PMOS: the strong-1 switch
Closes on a gate 0, and pulls firmly to power
A PMOS transistor conducts when its gate is a 0, and it passes a strong (full-rail) 1 but only a weak 0 that stops a threshold voltage short of ground, which is why PMOS is always used to pull an output up toward the power supply.
Builds onNMOS: the strong-0 switch
The PMOS is the NMOS turned inside out. Everything the NMOS does well, the PMOS does poorly, and vice versa. Once you see it as the exact mirror, you never have to memorize it separately: whatever is true for the NMOS, flip
0 and 1, flip strong and weak, flip up and down.When it conducts
A PMOS turns on when its gate is a
0 and is off when its gate is a 1. That inversion is why its symbol wears a small bubble on the gate. Its source is tied to the high side (VCC), and it conducts when the gate is pulled a threshold *below* that source, which for us means a gate 0.Strong 1, weak 0
- **Passing a
1(pulling the output up toVCC).** The output rises towardVCC. The source sits atVCCand the gate is at0, so the gate stays far below the source and the transistor keeps conducting hard all the way up. The output reaches a clean, full-strength1. This is a strong 1. - **Passing a
0(trying to pull the output down toGND).** As the output falls, it drags the PMOS source down with it. But the gate is fixed at0, so the gate-to-source voltage shrinks as the source approaches the gate. Once the output falls to withinVtof the gate, the channel cuts off. The output stops at roughlyVtaboveGND, short of a full0. This is a degraded weak 0.
A PMOS delivers a strong 1 and a weak 0, the exact mirror of the NMOS. It is superb at pushing a wire up to the supply and mediocre at dragging it down. That is why PMOS is used only in the pull-up network (the path to
VCC) in every static gate. Together the two devices cover both jobs perfectly: NMOS pulls down with a strong 0, PMOS pulls up with a strong 1.| gate | input passed | output |
|---|---|---|
| 1 | 0 | Z |
| 1 | 1 | Z |
| 0 | 1 | strong 1 |
| 0 | 0 | weak 0 |
1: off, nothing passes (output floats to Z). Gate 0: it passes a full-strength 1 but only a threshold-lifted weak 0.Why do the two threshold drops point opposite ways? Because each transistor loses grip as its source approaches its gate voltage. The NMOS gate is high, so it loses grip near the *top* of the range (weak
1). The PMOS gate is low, so it loses grip near the *bottom* (weak 0). Each device keeps its grip on the rail it is tied to, which is precisely the rail we ask it to reach.The PMOS costs a little more silicon than the NMOS for the same drive strength. Holes (the charge carriers in a PMOS) move more slowly than electrons (in an NMOS), so a PMOS must be made physically wider to conduct as hard. Keep this in mind now; it returns as a real design concern in transistor sizing, and it is one reason the naturally NMOS-heavy gates (NAND) are often preferred over their PMOS-heavy mirrors.
Check yourself
You want to connect an output firmly to
VCC (a solid 1) when a control signal is 0. Do you use an NMOS or a PMOS, and why?Answer
Use a PMOS. It conducts on a gate
0 (matching your control) and delivers a strong 1, reaching full VCC. An NMOS would be wrong twice over: it conducts on a 1, not a 0, and even if you inverted the control it would only manage a weak 1. Pulling up is the PMOS's strong side.Frequently asked
When does a PMOS transistor conduct?
A PMOS conducts when its gate is a
0 (pulled a threshold below its source, which sits at VCC) and is off when its gate is a 1. The bubble on its gate symbol marks this inversion.Why does a PMOS pass a strong 1 but a weak 0?
When it pulls an output up to
VCC, its source stays high so the gate-to-source voltage stays large and it conducts fully, giving a strong 1. When it tries to pass a 0, the falling output drags its source down toward the gate until the channel cuts off about a threshold above ground, so the 0 is weak.What is the difference between NMOS and PMOS?
They are mirror images. An NMOS conducts on a gate
1 and delivers a strong 0 (good at pulling down); a PMOS conducts on a gate 0 and delivers a strong 1 (good at pulling up). CMOS uses both so the pull-down is NMOS and the pull-up is PMOS, each on its strong side.Why is a PMOS made wider than an NMOS?
Because a PMOS conducts using holes, which move more slowly than the electrons an NMOS uses. To match the NMOS's drive strength, the PMOS is made physically wider so it can pass the same current.
You now have both switches on their strong sides. Next: wiring several NMOS into a pull-down network that pulls an output to
0 for exactly the input conditions you choose.Every lesson here builds toward one thing: a working CPU, from the transistor up.
Open the free lab →