How to increase the drive strength of a Amplifier model

Thread Starter

achen

Joined Sep 11, 2022
18
Below is the entire circuit, the Op Amp is generating a voltage to Vsink, which then connects to a gate of a NMOSFET.
When the Tox of the NMOSFET is 1e-7, the capacitance is small, so Vsink voltage profile is expected. (-15V->15V)
However for design reason, I need to have the Tox as 1e-9, but then the Op Amp is not strong enough to drive the Vsink because of the bigger oxide capacitance at the NMOSFET.
How should I modify the Op Amp (changing the parameters) to give expected Vsink at Tox=1e-9
View attachment 277924
View attachment 277925
View attachment 277926
 

Attachments

Last edited:
Could you please supply more information? There could be more than one reason for your problem.

I am not familiar with a Keysight AMP 155 and cannot find it on line. Can you supply a data sheet or a link to info on it? Is it an op amp, like an IC you put on a circuit board, or something in an instrumentation system? My search only brings up a 106 page manual for a "Keysight N5990A-155 Display Port".

Could you please give us the whole circuit - or at least the portion of the circuit that includes the op amp and the NMOS (FET). Is the op amp in a feedback circuit? What is the part number of the NMOS? Where does the input to the op amp come from?

If you say the problem is capacitance, it sound like you have a transient problem. What do you want your circuit to do? Switch on and off, or operate in some kind of linear fashion? Are you trying to make a fast transition and it reacts too slowly? Or, is it oscillating? Sometimes capacitive loads make op amps oscillate, especially in feedback circuits. Or, does it just never achieve whatever you want it to do?
 

Papabravo

Joined Feb 24, 2006
21,159
You might have a hard time finding an answer since most of us cannot afford a tool with such a hefty license fee. There is scant motivation for the expenditure of time and effort to support a company that keeps on taking money from your pocket. There is no evidence that anybody on this board can answer questions about Keysight ADS. Doesn't the company sponsor a user forum?
 
Last edited:

Thread Starter

achen

Joined Sep 11, 2022
18
Could you please supply more information? There could be more than one reason for your problem.

I am not familiar with a Keysight AMP 155 and cannot find it on line. Can you supply a data sheet or a link to info on it? Is it an op amp, like an IC you put on a circuit board, or something in an instrumentation system? My search only brings up a 106 page manual for a "Keysight N5990A-155 Display Port".

Could you please give us the whole circuit - or at least the portion of the circuit that includes the op amp and the NMOS (FET). Is the op amp in a feedback circuit? What is the part number of the NMOS? Where does the input to the op amp come from?

If you say the problem is capacitance, it sound like you have a transient problem. What do you want your circuit to do? Switch on and off, or operate in some kind of linear fashion? Are you trying to make a fast transition and it reacts too slowly? Or, is it oscillating? Sometimes capacitive loads make op amps oscillate, especially in feedback circuits. Or, does it just never achieve whatever you want it to do?
Below is the entire circuit, the Op Amp is generating a voltage to Vsink, which then connects to a gate of a NMOSFET.
When the Tox of the NMOSFET is 1e-7, the capacitance is small, so Vsink voltage profile is expected. (-15V->15V)
However for design reason, I need to have the Tox as 1e-9, but then the Op Amp is not strong enough to drive the Vsink because of the bigger oxide capacitance at the NMOSFET.
How should I modify the Op Amp (changing the parameters) to give expected Vsink at Tox=1e-9
1665185214397.png
1665185615244.png
1665185644785.png
 
Last edited:
It seems pretty obvious that the problem is the amplifier's 100 ohm Rout (output resistance). From your graph, the decay time constant is around 30 nanoseconds. Since time constant = R x C, this implies an NMOS capacitance around 300 pf. Does that sound reasonable?

The solution would be to make the amplifier's Rout MUCH lower, probably just an ohm or so. I'm totally unfamiliar with the Keysight system and so don't know that can be modified. Maybe another forum member knows. If tradeoffs would allow Rout to be reduced, an obvious one is, you don't need anywhere near 100 dB gain.

Since I don't know the Keysight system I'll blab on a bit more, though it may or may not be helpful. It looks like you're using the op amp as a comparator, not a controlled gain stage. If you can't reduce its Rout, could you replace it with a comparator with a low Rout? Or - can you add discrete components (ICs) to the system? If so, you might try putting a fast-response IC buffer with low Rout between the op amp and the NMOS. I don't have an immediate suggestion, but there are a number of CMOS drivers available. If the idea sounds good I could look a bit to see if I could suggest one. It looks like the buffer would need to run from plus and minus 15V supply.

Hope this helps.
 

