It would be interesting to see a picture of your centrifuge, which you appear to have built in a few days.
Centrifugation simply accentuates the effects of gravity, and since bacteria are denser than aqueous solutions they will move to the bottom of the sample to form a fairly tight pellet.
The conditions for your centrifugation will depend very much on the design of the centrifuge since as well as the rotational velocity, the radius of rotation is also important.
I found a suggestion on the internet to use a relative centrifugal force (RCF) of 12000 g for 2 minutes (I must admit that from memory I always spun slower and for longer than this suggests). Anyway...
using RCF = 1.12 * 10-5 * r *N2
r = radius of rotation
N is revolutions per minute
and plugging 12000 g into that, for a rotational radius of 10 cm, I get 3300 rpm.
Don't know if you can manage that with your machine, but as I say, a lower RCF with a longer spin will probably be ok as long as you can dilute the yoghurt sufficiently to reduce viscous drag. Of course if you have access to any microbiological growth media you could grow a standard liquid culture by inoculating some sort of broth with a small amount of yoghurt.
Try a Google image search for hand powered centrifuge to get an idea of what has been used in the past.
after comment from Mad Scientist
see here for a typical large table top centrifuge maximum RCF is 3120 g at 4600 rpm. You can certainly pellet bacteria in something like this in tens of minutes.