jerkosovo,
I think there might be some confusion about your question. In aerodynamics, the term downwash is used quite specifically. I think you are getting answers to a different question than you mean to be asking.
The way I read your question, you want to know what is the phenomenon that causes the air above the wing to be forced down. There are several ways to answer this. In a way, you've already answered it: the Coanda effect is simply the tendency of air ( and other fluids ) to follow a curved boundary, such as the upper surface of the wing. Like you, I find it useful to have a more intuitive understanding than that. Kind of a bottom line view is that the flow of the air is 'driven' by pressure differences and momentum. The motion of the air around the wing simply creates pressure differences and relatively high momentum regions that cause it to want to follow a curved boundary, under the correct conditions. If the angle of attack exceeds a certain value, of course, the conditions are not correct, and stall ( separation ) occurs, reducing or eliminating lift. In a way, just accepting that air is going to follow that curve, and consequently impart downward momentum in the air, is the most useful view.
Downwash is, speaking somewhat imprecisely, the relatively small downward component of the velocity of the oncoming flow which is due to the wing generating lift. This results in a small decrease in the actual angle of attack, which varies along the span of the wing. It is not caused by air spilling from the high pressure bottom of the wing to the low pressure top, it is caused by the very act of generating lift. This is very different from the downward velocity of the air above the wing as it approaches the trailing edge, which must exist if the flow remains attached.
vortexman