0
$\begingroup$

So I took the time to measure the current dependency on voltage of a diode I have. I applied an exponential fit to it, and have a pretty reliable equation (within 1%).

I'm interested in how a diode-resistor-capacitor series circuit response to different signals. Naturally, I'm starting with just DC voltage.

The equation that I have for the voltage/current dependency for the diode is of the form

$$ I=ae^{bV_D} \tag{1}$$

where $V_D$ is the voltage across the diode.

Using Kirchhoff's law, I get the following differential equation with an initial condition:

$$V = RQ' + \frac{1}c Q + \frac{1}b \ln\left(\frac{Q'}a\right)$$

$$Q(0)=0$$

where $R$ is the resistance of the resistor, $c$ is the capacitance of the capacitor, $a$ and $b$ are the exponential regression constants from equation (1), and $V$ is the applied DC voltage.

Does anyone know if it's possible to analytically solve this equation?

$\endgroup$
5
  • 1
    $\begingroup$ If you put a capacitor in series with anything, than at DC the current will be 0. $\endgroup$
    – The Photon
    Commented Jul 11, 2017 at 3:56
  • $\begingroup$ After enough time, yes that's correct. But not initially. I'm interested in modeling what happens in that first second. Presumably the current follows a type of exponential decay. $\endgroup$
    – Ben
    Commented Jul 11, 2017 at 4:08
  • $\begingroup$ When we talk about analyzing a circuit at DC, we usually mean DC steady state. If you want to talk about what happens when the input voltage is changed, we usually call that a transient analysis. $\endgroup$
    – The Photon
    Commented Jul 11, 2017 at 4:48
  • $\begingroup$ AFAIK, there is no analytical solution to this circuit. However there are dozens of different computer programs available that can provide a numerical solution as accurate as could possibly be useful. LTSpice is a well known free (as in beer) one. $\endgroup$
    – The Photon
    Commented Jul 11, 2017 at 4:49
  • $\begingroup$ Any non-linear term is usually a pain in the neck. Diodes are usually linealized because of that. $\endgroup$
    – FGSUZ
    Commented Apr 8, 2019 at 22:06

1 Answer 1

0
$\begingroup$

Well, we know that:

$$\text{V}_\text{in}\left(t\right)=\text{V}_\text{D}\left(t\right)+\text{V}_\text{R}\left(t\right)+\text{V}_\text{C}\left(t\right)\tag1$$

And we also know that:

  • $$\text{I}_\text{D}\left(t\right)=\text{I}_\text{S}\cdot\left(\exp\left(\frac{\epsilon\cdot\text{V}_\text{D}\left(t\right)}{\eta\cdot\text{k}\cdot\text{T}}\right)-1\right)\tag2$$

Where $\text{I}_\text{S}$ is the reverse saturation current, $\epsilon$ is the electronic charge, $\text{k}$ is the Boltzmann's constant and $\text{T}$ is the absolute temperature and $1\le\eta\le2$.

  • $$\text{V}_\text{R}\left(t\right)=\text{I}_\text{R}\left(t\right)\cdot\text{R}\tag3$$
  • $$\text{I}_\text{C}\left(t\right)=\text{V}_\text{C}'\left(t\right)\cdot\text{C}\tag4$$
  • $$\text{I}_\text{in}\left(t\right)=\text{I}_\text{D}\left(t\right)=\text{I}_\text{R}\left(t\right)=\text{I}_\text{C}\left(t\right)\tag5$$

So, we get:

$$\text{V}_\text{in}'\left(t\right)=\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}\cdot\frac{\text{I}_\text{in}'\left(t\right)}{\text{I}_\text{S}+\text{I}_\text{in}\left(t\right)}+\text{I}_\text{in}'\left(t\right)\cdot\text{R}+\text{I}_\text{in}\left(t\right)\cdot\frac{1}{\text{C}}\tag6$$


For example, when the input voltage is constant we get:

$$\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}\cdot\frac{\text{I}_\text{in}'\left(t\right)}{\text{I}_\text{S}+\text{I}_\text{in}\left(t\right)}+\text{I}_\text{in}'\left(t\right)\cdot\text{R}+\text{I}_\text{in}\left(t\right)\cdot\frac{1}{\text{C}}=0\space\Longleftrightarrow$$ $$\int\text{I}_\text{in}'\left(t\right)\cdot\frac{\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}\cdot\frac{1}{\text{I}_\text{S}+\text{I}_\text{in}\left(t\right)}+\text{R}}{\text{I}_\text{in}\left(t\right)\cdot\frac{1}{\text{C}}}\space\text{d}t=\int-1\space\text{d}t\tag6$$

Substitute $\text{u}:=\text{I}_\text{in}\left(t\right)$:

$$\int\frac{\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}\cdot\frac{1}{\text{I}_\text{S}+\text{u}}+\text{R}}{\text{u}\cdot\frac{1}{\text{C}}}\space\text{d}\text{u}=\text{C}\cdot\left\{\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}\int\frac{1}{\text{I}_\text{S}+\text{u}}\cdot\frac{1}{\text{u}}\space\text{d}\text{u}+\text{R}\int\frac{1}{\text{u}}\space\text{d}\text{u}\right\}=\mathcal{C}-t\tag7$$

So, we get:

$$\text{C}\cdot\left\{\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}\cdot\frac{1}{\text{I}_\text{S}}\cdot\ln\left|\frac{\text{I}_\text{in}\left(t\right)}{\text{I}_\text{in}\left(t\right)+\text{I}_\text{S}}\right|+\text{R}\cdot\ln\left|\text{I}_\text{in}\left(t\right)\right|\right\}=\mathcal{C}-t\tag8$$

Now, we can write:

$$0.02353823794935365<\frac{\eta\cdot\text{k}\cdot\text{T}}{\epsilon}<0.05569380628470534\tag9$$

When $0 ^\circ\text{C}=\frac{5463}{20}\space\text{K}\le\text{T}\le\frac{5463}{20}\space\text{K}=50 ^\circ\text{C}$

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.