Need help getting started on this problem

[TecHNooB]

http://cobweb.ecn.purdue.edu/~ee255/HW_Fall_08/Hwk5.pdf

Problem #1

Do I have a degree of freedom for choosing V_GS or is there an implied constraint given by the equations?

 

[blustori]

I hope you never TA a class dealing with MOSFETS. Anyways, to get started on problem 1, draw a small signal circuit to get a KCL equation. If you don't know what this is, I suggest googling somewhere like MIT or Berkeley course sites. Good luck. It looks like a pretty long assignment. Then again, it is grad school.

 

[TecHNooB]

Originally posted by: blustori
I hope you never TA a class dealing with MOSFETS. Anyways, to get started on problem 1, draw a small signal circuit to get a KCL equation. If you don't know what this is, I suggest googling somewhere like MIT or Berkeley course sites. Good luck. It looks like a pretty long assignment. Then again, it is grad school.

I feel inclined to ask: how old are you and what do you do for a living?

 

[us3rnotfound]

I gotta get started on mine, too

 

[blustori]

I'm an undergrad. Looking at the first problem again, the MOSFET is saturated since its tied to 16V. Use the saturation equation for current. You should be able to pick a value for the source resistance and go from there. BTW, I didn't mean to sound like a jerk.

 

[0]

Originally posted by: TecHNooB
http://cobweb.ecn.purdue.edu/~ee255/HW_Fall_08/Hwk5.pdf

Problem #1

Do I have a degree of freedom for choosing V_GS or is there an implied constraint given by the equations?

You need to plug in the extremes for Id in both equations. That'll give you the extremes of Vgs, which then will allow you to solve for the resistors. Remember to plug in the Id for the Vgs that you calculated to get the ranges for the resistors.

 

[0]

Originally posted by: blustori
I'm an undergrad. Looking at the first problem again, the MOSFET is saturated since its tied to 16V. Use the saturation equation for current. You should be able to pick a value for the source resistance and go from there. BTW, I didn't mean to sound like a jerk.

The mosfet is certainly NOT saturated. Id = Idss at saturation, and that is not the case here.

 

[RallyMaster]

Standard Purdue snowball question. You in Furgason or Krogmeier?

 

[TecHNooB]

Originally posted by: RallyMaster
Standard Purdue snowball question. You in Furgason or Krogmeier?

Krogmeier He's entertaining and he does not like how Furgason teaches the course (I kind of agree with him).

 

[RallyMaster]

Same teacher here, BTW. You on campus or are you somewhere else?

 

[TecHNooB]

Originally posted by: RallyMaster
Same teacher here, BTW. You on campus or are you somewhere else?

Campus T_T

Originally posted by: 0
The mosfet is certainly NOT saturated. Id = Idss at saturation, and that is not the case here.

I think it is saturated. Id = Idss at the pinch off point (for depletion). Since V_dd is so high, the operating point should be beyond pinchoff which is still in saturation but Id != Idss.

 

[RallyMaster]

Campus must be like a ghost town right now...

I'm going to go work on it now and I'll get back to this thread.

 

[TecHNooB]

Originally posted by: RallyMaster
Campus must be like a ghost town right now...

I'm going to go work on it now and I'll get back to this thread.

Kind of, but at least it's peaceful Thanks for your help!

 

[RallyMaster]

From what I"m understanding so far, the Id value should just be in between 4 and 6 mA, so I picked 5 mA for Id and set it to both the first and second Id equations. Solve for Vgs values for both and pick the one that looks seems the most likely. I do believe that in the 20(1+Vgs/4)^2 equation, 4 is the Vthreshold value. Same with 1.44 being the threshold value for the other equation. Since the transistor is a depletion mode NMOS, Vtn < 0. In order for conduction to occur, Vgs has to be less than 0 but greater than Vtn. In my case, Vgs is either -2V or -0.6067V. Check my answer please.

 

[TecHNooB]

Originally posted by: RallyMaster
From what I"m understanding so far, the Id value should just be in between 4 and 6 mA, so I picked 5 mA for Id and set it to both the first and second Id equations. Solve for Vgs values for both and pick the one that looks seems the most likely. I do believe that in the 20(1+Vgs/4)^2 equation, 4 is the Vthreshold value. Same with 1.44 being the threshold value for the other equation. Since the transistor is a depletion mode NMOS, Vtn < 0. In order for conduction to occur, Vgs has to be less than 0 but greater than Vtn. In my case, Vgs is either -2V or -0.6067V. Check my answer please.

I dont know where you got -0.6067V

So far what I've taken away from this problem is that.. given a range of Vp (-4 to -1.44), you should be able to choose values for Vgs that keep Id within 4 to 6 ma. I've found a range for Vgs at both 4ma and 6ma for Vp = -4 and Vp = -1.44 but there's no overlap. Why no overlap

 

[RallyMaster]

Where I got -.6067V:
5 mA = 7.2(1+Vgs/1.44)^2

-sqrt(5/7.2)-1.44 = -2.2733V invalid because it's more negative than Vtn
sqrt(5/7.2)-1.44 = -.6067V valid

So after that...I'm assuming we take the KVL of the loop starting from Vdd to Rss...
that leaves us with:
-Vdd +Id*Rd + Vds + Is*Rss = 0

Do we assume that Is = Id and that Rd can be any value?

 

[TecHNooB]

Originally posted by: RallyMaster
Where I got -.6067V:
5 mA = 7.2(1+Vgs/1.44)^2

-sqrt(5/7.2)-1.44 = -2.2733V invalid because it's more negative than Vtn
sqrt(5/7.2)-1.44 = -.6067V valid

So after that...I'm assuming we take the KVL of the loop starting from Vdd to Rss...
that leaves us with:
-Vdd +Id*Rd + Vds + Is*Rss = 0

Do we assume that Is = Id and that Rd can be any value?

Is = Id iff Ig = 0 (which we are assuming it is). I've switched gears to ee202 for now

Btw, your Vgs should be -2.64 or -0.24. From your equations, I'm assuming you're using a TI-83? No solve function for you

[ sqrt(5/7.2) - 1 ] * 1.44 is what you should have.

 

[RallyMaster]

Oh oops. Did that in MATLAB...Thanks for catching that.