Thomas has given some examples of situations that are conceivable but not possible (in that they conflict with, say, laws of nature): for instance, in some sense one can imagine a puddle's turning into a human being, though such a transformation flies in the face of what we believe is physically possible.

But there are also circumstances which are physically possible of which we can form no picture. For instance, it's physically possible for there to be a chunk of quartz with a thousand facets, though I cannot imagine such a thing.

The question presupposes that we use the terms "conceivable" and "possible" in just one way, which is doubtful. For a little more, see Question 71.

Sometimes people use the word “possible” to mean something like, “does not entail a contradiction.” This sense of “possible” is narrower than “consistent with the laws of nature.” I gather this is what Thomas Pogge was alluding to when he said that there is a sense of “possible” that coincides with “conceivable.” Some have raised worries about whether conceivability implies possibility in this more narrow sense. Here’s an interesting example from Barry Stroud’s book on Hume (p. 50). Goldbach’s Conjecture states that every even number is the sum of two primes. At this point (as far as I know) no one has offered a proof of Goldbach’s Conjecture, and no one has disproved it either. If Goldbach’s Conjecture is false, it presumably says something that is not possible (in the narrow sense). But one might argue that I can nevertheless conceive of a state of affairs in which someone proves that Goldbach’s Conjecture is true. In that case, it seems, I would have conceived of something impossible.

Here’s another famous example from Arnauld. Arnauld is responding to Descartes’ claim in the 6th Meditation that since we can conceive of the mind existing apart from the body, it is possible that the mind can exist without the body. Arnauld objects that someone uneducated in geometry might be able to conceive of a right triangle that does not obey the Pythagorean Theorem. However this doesn’t mean that it’s possible for there to be a right triangle that fails to obey the Pythagorean Theorem.

Part of the problem is that it's not altogether clear what it is for something to be "conceivable."