I've been studying some vlsi lately and came across instances where the author mentioned that it is easier to use MOSFETs at the micron level than it is to use a resistor. Therefore at many instances(such as load inverters) MOSFETs are used in place of resistors.

MOSFETs are complicated silicon devices with various n and p regions whose operation is much more complex than resistors. Reducing their size to micrometers/nano would make the task of creating a MOSFET very complicated and the operation difficult to sustain given the small size of the various regions in a MOSFET. Consdering all this how is it easier to fabricate and use MOSFETs than resistors in IC chips ?

  • \$\begingroup\$ You state a number of facts, reason to a conclusion, and lo and behold, it does not match what experts say. If you don't want to dismiss the expert, either your facts or your reasoning must be wrong. For a start, how is "whose operation is much more complex" relevant? And why would (without further proof) "operation difficult to sustain given the small size" hold more strongly for a FET than for a resistor. \$\endgroup\$ – Wouter van Ooijen Nov 13 '13 at 8:20
  • \$\begingroup\$ @WoutervanOoijen I know I am going wrong somewhere. I don't know where so I asked a question \$\endgroup\$ – Vineet Kaushik Nov 13 '13 at 8:22
  • \$\begingroup\$ My point was that there are some things in your argument that you can spot as wrong, without any knowledge of the subject. \$\endgroup\$ – Wouter van Ooijen Nov 13 '13 at 9:41

The resistance of a resistor is proportional to it's length, and resistivity, and inversely proportional to it's width.

The smallest (cheapest) way to make, on silicon, a narrow isolated resistor with high resistivity, is to control the resistivity by controlling the number of electrons/holes, and to pinch the width of the resistor by applying an electric field to the region.

When you make a structure like that, it's called a "MOSFET". The Metal is a conductor at the edge of the resistor, the Oxide is the isolating region around the resistor, the Silicon is the controled resistive region, and the Field Effect controls the size of the resistor.

If you are making a structure that already includes some MOSFETs, then you've already got the material to make more MOSFETs. Adding a different kind of resistor would not only be larger, it would add more process steps, be more expensive, and probably mean you couldn't make your other MOSFETs as small and cheap.

If you don't use a Field to control the size of your resistor, you will have to make a much bigger resistor.

  • \$\begingroup\$ Not only is deriving a resistor from the same monolith more economical, adding a real resistor in place of a MOSFET in amplifier configurations (for biasing) has the overhead of using very large resistance values for R(drain) and R(source) in order to have a very large input and output impedance for the configuration. And for large passive resistors, in order to maintain proper bias in transistors for proper operation, we'll need very large power rails. In fact, it is due to MOSFETs that VLSI has been able to reduce the size of electronics to such levels :-) \$\endgroup\$ – Anshul Nov 13 '13 at 10:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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