Max Current in CMOS Inverter with Given Parameters

  • Thread starter Thread starter perplexabot
  • Start date Start date
  • Tags Tags
    Cmos Current Max
AI Thread Summary
The discussion centers on calculating the maximum current in a CMOS inverter given specific parameters, including transistor characteristics and voltage levels. Participants express confusion about the equations and the operational modes of the NMOS and PMOS transistors, particularly regarding their behavior when the output is constrained to within 0.1V of ground or VDD. Key points include the need to clarify the input voltages and the conditions under which the NMOS operates in triode mode, as well as how to derive the gate voltages for both transistors based on the output conditions. The conversation emphasizes the importance of understanding the relationships between gate-source and drain-source voltages for accurate current calculations. Overall, the thread highlights the complexities of analyzing CMOS inverters and the necessity of clear communication regarding circuit parameters.
perplexabot
Gold Member
Messages
328
Reaction score
5
Hi all, I have gave this question a lot of thought but can't seem to get anywhere. Any help will be much appreciated.

Homework Statement


For a digital logic inverter for which k'n = 120 uA/V^2, k'p = 60 uA/V^2, Vtn = |Vtp| = .7V, VDD = 3V, Ln = Lp = .8 um, Wn = 1.2 um and Wp = 2.4 um, find:

the maximum current that the inverter can sink or source while the output remains within .1V of ground or VDD respectively.


Homework Equations


(attached as an image)


The Attempt at a Solution


No idea what to do. I can't even start to think about this one. I know that when Vout = Vdd/2, max current is achieved since (if Vdd is large enough) both mosfets are in saturation. Is that right? But I don't think this has anything to do with this question, since we are restriced to .1V from ground or VDD. I do not understand what is meant by the solution manual, please HELP.

I have also attached the answer in the solution manual, it makes no sense to me. Please help me out. This question is bothering me. Thank you
 

Attachments

  • eqns.png
    eqns.png
    10.6 KB · Views: 1,104
  • eqns1.jpg
    eqns1.jpg
    12.6 KB · Views: 731
Last edited:
Physics news on Phys.org
Attachments?

Anyway, you're right about this not being a linear mode of operation.
 
rude man said:
Attachments?

Anyway, you're right about this not being a linear mode of operation.

Oops. Forgot about the attachment. I have edited my post with the attachment. Thank you.
 
EDIT

never mind, I see it's supposed to be a CMOS output stage.
 
Last edited:
rude man said:
The problem is I don't know what the equations for modeling a MOS transistor you were given. Those symbols are not standard. You need to give us your model using those symbols, or at least write their definitions out in full.

In other words,
Ids = Ids(k,Vtn,L,W,Vgate-Vsource, Vdrain-Vsource, etc. ).

Also: is this a CMOS output or just an NMOS? If CMOS there are two answers to this question, one when the output is close to Vdd and one when it's close to ground. (NMOS and PMOS have different models).

Hey, thanks for your help. I have attached an image of the Id - Vgs relations that we use. I have also included two equations that greatly resemble the solution manual. Let me know what you think.
 
Good , but where are the image and equations?
 
rude man said:
Good , but where are the image and equations?

In my original post (first post). I now have two attachments. I should have mentioned that, sorry.
 
I still don't see a circuit diagram, but I'm going to assume the circuit applies VDD to the gate of the N channel device:

OK, so from what I gleaned from that,
for the N channel device,

Ids = k(W/L){(Vgs - VT)Vds - (1/2)Vds2}
where
Vgs = voltage from source to gate
Vds = voltage from source to drain.
Both voltages are > 0.

So now use that equation to solve for Ids which is the output current (sinking current) of your circuit when the output voltage is to be +0.1V. Hint: (1/2Vds2 << (Vgs - VT)Vds).

Then do the same for the P channel device using the p constants given you. Except now the output voltage is to be VDD - 0.1V. Watch your polarities for the P device!


BTW the equation for "I_peak" in your 1st image is totally irrelevant.
 
rude man said:
I still don't see a circuit diagram, but I'm going to assume the circuit applies VDD to the gate of the N channel device:

OK, so from what I gleaned from that,
for the N channel device,

Ids = k(W/L){(Vgs - VT)Vds - (1/2)Vds2}
where
Vgs = voltage from source to gate
Vds = voltage from source to drain.
Both voltages are > 0.

So now use that equation to solve for Ids which is the output current (sinking current) of your circuit when the output voltage is to be +0.1V. Hint: (1/2Vds2 << (Vgs - VT)Vds).

Then do the same for the P channel device using the p constants given you. Except now the output voltage is to be VDD - 0.1V. Watch your polarities for the P device!


BTW the equation for "I_peak" in your 1st image is totally irrelevant.

Thanks for your reply. I have 2 questions though. Before I ask my questions, I need to clarify that VDD is NOT applied to the NMOS gate. I have finally attached a schematic for reference.

First, how do u know that the NMOS is in triode mode?
Second, how do I know what Vin (AKA Vg for NMOS AKA Vg for PMOS) is for a given Vout. For example, how can I get Vg for the NMOS knowing that the output is .1V from ground? I know it will be around VDD since this is a CMOS inverter, but how can I get a precise answer?
 

Attachments

  • schematic.png
    schematic.png
    2.1 KB · Views: 707
  • #10
perplexabot said:
Thanks for your reply. I have 2 questions though. Before I ask my questions, I need to clarify that VDD is NOT applied to the NMOS gate. I have finally attached a schematic for reference.
Oh, but I think it is. I'm talking about what's labeled vl on your schematic diagram. That input comes from a similar circuit's output and will be very close to VDD when the previous stage output is high,

The reason I'm confident is that that way we get the right answer!

First, how do u know that the NMOS is in triode mode?
Because I picked the equation for the case where Vds < (Vgs - VT). You understand that? Why they call it 'triode' mode I have no idea. Probably dumb. A triode is a vacuum tube!

Second, how do I know what Vin (AKA Vg for NMOS AKA Vg for PMOS) is for a given Vout. For example, how can I get Vg for the NMOS knowing that the output is .1V from ground? I know it will be around VDD since this is a CMOS inverter, but how can I get a precise answer?

It's an inverter. The N device turns on and the P device turns off. So that means that if the input is +VDD, so is Vgs for the N device and the output tries to go to 0V. The reason the output goes only to 0.1V in your case is that there is a source of current like a load resistor tied to the output going to VDD that is sourcing current into the drain. Without a source of current the output will be very close to VDD. That current is what you're computing as your answer.

Similarly, when the input is close to 0V, the output tries to swing to +VDD via the P channel device turning on. For this device, Vsg = VDD also. The output won't quite make it to VDD, again if there is a current sink tied to the output, for example a load resistor going to ground. It's exactly analogous to the N channel device.

Try to get comfortable with the polarities involved with N vs. P channel devices:
N channel: gate voltage > source voltage to turn on. Vd > Vs. Vs is typically at ground.
P channel: gate voltage < source voltage to turn on. Vs > Vd. Vs is typically at VDD.

 
Back
Top