error analysis

i hv two or three doubts in error analysis
first when we are calculating zero errors for vernier calipers we find that the positive and negaticve zero errors are different

mainly i cant understand this that wjen nothing is put between a screw gauge and the zero of the vernier scale lie 6 div below the refrence line

then we are minusing the zero error but logically if we think then we should add as the circular sclae reads lesss than the actual value??

2)

if the error measuremnt of momentum of a particle is 100% find the error in measuring its kinetic enrgy

3)

sometimes we are asked in exms like if the value of g decreases by 2% find the decrease in time period

my doubt is that we can obviuosly do the method by the long way but is there any short method to it from reading error analysis chap i realise that the best method is by using the formula for caculating errors when quantites are raised to some power.... am i rite????

4)

there was thiss some given in which u had to t=find the focal length of the lens

and as we know 1/f=1/v+1/u

now in dc pandey i could understand the steps for calculating the errors can anyone give me methods