DIY transformer

Status
Not open for further replies.
Electromatic

Electromatic

Senior Member
Location
Virginia
Occupation
Master Electrician
I know the thread title sounds bad, but...
I'm trying to build a transformer for some in-house training based on this link:
https://www.allaboutcircuits.com/textbook/experiments/chpt-4/build-a-transformer/

I'm using stuff around the shop, so:
--individual strand of #24 Cat5 wire instead of magnet wire
--I've tried a 120-12V and 120-30V doorbell transformer to keep things kind of safe
--I've tried wrapping some 3/8 all-thread and a couple of long 1/4-20 bolts
--joining the 2 coils with a large ground bar instead of steel bars and bolts

I definitely get nothing out of the "secondary" and both my doorbell transformers drop down to about 1V when the "primary" is connected and no load on the "secondary".

Any tips for what I'm doing wrong? Better materials? More windings (I'd say I got around 100 on both tries)?

I'm fine with understanding what transformers do, but I admit I don't know that much about their construction.

Thanks in advance.
 
The impedance is too low for the source.

Try this- (i have not :D)
make a core of a hand-full of 10d nails with the heads cut off then dipped in varnish (the nails, not the heads)
wrap with 100's of turns of 24g or smaller wire for the primary
wrap that with a layer of vinyl tape
add another 100's of turns of wire as the secondary
 
I would suggest putting a 12V automotive incandescent lamp in series with the output of the 120V/12V transformer to limit the current.
Another alternative is to cut small gauge soft, easily bendable iron wire (non-galvanized) into a bunch of equal lengths and use that for the transformer core. You don't want hardened bolts because they are more easily magnetized and not suitable for AC magnetic fields.
As zbang said it's best to insulate the individual wires or nails from each other with varnish. But I think it should work well enough just for demonstration purposes without varnish, because I don't think they make very good electrical contact with each other compared to the impedances involved in the eddy current paths. That is, unless you sand and polish them ;). This is based on experience making an induction coil according to Alfred Morgan's Boy Electrician when I was young. But if you have some varnish on hand feel free to use it.
 
I like the idea of two separate rods and coils as in the link I posted. It seems like it could better illustrate the coils. Does that design stand a chance of working? Would rebar be any better than the threaded rods and bolts? Does the material linking the two rods/coils need to be particularly ferrous as well?

(Using some kind of rod is nice because I can chuck them in a drill to make the windings.)

Thanks again.
 
Electromatic said:
I like the idea of two separate rods and coils as in the link I posted. It seems like it could better illustrate the coils. Does that design stand a chance of working? Would rebar be any better than the threaded rods and bolts? Does the material linking the two rods/coils need to be particularly ferrous as well?

(Using some kind of rod is nice because I can chuck them in a drill to make the windings.)

Thanks again.
Click to expand...

The material linking the rods definitely needs to be ferrous so that it forms a complete loop that the magnetic field can pass thru.
If could mount a piece of small diameter plastic pipe on an arbor of some type (even a wooden dowel) and then rotate it with a drill that might work. If you can find one, an old time hand drill mounted in a vise would work better. Of course a lathe would be even better. Then when the coil is done you can insert the ferrous core through the plastic pipe.
Rebar would probably work OK because it is relatively "soft" iron.
 
Some transformers have an "O"-shaped core, with one winding on the left side, and the other on the right. Another company makes a 'Theta' shaped core (An O with a cross bar, a rectangular figure 8). One coil around each loop. To assist in winding the coils, one company made plastic half-spools, and glued them around the laminations. To wind the coil, they'd start the coil, then press the spool against a spinning 'tire,' which spun the spool and wound the coil. Made it possible to make the laminations in 1 piece, which greatly reduced 'transformer hummmm.'

ps- the spinning spools was for tiny audio transformers, not industrial 480v and up!
 
Even a small transformer used to demonstrate the working principle needs a CONTINOUS magnetic circuit, in the form of a circle, rectangle or other closed loop. A length of steel studding, pile of bolts or the like with exposed ends wont work well and probably wont work at all.

One way of achieving this is say 100 turns of iron fencing wire wound in a circular or doughnut formation. Keep all the strands close to each other with cable ties etc.
Wind the primary with say a few dozen turns of relatively thick wire, #14 single core would suit.
Wind the secondary the other side of the doughnut with about ten times the number of turns of thinner wire. Something like #20 or #22 might serve. Winding this is very tedious since the whole length of the wire must be passed through the core, hundreds of times.
This work may be eased by obtaining a small spool or roll of the secondary wire, NOT a loose length. Make the doughnut shaped core large enough that this small roll of wire may be passed through repeatedly during the winding of the secondary. (a young child is useful at this point)

Connect a very low voltage AC supply to the primary winding, try 4 volts or 6 volts.
The secondary winding should produce about ten times the input voltage. This may be demonstrated by connecting a small 50 volt incandescent pilot lamp, or several small 12 volt lamps in series.

Such a crude transformer will be of very low efficiency, so if the primary absorbs 10 amps at 6 volts, that is 60 watts, don't expect more than perhaps 10 or 12 watts output. say a small fraction of an amp at 50 volts.
 
Know the basic first.

Basic magnetic law:
Integral of V*dt = N*B* A,
N is # of turns
B is saturation flux (say 1.5Teslas for plain iron or steel)
A is core are in meters sq.

For 60 Hz, this reduces to
Volts on winding = N*1.5*area of core*377

Quick calculation for your 3/8" threaded rod (0.1 in sq) ( you need 28 turns per volt, or about 3500 turns for 120 V at 60 Hz.
A toroid made of 100 T of 0.06 dia iron fence wire (0.283 in sq) would need about 5 dozen turns for just 6V.
 
Yes, I did say "a few dozen" turns for 6 volts, based not on calculation but on experience, having made such a device which worked fine and was a good demonstration of the principles involved.
 
Status
Not open for further replies.
Top