I want to write a function that takes the derivative of `f` wrt `y1` and then find the root of this derivative when `y1` equals `y2`. The function `f` contains several integrals, `y1`, `y2` and `x` also appear as limits of the integrals. Eventually I would like to be able to vary `x` and see how the root changes. 

Below is the closest I've gotten, but doesn't work.
```
fd[x_?NumericQ, y1_?NumericQ, y2_?NumericQ] := D[f[x, y1, y2], y1]
fr[x_?NumericQ, y10_?NumericQ] := 
 y1 /. FindRoot[fd[x, y1, y1]==0, {y1, y10}]
```
Am I missing a HoldAll condition for `y2` in `fd`?