How is diode measurement accuracy determined?

[giosif]

Hi,

Some days ago, I was measuring the voltage drop across a diode and noticed that different multimeters I have would give quite different results.
This got me wondering and the exaplantion I could think of for the differences in measurements is multi-factor:

• Multimeter accuracy for diode measurement
• Multimeter current being generated in diode measurement mode
• Diode specs with regards to voltage drop vs. current flowing through it

But then, looking at the datasheet for a diode and seeing that voltage drop vs. current is not a linear curve, I am asking myself how a multimeter manufactuter can determine the measurement accuracy for one of their multimeters, since the current that multimeter generates is not a "standard" value ("standard" = a value agreed upon in the industry)?
In other words, how do they now the precise value of the voltage drop for the diode under test @ the current value their multimeter is designed to generate?

Thanks!

[Rerouter]

If you look at the datasheet they will generally give a typical voltage at a given current, most multi-meters use a constant current source, something between 10mA and 100uA to drive the diode, however these are not intended to tell you everything about the diode, more that the part is working as intended,

Your main deviation would be the differences in test current, to find this out, put one meter on diode test, and another on mA, and cross connect the 2,

[Audioguru]

Why do you need the exact forward voltage of a diode? The voltage is a little different even if different diodes have the same part number. Temperature and current also change the voltage.

[munkeyman1985]

I can't say why your meter had different results. How it works is simple. Look at the data sheet again. The Y axis should be current the X axis should be voltage. Please note, after the steep curve the voltage dose not change much with current. The meter simply a sufficient current to get past that curve, and display that voltage. It should be fairly close from one meter to the next about 0.7 for silicon and 0.25 for geranium in foreword bias, once the current is past that curve your see. Even if it is just pasting with a 1ma or way past it with 10ma, the voltage drop will be about the same. This voltage drop is called the forward voltage drop. Hope this is what you were looking for.

[TimFox]

When sorting diodes into matched pairs or quads using a digital multimeter in ohms mode, you will note that the reading (actually volts at the meter-determined current) will change with time as the diode heats up with the dissipated power.  Therefore, you must be careful to record the value at a fixed time after applying the leads to get consistent answers.
As mentioned above, the actual voltage drop across a physical diode is a function of the current through the diode and the temperature of the semiconductor junction.  Furthermore, glass-packaged diodes may exhibit photocurrent that will affect the reading, so a simple dark shield is advisable.

[BMack]

Why do you need the exact forward voltage of a diode? The voltage is a little different even if different diodes have the same part number. Temperature and current also change the voltage.

Same thing I was wondering.