Taking time control into account, how long should one take when looking for a combination? Also, what factors should one consider when looking for a combination?
That's a very general question that I don't think you will get a solid answer for. The chance of a powerful combination being available depends largely on the pieces each side has available and developed. So the answer varies based on what stage the game is in and the positions of each side.
Instead I would focus on identifying the types of combinations you are comfortable with, and start to learn to identify boards where those combinations MAY be present, and others where the combinations is not present. If you can speed up your identification you can determine on the fly how much time to spend on evaluating each potential move.
The time spent on looking for combinations is a function of board position.
If you have few active pieces, spend minimal time looking for combinations, and more time developing strategic plans.
As more pieces come into play you will look for combinations more, particularly if their are open lines for attack. Ideally, you eventually arrive at a board position where there simply must be a combination available. At this point you search until you find it.
While good players are always alert for tactics, there are a few key "signals" that indicate that tactics and combinations might be available.
If there are one or more enemy pieces that are hanging or underdefended, you should look for tactics that attack that piece. For example, if a bishop is defended by an enemy queen and attacked by a friendly rook then the bishop is underdefended.
If the enemy king does not have many friendly pieces around it, then it might be possible to stir up an attack. If the king is castled, look for ways to disturb the pawn shield in front of the king, or look to see if the h7 square can be attacked. For example, a Greek gift sacrifice might be possible.
If the king has not castled, then be alert for tactics based on the f7 square. f7 is frequently weak because only the king defends this pawn.
Lead in Development
If you have a moderate lead in development (an extra piece or two developed), then it is frequently possible to do something useful with the extra forces. Most usually this is an attack on the enemy king, but it could be something as simple as gaining a space advantage too.
Enemy Piece Low on Squares
If an enemy piece doesn't have many squares to which it can safely move, it might be possible to trap the piece. This most often happens with the enemy queen, but it is possible to trap rooks, knights, and bishops as well. Here's an example from one of my own games:
1...Re1+ 2.Kg2 Be3! -+
Moves to Consider
If you have decided to look for tactics, you should consider all of the checks available as well as all of the captures. This may sound slightly radical, but almost all tactics will fall into one of these two categories. Moves that place a piece close to the enemy king should also be studied. Almost all chess tactics are based on double attacks (or double "threats"), so look for moves that unmask an attack (discovered attack) or that fork two pieces.
Amount of Time to Spend
Depending on how much time you have left, you should spend as much time as you need to look at all of the possibilities. If you have only 5 minutes left, then it might be prudent to just look at obvious moves fairly quickly, but if you have more than an hour left on the clock, there is no excuse for missing a winning tactic.
With practice you will be able to look at all of the captures and whittle them down to one to three worth considering very quickly. For example, sacrificing a queen for a bishop should at least be checked out (it's a capture), but almost all the time it is sufficient to look just another move further and decide that there is no follow up. The best players spend over an hour on a single move if it is possible that there is a winning attack in the cards. The biggest reason for this is that they are able to win with just a few minutes left on the clock, so it is worth the initial time investment in order to gain a winning position.
To conclude, here's another example from one of my own games. I was black, and I got to this position. Notice that I am up a piece, but black now has a knockout blow to end the game immediately. I spent some time looking for tactics because the enemy king is weak and the king is also low on squares (only can play
Kxg2). Since it is likely that a successful tactic will win the game, I was willing to spend about half of my remaining time on the move (although I spent significantly less than that in the game). I had about 35 minutes left.
1...Bxg3! 2.hxg3 Qh5 0-1