Controlling temperature of a water bath with induction plate

I am thinking of building a sous-vide cooker using some kind of prebuild heater, controled via an AVR and temperature sensor like the LM35. There are a lot of references around using rice cookers, crook cookers or immersion heaters. As induction plates are getting cheaper, this might be a notable alternative, not only regarding reaction speed (they seem to be close to water cookers) and evenness of of the heating, but also the space for putting it away.

My main question might be a little bit physical, but I assume it is still better asked here: While electric resistance heater can just be switched on and off to regulate the temperature, I am not sure how well that works for induction, as there are strong magnetic fields involved. Does anyone have experience there? I could, instead of simply switching it on and off repeatedly, interface the control panel of the induction plate from the AVR (what would be the best way to do it? I feel the heater should be galvanically separated from the controlling circuit, so just soldering a transistor parallel to the switch doesn't sound like a good idea ...). But the lowest setting might still be too high, thus completely switching off the heat source might be necessary ...

Any experiences or suggestions?

• I don't know of a means of "switching" thermal resistance other than mechanically moving things. Other parts of your question make less sense. Whether thermal resistance can be switched or not, I don't see how that relates to induction. What the AVR? You hadn't previoulsy mentioned one. What should be galvanically separated? It is difficult to tell what is being asked here. – Olin Lathrop Aug 24 '12 at 11:18
• Sorry, was a bit early this morning. I hope I clarified the post. The main error was that I confused "thermal resistance" with "electric resistance heater" but I added more background and connections. – mcandril Aug 24 '12 at 11:45
• Wasn't there a second answer yesterday talking about isolations? Just saw it on my phone ... – mcandril Aug 26 '12 at 14:43
• Yes, there was a second "answer", but it was from someone who has been banned here for a time. He keeps trying to sneak back by using various aliases, noising up the site, making busy-work for the mods, and pissing off everyone else in general. His posts are deleted whenever this happens since they weren't supposed to be here in the first place. No loss. Most of his posts contain a lot of incoherent babble. The guy is incapable of writing in a clear fashion taking the context of his audience into account. – Olin Lathrop Aug 26 '12 at 16:10
• Here's a nice sous vide implementation that may give you some ideas. flashingleds.net/sousvader/sousvader.html – Russell McMahon Jan 23 '13 at 16:54

Induction heating switches the current anyways, since a change in magnetic flux is required to induce eddy current. $$\varepsilon = {\Delta\Phi\over\Delta t}$$ Plain DC would give you nothing more than an electromagnet. Therefore cycling power wouldn't really hurt anything.