Not able to match the number of components in a system with the "species-approach" to the Gibbs phase rule

Question

The Gibbs' phase rule states that the number of degrees of freedom in a system $F = C - P + 2$ where $C$ are the number of components and $P$ are the number of phases. The term $C$ can be expanded to $C = s-r$, where $s$ are the number of species, and $r$ are the number of independent relations.

I'm trying to find number of components for two systems, but the two approaches doesn't add up. Where am I wrong?

System 1:

A closed, evacuated container with $NH_4Br(s)$ is heated, and the following equilibrium takes place: $NH_4Br(s) ⇌ NH_3(g) + HBr(g)$.

I believe the number of components C to be 1, $NH_4Br(s)$, since this substance alone can produce the gas mixture phase, and the solid phase. ($NH_3(g) + HBr(g)$ can also do this, but C should be the lowest number of components needed..).

Using the species approach, $s = NH_4Br, NH_3, HBr = 3$, and $r = $ 1 equilibrium reaction. Which gives that $C = s-r=3-1=$ 2 which doesn't match what I wrote above. Why?

System 2:

A saturated aqueous solution of $KCl(aq)$ with $KCl(s)$. There is also water vapor. I also assume this is an evacuated container, but the problem doesn't mention this.

I believe the number of components to be 2, $H_2O$ and $KCl$, since these substances can produce the solid phase $KCl(s)$, the gaseuous phase $H_2O(g)$, and the liquid phase $H_2O(l)+Cl^-(aq)+K^+(aq)$.

Using the species approach, $s = H_2O(aq/g), K^+(aq), Cl^-(aq), KCl(s)$, and $r =$ one phase with ion-balance + one equilibrium between water liquid and gas + one equlilibrium between $KCl(s)$ and the dissolved ions = 3. Which gives that $C = s-r = 4-3 = $ 1, which again doesn't match what I wrote above.

Answer 1

The Gibbs phase rule for reacting systems is $$F = 2 - \pi + N - r - s$$ Where $F$ is number of degrees of freedom, $\pi$ is number of phases, $N$ is the number of components, $r$ the number of independent reactions and $s$ the number of extra constraints on the phase rule variables. These constraints originate by the way the system is created, for e.g. in the reaction $$\ce{NH_4Cl(s) -> NH_3(g) + HCl(g)}$$ If the only source of products is the reaction, there is the special constraint $y_\ce{NH_3} = y_\ce{HCl}$ ($= 0.5$ if these are the sole components). This counts as a constraint in this case ($s = 1$). (This example appears in the reference)

In the reaction $$\ce{NH_4Br(s) <=> NH_3(g) + HBr(g)}$$

$\pi = 2$, $C = 3$ (note that all components are considered), $r = 1$. Given the constraint $y_\ce{NH_3} = y_\ce{HBr}$, $s = 1$. So $F = 2 - 2 + 3 - 1 - 1 = 1$.

In the system $\ce{KCl(aq)}$ plus a vapor phase, I guess you model either $\ce{K^+}$ and $\ce{Cl^-}$ as independent components, or you could consider $\ce{KCl(aq)}$ as a component in equilibrium with $\ce{KCl(s)}$.

1st case: $\pi = 3$, $C = 4$, $r = 1$ (dissolution reaction), $s = 1$ (constraint given by mole balance), $F = 2 - 3 + 4 - 1 - 1 = 1$.

2nd case: $\pi = 3$, $C = 2$ ($\ce{KCl}$ treated as a single component), $s = 0$ (there is no extra constraint on $x_{\ce{KCl}}$), $r$ = 0 (no dissociation considerd, $\ce{KCl(s) -> KCl(aq)}$ is modeled as phase change). $F = 2 - 3 + 2 - 0 - 0 = 1$.

Reference: Smith, Joseph Mauk. "Introduction to chemical engineering thermodynamics." (1950): 584. (Chapter 14 - Chemical-Reaction Equilibria)