# I have used a diode having a knee voltage of around 0.72V. But in simulation, in a graph, I m not getting that value. is that because of the resistor

#### valorous

Joined Oct 28, 2021
18

#### boostbuck

Joined Oct 5, 2017
260
Yes - the current in your sim is very small. The datasheet shows a 0.7 knee at If = 10mA, which in your circuit would be provided by a 2k resistor.

#### valorous

Joined Oct 28, 2021
18
Yes - the current in your sim is very small. The datasheet shows a 0.7 knee at If = 10mA, which in your circuit would be provided by a 2k resistor.
thanx bro. Is datasheet given in Ltspice??
and also when i tried 2k resistor i m not getting correct result, so i used resistor of around 0.1k , and so i m getting correct value now.

#### valorous

Joined Oct 28, 2021
18
thanx bro. Is datasheet given in Ltspice??
and also when i tried 2k resistor i m not getting correct result, so i used resistor of around 0.1k , and so i m getting correct value now.
and also is it necessary to show the knee voltage = 0.72V.
i mean, in my case, i m getting a diff value(in graph , where R=100k) for forward characteristics of diode , is that will be okay or not??

#### Papabravo

Joined Feb 24, 2006
19,576
You should not expect a simulation to match a datasheet exactly. The datasheet numbers should be treated as if they were normally distributed random variables with a mean and a variance. A datasheet value listed as a MINIMUM or a MAXIMUM should be interpreted as being 3 standard deviations away (+ or -) from the mean. Any particular part that you buy from a supplier has a 99.5% chance of being within ±3σ of the mean. A simulation model should also have the same 99.5% probability of being within ±3σ of the mean also. If you get really clever and decide to tweak the model(s) for your purposes like Alex ( @Bordodynov ) does from time to time you can adjust your result to your preference.

Last edited:

#### valorous

Joined Oct 28, 2021
18
is this count as a right or wrong, ??
and so, it is not necessary to go along with datasheet, right?

#### MrChips

Joined Oct 2, 2009
27,679
I don’t know if there is an exact definition of knee voltage.
Plot the current on a logarithmic scale and let’s see what you get.
You need to take the diode current beyond 10mA.

#### Papabravo

Joined Feb 24, 2006
19,576
is this count as a right or wrong, ??
and so, it is not necessary to go along with datasheet, right?
It counts as incomplete. You need to explore a much wider range of values. In particular you must extend the potential current to 10 mA and beyond. Like this:
At about 0.72V forward bias the current is in excess of 16.36 mA. At ≈12 mW of power dissipation it won't even get warm.

Last edited:

#### valorous

Joined Oct 28, 2021
18
so when i am using resistor in ckt, i have to set the values such that it
It counts as incomplete. You need to explore a much wider range of values. In particular you must extend the potential current to 10 mA and beyond. Like this:
At about 0.72V forward bias the current is in excess of 16.36 mA. At ≈12 mW of power dissipation it won't even get warm.

View attachment 251408

View attachment 251409
so when i am using resistor in ckt, i have to adjust resistor value such that it explores wide range of values of current or to get the desired knee voltage(0.7V) , right?

#### Papabravo

Joined Feb 24, 2006
19,576
so when i am using resistor in ckt, i have to set the values such that it

so when i am using resistor in ckt, i have to adjust resistor value such that it explores wide range of values of current or to get the desired knee voltage(0.7V) , right?
That is not usually what you do. Normally you decide on what current you want in the diode and let the forward voltage be whatever it needs to be. In a career spanning half a century I've never seen somebody try to design a diode to operate at a particular forward voltage. Since every diode is different, this is a ridiculous goal, which cannot be achieved in practice. The point on the IV curve where the diode conducts a substantial current, like 10 mA, is not called the KNEE. It is just called the forward voltage. A KNEE occurs in a REVERSE BIASED Zener diode.

#### crutschow

Joined Mar 14, 2008
31,116
Since the diode current is a logarithm function of voltage, the apparent "knee", using linear scales, depends upon the current range of the test.
The simulation below, with a logarithmic horizontal current scale, gives a straight line, showing this log relation between voltage and current.
The slight curve at above about 10mA is due to the intrinsic ohmic resistance of the diode.

#### Papabravo

Joined Feb 24, 2006
19,576
You can get current as a function of voltage or voltage as a function of current and both representations should give you the same set of points. Straight lines on a semi-log grid imply an exponential relationship.