I never had a rock hole collapse, per se, that I can remember. Mainly the drilling was in the Canadian shield and did include 300m holes in Saskatchewan and several 100m holes in Labrador. Normally, even in Ontario shales, we would drill in NW casing with a casing shoe and seat it into the shale about a foot or so. You might have to go a bit deeper if your rock surface is highly fractured. Mainly, in my view, this is to prevent any dislodgment of fractured rock that would block the retrieval of your core barrel.
Having said that, you may hit brecciated zones within the rock that might cause a problem - again, in blocking the retrieval of the core barrel. In such cases, you would grout the brecciated region of the rock, let it set and then drill it out. Similarly, there may be clay seams that might wash out but I doubt it would leave to 'cave' - perhaps in sandstones, you might run into sandy seams.
The point is that unless you know the geology quite well, how would you know what is there that might cause a 'cave'?
Anyway - that's my take for now! Chou and
When rock quality is bad, you are more likely to get fluid loss than collapse. Fluid loss indicates voids in rock structure, particularly in poor quality limestone or areas of none solution cavities.
Mike00,
your concern is a typical issue in oil drilling engineering, where stability of holes located in significant stress fields (thousands of feet of lithostatic burden)
may be critical, expecially in horizontal drilling.
While BigH and Ron said just about everything we need to know, if you want to pursue the subject further I think you have to post your question in some petroleum or drilling engineering forum (sorry, I don't know any).