Thank you for your analysis.

I put your coeffiecents into Execl and found error is not acceptable.

Instead of using actual count i have use the weight calculated from the same ignoring fractions gives me some what good curve fit.
x = [0,20,40,60,80,100,120,140,160,180,200,220,240,260,280,300,320,340,360,380,400,420,440,460,480,500,520,540,560,580,600]
y = [0,37,80,114,150,178,199,223,240,256,270,286,296,300,304,303,303,297,289,275,268,245,224,202,179,155,124,90,48,18,0]

polyfit(x,y,2) gives me the following.

a0 = 4.2559
a1 = 2.0313
a2 = -0.0035


Ideally i want to find the error as a function of actual count.

Could you through some light on this.

V.Nagarajan