Support zones are usually found by
Price getting supported at a particular price zone.
Ideally when a few of the above ranges coincide at a particular support zone it assumes a lot of significance. .eg 200 day moving avg coinciding with 50 or 61.8% retracement level and we see stock holding up there and then bouncing with volumes.
Coming pointwise to your queries
In the absence of sufficient data, its very difficult to make out support levels. One way could be to find out retracement levels from bottom or top and find out the support zones at retracement levels.
If a stock remains between 100-115 for 6 months, support is nearer to 100 and resistance nearer to 115.
A breakdown/breakout from a range lasting only a day or two could be ignored as a whipsaw, provided previous range has been established for a lot of days.
Support zones are areas where stock price gets support on declines. In between there would be bounces and stock comes back to test the support zones and if support holds we can conclude that its a strong support. While looking at supports one cannot stick to only one level e.g 135 in case of equitas. It would have to be say 133-135 or so. Regarding time frame, there is no precise rule. The level from where a good bounce with volumes occurs is a good support.
Regarding my buying I usually look at breakouts and then some pullback post that to take a position. This is especially for technofunda bets. For bets where I like the fundamentals and management of the company (e.g MAS) I dont attach too much importance to support etc. If there is a clear picture its well and good otherwise I buy without looking at technicals too much.
Regarding absence of supports this happens if a stock corrects after it has had a parabolic rise. Usually seen at the fag end of a bull run. This happens because buying frenzy/panic had led to the parabolic rise and once the stock starts going down, everyone wants to bail out at the same time and there is a huge selling pressure. Or else it happens in case of negative news/frauds etc. In case of KNR it went up from 200 to 350 levels in around 2 months which is almost 75% gains. So once correction was to set in it can assume such steep cuts. Now its almost at the level from where the whole parabolic rise started. Specifically in case of KNR it would be advisable to wait and see how the stock establishes a range and consolidates. In the near future as there dont seem to be no triggers its better to be patient. Once the stock price shows some strength and comes out of a consolidation zone its ripe for the picking.