homemade solenoid coil

hi folks,

i'm having some really bizarre behavior out of a homemade solenoid,
please read and help!

i've got some really thin "magnetic wire" from radio shack, i think
it's 30 gauge, that i'm using to create a solenoid coil. what i did
was took a drinking straw from one of those kid's box drinks, the
really thin straws, and i wrapped two layers of the wire around the
straw and solder the ends to a breadboard and attached leads to it.
then i stuck two different sized nails inside the straw and applied
current from a 12 volt radio shack power supply (actually 14 volts). i
am touching the coil so i can disconnect it when/if it gets hot because
it isn't protected by any kind of resistor or anything.

the behavior i get is unusual, or at least i didn't expect it. i apply
power and i get this rhymic kind of "jump" out of the nail. it will
get pulled to the center of the coil and kind of align itself so that
equal amounts of the nail are on either side of the coil. if i push
the nail out of the coil it will jump back into the coil, which is
great. but the unusual thing is that the behavior is in pulses, it's
not like there is a constant kind of "push" or "pull", but instead it
sort of sits there and has sort of a "push, push, push, push ..." kind
of heartbeat to it every second or so, or 1/2 second maybe. it's cycle
is very predictable. if i tilt the coil so that the nail kind of
slides out i can get it in equalibrium where the nail will slide down,
get jerked back up, slide down, jerked back up, etc, and move like a
cylinder in a car engine, up and down.

so my question is, i suppose, why does it pulse like this instead of
just having a constant force ? i was thinking it was something to do
with overcoming friction, that the nail would just sit still until some
kind of potential was high enough to make it overcome the friction it
has against the straw, but that's not it. intuitively, just watching
it, i can tell this is not what's going on. another thing i thought is
maybe the power supply was pulsing, but that doesn't seem to be the
case either.

what's going on ?

a basic electronics group seems like the right place to ask this