I agree 100% but then I think FesZ's topics are intermediate to advanced and the market for anything beyond lighting LEDs with microcontrollers or using a BJT as a switch or common emitter is limited. Hard to find a tutorial using a FET as a switch, despite being vastly superior to a BJT for this purpose. Or really even beginner content for FETs at all.
You could also run a "DC sweep" analysis in LTspice instead of varying the gate voltage source with time - the advantage is that you immediately get the correct x-axis.
Hey man, always love your videos! I recommend them to all my EE friends. Ever planned on doing anything to do with control theory? Like PID controllers and stuff related to that? I think your natural style of video REALLY helps me. Only regret is that it took me so long to find you haha - would have helped a lot in college!
Again, it's possible to improve on the dc stability against variation of threshold voltage and temperature by using a combination of feedback biasing and source resistor. You can do this by allowing more voltage drop across the source resistor and biasing the gate above ground by a resistor divider from drain to gate and gate to ground. For example, set a drain current of 10mA and allow 4V across the source resistor. So Rs=400R. Then split the remaining 6V equally between the transistor and the drain resistor. So each drops 3V and Rd=300R. The target quiescent bias point is 7V. Since the source is at 4V, the gate will need to be at 4V - 1.8V = 2.2V. Using a current of 0.1mA through the bias resistors gives Rg (from gate to ground) = 22K. Because the drain is at 7V, the feedback resistor Rf (from drain to gate) = (7V - 2.2V) /0.01mA = 48K. Running the simulation with the same extreme models gives a change of quiescent drain voltage from 7.7V to 6.0V as Vth changes from -2 to -6.5V. The change over temperature is 0.1V across the temperature range -20°C to 80°C. The corresponding drain current change is from 7.5mA to 13mA and 0.3mA across temperature.
In engineering school we were taught that in a FET, the linear operating mode is called the saturation region. The confusion comes from how the term is used with a BJT, where saturation is a nonlinear region of operation….can’t get more collector current with more base current. It’s really an unfortunate use of terminology, confusing many people! 🤦♂️
That's exactly right. The concept of "saturation" mode is normally used in the context of a BJT where the collector current is limited by the collector resistance instead of the (excess) base current, and the resulting collector-emitter voltage is small. In the case of an FET, the ohmic region occurs under analogous conditions: the drain current is limited by the drain resistance instead of the (excess) gate-source voltage, and the resulting drain-source voltage is small. That seems to be sufficient justification to describe the FET as "in saturation" as the meaning should be familiar.
It's really difficult to create a current mirror with discrete JFETs because of the difficulty in matching Vth between samples. Usually we'd use a matched pair on the same die, which tends to be expensive. In such cases, we would simply use a JFET input opamp like the TL072 for cost reasons. Usually, active biasing for a common source amplifier stage would mean using an active load (current source) instead of the drain resistor. That allows a larger gain without having to use very high supply voltages.
Why did you use ".op 1" at 5:45 and in other simulations, what is the meaning of the "1" argument? I get the same results with just ".op" command without the extra argument.
I think that when I started working on the simulation it was a transient type of 1 second - ".tran 1"; but then I switched it to ".op" and LTspice just didn't delete the argument. The "1" was not added intentionally...
Dunno how to write it without being rude, but here it goes anyway: please stop smoking. I started watching your videos from the very first ones, and could not help but notice the detrimental effect. To be on topic, I really liked your project videos, from the every day problem until the finished project. The light sensor, microphone, current probe just to name a few. Do you not do any hobby projects anymore? I mean the theory videos are good and fine (and necessary! you are basically the only youtuber covering eletronics so deeply and methodically). But some lightweight videos would be a breeze of fresh air. Keep up the good work.
Hello Laszlo; To start, I do not smoke; I guess my video editing, lighting and youtube video compression make me look different in videos than in real life; but thank you for your concerns. As you pointed out, I did start working more on educational content rather than projects, on the one side, since these videos "do better" (as far as views go) and also its somewhat easier to make - you never know how a project turns out, and I do not want to make a video series about building something that goes nowhere. With the educational stuff, it is guarantied to work. I will try to integrate a "project" into some future content, just to somehow link various principles and theories to a practical usecase.
😃😄😁😆😅 5 Smile's. You have to love negative gain in a circuit like a 555 on a negative and if you put a director to it on the in supply it just will not turn on???? strange no😁you could try light to drain to a negative use one to balance a relay line from flicking due to signal on negative but unlike the relay flyback put that director in and it doesn't work!! again strange no!😁.
You should have more subscribers. You have some of the best electronics content on RUclips.
@G E T R E K T Good points. I’m an EE.
Agreed
I agree 100% but then I think FesZ's topics are intermediate to advanced and the market for anything beyond lighting LEDs with microcontrollers or using a BJT as a switch or common emitter is limited. Hard to find a tutorial using a FET as a switch, despite being vastly superior to a BJT for this purpose. Or really even beginner content for FETs at all.
Videos should be required watching for all 1st year EE students!
