Hello,

I am trying to simulate an air core indutance with LTSpice. The simulation gives ideal values, and my pulsing waveform looks perfect no matter what frequency I used. I know in real life the waveform gets deformed and looks more like a sawtooth wave with a fraction of the voltage.
I found this link and many others like it:
https://www.allaboutcircuits.com/technical-articles/modeling-inductors-with-ltspice/

But nothing has worked so far, I still get ideal inductance. Has someone made a LTspice model with an good representation of a real inductor.

I put a screen shot, I just realized that the TanH(5*x) is what I had initially. I tried many values and also random values. So the value 222 in there is just one on the many values I tried.
I didnt realize it was that difficult to model a real life indutor.

Ken

 

hi cap,
Please post your LTS asc file.
E

Update:
Check out this video.

 

First you need to know the characteristics of your real life inductor before you can generate a model (inductance, resistance, stray capacitance, etc.) which you have not provided.
An air core inductor should be fairly close to the ideal.

 

Here is my .asc file.
I have measured the inductance to be 0.175mH

 

hi cap,
I have run your simulation, what is the application of the inductor
E

 

I know in real life the waveform gets deformed and looks more like a sawtooth wave with a fraction of the voltage.

Did you observe that, or is it just an assumption?

 

What part of an ideal inductor are you trying to understand?
There should be a self resonant frequency in a real inductor. Add 5pF (more or less) across the inductor.
Also there is some wire resistance. Maybe add 1 ohm
Right click on the inductor and add 1 ohm series resistance and 5pF parallel capacitance.

 

At any frequency less than 4.5MHz the 5k resistance is going to dominate. At you test frequency 250Hz you will hardly notice that there is any inductance.

 

Did you observe that, or is it just an assumption?

I did observe this, I can take a screen shot, but its exactly as I said.
The goal is to make a simulation very close to the real life so I can compensate electronically (with a circuit of some kind) and obtain the full pulse wave in the air core inductance at higher frequencies.

 

I can take a screen shot

Please do, with where you are measuring the voltages, and the sweep-speed and generator frequency.

 

Here are some snapshots I got, from 1khz to 76 Khz.

 

hi @captoro
This is what LTS shows for the AC analysis, looks OK to me.
Can you confirm the quality of the signal generator you are using as an input?
Say a screenshot of the Scope showing the Input to the MOS Gate.?
Also, the SG connected directly into your Scope, showing frequency response.
E

 

Here is a screen of the source at the base of the MOSFET. I am using an ISl55110 MOSFET DRIVER

 

hi @captoro
Thanks for the information, very puzzling.
Can you do just one more test.?
Remove the Inductor from the Drain, so only the 5K resistor is connected, run the project and check the Drain waveform.

E

 

hi cap,
It seems that N MOSFET has little history about its performance in LTSpice.
It maybe a poorly specified model.?
E
https://www.electronics-related.com/showthread/sci.electronics.design/772550-1.php

 

Here's the simulation with the 10μs pulse period you show in post #13:

 

hi @captoro
Thanks for the information, very puzzling.
Can you do just one more test.?
Remove the Inductor from the Drain, so only the 5K resistor is connected, run the project and check the Drain waveform.

E

HI,
I did the test without the coil and I am getting the exact same results...
So I decided to swap resistors, and same issue again. So my only other variable is the MOSFET !!.

I am going to change it and try again

ken

 

Here's the simulation with the 10μs pulse period you show in post #13:

View attachment 287627

Yes the simulation if good. I was simulating with frequency not high enough to see the detoriated effects. But then again, I will need to swap MOSFET and see.

 

I think I found my issue. I have the mosfet Driver heating up too much and actually overheating and basically self destroyed itself.
That will take some time to change it, The PCB has restrictd size. This is in a TSSOP package..
Passing heat sink or Fan would be better ?

 

hi cap,
A photo shot of the PCB would help us to suggest a fix.
E