ITM 0.1 km/h/year, and yes. The Pioneer Anomoly is a photon drive: the RTG gets hot, infrared photons get emitted preferentially in a particular direction (because the body of the spacecraft blocks them), and it's a simple reaction drive. Nothing fancy there.
You could make it loads more efficient by putting the RTG at the focus of a parabolic reflector. Now all the photons are being emitted in a single direction, which means you get much more thrust.
...a perfectly collimated photonic rocket runs at about 300MW/N. Assuming you're New Horizons, you have a mass of about 500kg and a 4kW RTG, so you'll get a thrust of 0.1mN, which means an acceleration of 5x10^-8 m/s^2. Assuming my arithmetic is correct (a totally unwarranted assumption!), then over a year this adds up to a velocity change of 1.5 m/s.
No, your maths is wrong.. in fact after 10 years the acceleration had provided a 1km/hr velocity, resulting in a 400km displacement from its projected position.
You could make it loads more efficient by putting the RTG at the focus of a parabolic reflector. Now all the photons are being emitted in a single direction, which means you get much more thrust.
According to this article:
https://en.wikipedia.org/wiki/Nuclear_photonic_rocket
...a perfectly collimated photonic rocket runs at about 300MW/N. Assuming you're New Horizons, you have a mass of about 500kg and a 4kW RTG, so you'll get a thrust of 0.1mN, which means an acceleration of 5x10^-8 m/s^2. Assuming my arithmetic is correct (a totally unwarranted assumption!), then over a year this adds up to a velocity change of 1.5 m/s.