Let's say I am given the following circuit: along with $$dV_p/dT=0.139mV/K$$:
Defining V1 the voltage above Diode D1 and V2 the voltage above diode D2 , we use the expression for the current of a diode for both diodes :
$$I_{D1}=I_{S1}e^{\frac{V1}{V_T}}$$ $$I_{D2}=I_{S2}e^{\frac{V2}{V_T}}$$
If we solve with respect to the voltages, we get :
$$V_1=V_T\ln{\frac{I_{D1}}{I_{S1}}}$$ $$V_2=V_T\ln{\frac{I_{D2}}{I_{S2}}}$$
Subtracting, we finally get Vp :
$$Vp=V_T\ln{\frac{I_{D1}I_{S2}}{I_{D2}I_{S1}}}$$
What I know for sure, is that the thermal voltage VT depends on temperature and it is given by :
$$V_T=\frac{kT}{q}$$
I think that the saturation currents depend on the temperature as well. However, will these currents be different? Or can I just cancel them out in the logarithm above?
Cancelling them out, the result doesn't make sense:
$$\frac{dV_p}{dT}=\frac{k}{q}\ln{\frac{I_{D1}}{I_{D2}}}$$
The result is different form the given value of 0.139 mV/K and does not depend on T so I get no information from here.
So, it seems I can't cancel out the saturation currents .
After looking it up, saturation currents seem to depend on :
$$n_i^2=(5.2\times10^{15}T^{\frac{3}{2}}e^{\frac{-E_g}{2kT}})^2$$
However, even if I use the long expression for the saturation currents, ni1 and ni2 will still cancel out in the logarithm. Therefore the dependence of Vp on T vanishes even if the saturation currents depend on it.
Long story short, this was my attempt at a solution. How do I find the temperature?