Here I find the most advanced topics discussed! Congrats!! Greetings from Brazil!
I was serching this Topic, and my favorite channel made videa about It!
You could also run a "DC sweep" analysis in LTspice instead of varying the gate voltage source with time - the advantage is that you immediately get the correct x-axis.
Excellent content as always, Fesz! 👌
Hey man, always love your videos! I recommend them to all my EE friends. Ever planned on doing anything to do with control theory? Like PID controllers and stuff related to that? I think your natural style of video REALLY helps me. Only regret is that it took me so long to find you haha - would have helped a lot in college!
Good videos! Can't wait for the next topic.
Very useful informations as usual.
Thank you 😊😊😊😊 very nice topics
Simply excellent, thanks
Again, it's possible to improve on the dc stability against variation of threshold voltage and temperature by using a combination of feedback biasing and source resistor. You can do this by allowing more voltage drop across the source resistor and biasing the gate above ground by a resistor divider from drain to gate and gate to ground.
For example, set a drain current of 10mA and allow 4V across the source resistor. So Rs=400R. Then split the remaining 6V equally between the transistor and the drain resistor. So each drops 3V and Rd=300R. The target quiescent bias point is 7V.
Since the source is at 4V, the gate will need to be at 4V - 1.8V = 2.2V. Using a current of 0.1mA through the bias resistors gives Rg (from gate to ground) = 22K. Because the drain is at 7V, the feedback resistor Rf (from drain to gate) = (7V - 2.2V) /0.01mA = 48K.
Running the simulation with the same extreme models gives a change of quiescent drain voltage from 7.7V to 6.0V as Vth changes from -2 to -6.5V. The change over temperature is 0.1V across the temperature range -20°C to 80°C. The corresponding drain current change is from 7.5mA to 13mA and 0.3mA across temperature.
Great video.
Hey, great video! Just one question. When do you say saturation mode is the Fet working in linear/ohmic region it isn't? Thanks!
In engineering school we were taught that in a FET, the linear operating mode is called the saturation region. The confusion comes from how the term is used with a BJT, where saturation is a nonlinear region of operation….can’t get more collector current with more base current. It’s really an unfortunate use of terminology, confusing many people! 🤦♂️
That's exactly right. The concept of "saturation" mode is normally used in the context of a BJT where the collector current is limited by the collector resistance instead of the (excess) base current, and the resulting collector-emitter voltage is small.
In the case of an FET, the ohmic region occurs under analogous conditions: the drain current is limited by the drain resistance instead of the (excess) gate-source voltage, and the resulting drain-source voltage is small. That seems to be sufficient justification to describe the FET as "in saturation" as the meaning should be familiar.
"Well that's not good." - FesZ
an active way......
so some current mirror ( i hope this is how it translates)?
or maybe some zener ?
It's really difficult to create a current mirror with discrete JFETs because of the difficulty in matching Vth between samples. Usually we'd use a matched pair on the same die, which tends to be expensive. In such cases, we would simply use a JFET input opamp like the TL072 for cost reasons.
Usually, active biasing for a common source amplifier stage would mean using an active load (current source) instead of the drain resistor. That allows a larger gain without having to use very high supply voltages.
@@RexxSchneider i know this is why i mentioned the zener since for this is an ok ish reference
Why did you use ".op 1" at 5:45 and in other simulations, what is the meaning of the "1" argument?
I get the same results with just ".op" command without the extra argument.
I think that when I started working on the simulation it was a transient type of 1 second - ".tran 1"; but then I switched it to ".op" and LTspice just didn't delete the argument. The "1" was not added intentionally...
Dunno how to write it without being rude, but here it goes anyway: please stop smoking. I started watching your videos from the very first ones, and could not help but notice the detrimental effect. To be on topic, I really liked your project videos, from the every day problem until the finished project. The light sensor, microphone, current probe just to name a few. Do you not do any hobby projects anymore? I mean the theory videos are good and fine (and necessary! you are basically the only youtuber covering eletronics so deeply and methodically). But some lightweight videos would be a breeze of fresh air. Keep up the good work.
Hello Laszlo; To start, I do not smoke; I guess my video editing, lighting and youtube video compression make me look different in videos than in real life; but thank you for your concerns. As you pointed out, I did start working more on educational content rather than projects, on the one side, since these videos "do better" (as far as views go) and also its somewhat easier to make - you never know how a project turns out, and I do not want to make a video series about building something that goes nowhere. With the educational stuff, it is guarantied to work. I will try to integrate a "project" into some future content, just to somehow link various principles and theories to a practical usecase.
at just before 11:00 the text was white on a white paper background, maybe not the best choice
Indeed.. I will try to take more care in the color selection...
Man, you have to take some sleep... Take care!
😃😄😁😆😅 5 Smile's. You have to love negative gain in a circuit like a 555 on a negative and if you put a director to it on the in supply it just will not turn on???? strange no😁you could try light to drain to a negative use one to balance a relay line from flicking due to signal on negative but unlike the relay flyback put that director in and it doesn't work!! again strange no!😁.