KeithWalker

Joined Jul 10, 2017
3,063
You might have a hard time finding an answer since most of us cannot afford a tool with such a hefty license fee. There is scant motivation for the expenditure of time and effort to support a company that keeps on taking money from your pocket. There is no evidence that anybody on this board can answer questions about Keysight ADS. Doesnt' the company sponsor a user forum?
Keysight Technologies is the Test and Measurement company that evolved from a split in Hewlett Packard between the Computer group and the Test and measurement group, via Agilent Technologies. I worked for them for 33 years as a measurement specialist and technical consultant.
The company has always offered the very best of post sales technical support but if you are trying to modify one of its standard products, I doubt whether they will offer you any help.
What exactly are you trying to achieve? If we knew more about your problem, maybe we could be more helpful.
 

Papabravo

Joined Feb 24, 2006
21,159
Keysight Technologies is the Test and Measurement company that evolved from a split in Hewlett Packard between the Computer group and the Test and measurement group, via Agilent Technologies. I worked for them for 33 years as a measurement specialist and technical consultant.
The company has always offered the very best of post sales technical support but if you are trying to modify one of its standard products, I doubt whether they will offer you any help.
What exactly are you trying to achieve? If we knew more about your problem, maybe we could be more helpful.
I am not the one with a problem. The TS, @achen has made a series of posts asking questions about the Keysight ADS simulator. AFAIK he has received no meaningful replies from anybody familiar with that particular simulator. Keysight apparently offers a free license to students, but they steadfastly REFUSE to say what the price of a non-student license is. That to me is the biggest of RED flags that the cost is an eye-popping number that only a corporation with very specific needs could afford.

I merely expressed my opinion that this forum is probably the wrong place to get help using the Keysight ADS simulator because ordinary hobbyists would have scant motivation to pay for and learn such a simulator. If you think any of my opinions or assumptions are wrong or even misguided, I can respect that.
 

KeithWalker

Joined Jul 10, 2017
3,063
I am not the one with a problem. The TS, @achen has made a series of posts asking questions about the Keysight ADS simulator. AFAIK he has received no meaningful replies from anybody familiar with that particular simulator. Keysight apparently offers a free license to students, but they steadfastly REFUSE to say what the price of a non-student license is. That to me is the biggest of RED flags that the cost is an eye-popping number that only a corporation with very specific needs could afford.

I merely expressed my opinion that this forum is probably the wrong place to get help using the Keysight ADS simulator because ordinary hobbyists would have scant motivation to pay for and learn such a simulator. If you think any of my opinions or assumptions are wrong or even misguided, I can respect that.
Papabravo, I was not directing any criticism towards your comments. I agree with what you posted. I was pointing out to the TS that Keysight does offer excellent technical help, but will not help to modify a standard product. I was not aware that he has made other posts about this. Obviously, from the small amount we have been told about this project, an interface is required between the amplifies and FET.
 

Papabravo

Joined Feb 24, 2006
21,159
Papabravo, I was not directing any criticism towards your comments. I agree with what you posted. I was pointing out to the TS that Keysight does offer excellent technical help, but will not help to modify a standard product. I was not aware that he has made other posts about this. Obviously, from the small amount we have been told about this project, an interface is required between the amplifies and FET.
OK - I guess the quoting of my post confused me. You can go to the username and examine all of the threads he has started. In almost all cases he asks a question about modifying the parameters of a blank model and nobody responds, I'm guessing, that is because nobody here is familiar with that simulator because we mostly use LTspice since it is still FREE!

I also have a pathologically bad attitude toward companies that have all kinds of shiny promotional pieces on the web without so much as a whiff of the pricing. If their products are so great, why the reluctance to advertise the price.

If Keysignt ADS was a free download, I might have been willing to download it and give it a shot. What else is a retiree supposed to do with his time.
 

KeithWalker

Joined Jul 10, 2017
3,063
Now I understand. The TS is talking about a circuit simulator not a hardware amplifier. He should have specified that to avoid confusion. If he does not understand how to drive a power MOSFET, that is what he should be asking about. He doesn't mention whether he is driving it linearly or into saturation so I can't offer much advice.
 

Papabravo

Joined Feb 24, 2006
21,159
Now I understand. The TS is talking about a circuit simulator not a hardware amplifier. He should have specified that to avoid confusion. If he does not understand how to drive a power MOSFET, that is what he should be asking about. He doesn't mention whether he is driving it linearly or into saturation so I can't offer much advice.
I don't know if I'm jumping the gun here, but this behavior suggests the possibility that the TS is a troll. He asks a question and then in some cases he fails to interact further with the answers.
 
Top