Hi, i have chosen propeller clock as a school project and i could use some help from more experienced people, that I'm sure i will find here

for starters the project is few internet sources connected to something i think i like the most, but I'm open to new ideas. the motor is chosen to be 12cm 12V fan(idea by
http://hawelson.blog.cz/0705/virtualni-tocici-se-hodiny-propeller-clock) and from same design i like the idea of air core transformer for power to the rotating board
as for my part of this project i decided to base the "power supply" on the tiny 13 on which I'm using timers to generate
1x +-37,5kHz PWM for speed control of fan
1x +-37,5kHz PWM with duty cycle set to 50% just as square wave generator for IR led(I'm using integrated receiver)
and 1x software based +-120kHz 50% duty cycle square wave for air core transformer(the value is just randomly chosen by me with "higher f=better efficiency" in mind)
now there is button and led on board used for switching and signalling 3 modes just in case i could use them

Schematics:

OK so i have built this on the breadboard checked on cheap Chinese logic analyzer. all looked fine but when i put signals in transistor(on breadboard i used irlz24n), the output waveform Had huge spikes on rising edge
scope outputs
http://imageshack.us/g/1/9819740/first two are outputs on mcu pins(5V)
then there are mix of resistive load and tiny fan attached(i see no or very little difference) and last one is big 20W fan attached
this is the point I'm stuck on because i have no idea why is the ringing there and how to get rid of it. on the final board there will be different transistors(haven't arrived yet) but I'm not sure if transistor could be the problem. i was thinking of using 13V zener to cut "excess voltage" but haven't tried it yet.
so this is my little school project if anyone can help me with spikes or have any other ideas advices or just something in mind please say it so i can one day be at least as good as Dave
Thanks