is proportional to the
number of turns (hence length) so both terms cancel out?

---
Both lengths _don't_ cancel out since one is the length of the wire used
to wind the coil and the other is the length of the wound solenoid.
---

So presumably the only electrical considerations as far as a resistive
electromagnet is concerned is how many amps you can put through it
before you burn it out.

---
In a sense, yes, but you also need to be concerned with how many turn of
wire you can get on the thing because that, and the current in the coil
(ampere - turns) will determine the strength of the magnet.
---

Does this imply that thick copper wire (a few
mm at least) with a large thermal mass and a very high voltage, high
capacitance power supply is the way to go?

---
Not necessarily.

If you want to get a feel for the numbers, here ya go:

Let's say we have a core that looks like this:

.            |<---1"--->|   |<--1/2"
.        +---+          +---+          +---+ ---
.        |   |          |   |          |   |  |
.        |   |          |   |          |   |  |
.        |   |          |   |          |   |  1"
.        |   |          |   |          |   |  |
.        |   |          |   |          |   |  |
.        |   +----------+   +----------+   | ---
.        |                                 |
.        +---------------------------------+

And we want to see how many turns of #10AWG magnet wire we can get on
it.

From copper wire tables we find that #10AWG has a nominal diameter of
0.102", so we can get 10 turns around the middle leg, like this:

.        +---+          +---+          +---+
.        |   |         O|   |O         |   |
.        |   |         O|   |O         |   |
.        |   |         O|   |O         |   |
.        |   |         O|   |O         |   |
.        |   |         O|   |O         |   |      
.        |   |         O|   |O         |   |        
.        |   |         O|   |O         |   |      
.        |   |         O|   |O         |   |          
.        |   |         O|   |O         |   |          
.        |   |         O|   |O         |   |        
.        |   +----------+   +----------+   |          
.        |                                 |
.        +---------------------------------+

Now, with the leg being 1/2" in diameter, each turn will have a length
of:

    C = pi D = 3.14 * 0.6" = 1.88"

and the length of the 10 turn layer will be 18.8"

Tabulating the results for all ten layers:

 LAYER     LENGTH
           INCHES
--------+-----------
   1       18.8
   2       25.1
   3       31.4
   4       37.7
   5       44.0
   6       50.3
   7       56.5
   8       62.8
   9       69.1
  10       75.4
          ------
    TOTAL 471.1" ~ 39'

So now our electromagnet will look like this:

.        +---+          +---+          +---+
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   |OOOOOOOOOO|   |OOOOOOOOOO|   |
.        |   +----------+   +----------+   |
.        |                                 |
.        +---------------------------------+

and with 39 feet of wire in it, since #10 has a resistance of about 1
milliohm per foot, it'll have a resistance of 39 milliohms.

Now, let's say that this thing can dissipate 100 watts, continuously.

Since:

    P = I²R

then:
              P            100W
    I = sqrt --- = sqrt --------- = 50.64 ~ 50 amperes,
              R          3.9e-2R

the voltage across the coil would be,

    E = IR ~ 50A * 3.9e-2R ~ 2 volts

and its IT would be 50 amperes * 100 turns = 5000 ampere - turns.

Google "ampere-turns" for some rather more detailed info.

If an electromagnet is thermally limited, and you plan to fill the
available winding window, I think it doesn't matter what size wire you
use. So size the wire to match whatever power supply you have, to get
as many watts into the coil as it can stand.

It's like a transformer. A 100 VA transformer is the same size whether
it has a 120 volt primary, a 240 volt primary, or a dual primary.

Square wire *is* better for a couple of reasons.

John- Hide quoted text -

- Show quoted text -


"> Square wire *is* better for a couple of reasons."

Yeah!!! I've always wanted to try a coil with square wire. It must
lay in the groove real nicely! Good for precision coils.

George H.


.

Quantcast