Discriminating Between MOSFET Drain & Source Terminals

[muh_j18]

If the MOSFET is a symmetrical device, i.e the current can flow either from the drain to the source, or vice versa, depending on wheather you apply the voltage VDD to the drain or to the sourse, then, what is the point of "discrimination" between these two terminals? why not just mark the gate of the transistor, as these two nodes can be connected either way?

 

[╔(σ_σ)╝]

From my basic understanding of MOSFET the voltage you apply to the drain or source depends on if the semiconductor is n or p channel and what characteristics you want.

I think I may understand where you are coming from.

But I think the "discrimination" is partly due to convention also.

If we have an n-channel device we would want to apply a positive voltage to the drain terminal so that current ( conventional current, which means positive charges moving) flows from a higer potential to a lower potential. In this case from the drain to the source.

Now if we had a p-channel device the situation would be reversed.

I hope that helps in some way.

 

[ranger]

In principle, you can interchange the two terminals of depletion MOSFETs. However, if the gate is offset from the drain to reduce gate-drain capacitance, then the terminals cannot be interchanged. Additionally, in deep sub-micron processes, there is asymmetry between Rs and Rd arising from layout and process variations. Therefore, any interchanging will alter the drain current considerably.

 

[cabraham]

With an n-channel device, biasing the gate positive wrt source will create a conducting channel from drain to source. This channel will conduct in either direction. The characteristics when conducting drain to source are slightly different than that for source to drain conduction. In synchronous rectification, the FET conducts source to drain. Otherwise, the normal mode is drain to source.

Claude

 

[Adjuster]

Most discrete MOSFETS are made in a way that produces an inherent diode (called the body diode) connecting source and drain. In normal operation the diode is reverse-biased and so does not conduct. Reversing the drain-source voltage will turn this diode on.

 

[muh_j18]

From my basic understanding of MOSFET the voltage you apply to the drain or source depends on if the semiconductor is n or p channel and what characteristics you want.

I think I may understand where you are coming from.

But I think the "discrimination" is partly due to convention also.

If we have an n-channel device we would want to apply a positive voltage to the drain terminal so that current ( conventional current, which means positive charges moving) flows from a higer potential to a lower potential. In this case from the drain to the source.

Now if we had a p-channel device the situation would be reversed.

I hope that helps in some way.

what do you mean by "you may understand where i am coming from" ?

what I'm trying to say is : in the depleation MOSFET, a current path exist between the drain and the source even if VGS=0, so, if we short the gate with the source, the current can flow in both directions, am i wrong?

what made me confused about the MOSFETs is the CMOS gates, if you have studed this area, you will understand my question, in whih a transisor can drive the current in two pathes, each is the opposite of the other.

 

[cmos]

I believe ranger (post #3) gave the best reply in that he/she actually answer your question. To reiterate: yes, you are correct, in the most basic situation, source and drain are interchangeable. The reason to make a discrimination is so that you may intelligibly describe the physical operation of the MOSFET and also to keep you honest when designing a full circuit.

In modern ULSI, the MOSFET structure may be designed to contain some sort of asymmetry. This is done by the device engineer to enhance FET performance. In this case, it is extremely important to discriminate between source and drain.

 

[Mike_In_Plano]

The MOSFET is intrinsically a four terminal device, with the substrate acting as the fourth. Some, not many, transistors are available with the substrate brought out as a separate connection.

Two things are important about the substrate connection :
1 - The difference and polarity in potential between the gate and substrate is the means of control
2 - There are two diode junctions to the substrate. One from the Source, and one from the Drain

Thus, for an N-channel MOSFET, the substrate must always be at or be below the Source and Drain voltages. Given that, you can operate bilaterally. That is, with the Source or Drain acting to block voltage or source current.

In most discrete devices, the substrate is tied to the source. Thus the diode structure between the Drain and substrate is now between the Drain and Source.

- Mike

 

[muh_j18]

ok guys, thanks for all your replies, i think now i got the idea :P