View attachment 257990
The SpiceModel line. That is where I specified the external PSpice model from TI when I used the opamp2.asy file to build my TL071.asy model that worked for a while and then quit.

Yes. DON'T do that on my symbols yet.

 

BINGO! That was the missing piece and probably would have made my frankenstein model work! Thx!

Yes...LTspice remembers where you last selected the file.

 

View attachment 257990
The SpiceModel line. That is where I specified the external PSpice model from TI when I used the opamp2.asy file to build my TL071.asy model that worked for a while and then quit.

If you do that, LTspice will look in the default lib/folder for the filename, even if its already defined on the schematic.

 

Only downside is it is using the entire long file path which is quite a few subdirectories down, but it works! I can live with this but filesharing is not possible it seems.

 

LTspice will look in the default lib/folder for the filename

Which is where? The subdirectory the symbol came from?

 

Only downside is it is using the entire long file path which is quite a few subdirectories down, but it works! I can live with this but filesharing is not possible it seems.

The reason I structure the zip file that way is so that IT CAN BE shared easily.

 

Which is where? The subdirectory the symbol came from?

No...the default library installation folder.
If using windows 10:

Documents\LTspiceXVII\lib\sub

 

For me it is using on Win10 "C:\Users\SamR\Documents\LTspiceXVII\lib\sym\EETech\TL07xTI.txt"

Which is where the symbol came from and without the pointer could not find.

 

For me it is using on Win10 "C:\Users\SamR\Documents\LTspiceXVII\lib\sym\EETech\TL07xTI.txt"

Which is where the symbol came from and without the pointer could not find.

Please be aware the following IS NOT a recommendation, but will allow you to use the default folder structure:

To use the LTspice default folder structure:
The symbol file .asy should go in lib\sym\ folder
The model file .anything should go in lib\sub folder

The following symbol attributes should be populated:
SpiceModel: the name of spice model
Example:
SpiceModel: TL071
Modelfile: the filename of spice model definition file.
Example:
ModelFile: TL07xTI.txt

These attributes should not contain folder paths to use the default library structure.

However,
The model file can be located in a specific folder below lib\sub\ but the "ModelFile:" attribute must
contain the complete path relative to lib\sub\ for LTspice to find it.

Example:
The spice model "TL071" is defined in spice model definition file "TL07xTI.txt".
The model file "TL07xTI.txt" is located in folder Documents\LTspiceXVII\lib\sub\Contrib\eetech00\

The SpiceModel: attribute should contain:

SpiceModel: TL071

The ModelFile: attribute should contain:

ModelFile: Contrib\eetech00\TL07xTI.txt

Edit: I've attached a readme file that further clarifies use of the TL071,2,4 symbols. It is important to use the symbol that is intended to be used with the model file, otherwise, erroneous simulation results can occur.

 

Aha! Moved the TL07xTI.txt file to the sub folder and now long path not needed. Got it! I am having an issue with the simulation plot output. I'll post about it later but seems the model giving erroneous data output. I'm getting only -dB output plots for noninverted use.

 

Not sure why it's called a 3dB error since that's the normal rolloff of a 1-pole LP filter, but the rolloff is indeed -3dB at the corner frequency.

X1 just appears to be an arbitrary label at the unity gain point.

It is called a "3dB error" because there is an error between the exact frequency response and the asymptotic Bode approximation.

Sometimes the exact frequency response is called a "Bode Plot" but the original Bode idea was to analyze the circuit using an asymptotic approximation which involves straight lines for simple responses rather than curves. This was originally to make an analysis easier so it can often be calculated by inspection with no need for slide rule or calculator. Since the gain at the-3dB point on the asymptotic Bode plot is 3dB higher than the actual exact frequency response, it can be called an error because it's not really that high it is really 3dB lower as we see in the plots.

It may make sense to call the exact frequency response a "Bode Plot" since it is often taken to be that, but the "Asymptotic Bode Approximation" is the straight line approximation except in the case of the 2nd order response and then predetermined standardized curves are used to get a close approximation still without the need for much real calculation effort.

In the straight line approximations, a positive slope indicates a high pass function, a negative slope indicates a low pass function, and when a high pass is followed by a low pass at a higher frequency the result is a flat response. So a positive slope at say 100Hz followed by a negative slope at 500Hz followed by another negative slope at 5000Hz would indicate a bandpass filter. So approximately, the response goes up at around 100Hz, levels off to flat around 500Hz, then decreases around 5000Hz. It's not always that simple of course but often it can be.

 

Aha! Moved the TL07xTI.txt file to the sub folder and now long path not needed. Got it! I am having an issue with the simulation plot output. I'll post about it later but seems the model giving erroneous data output. I'm getting only -dB output plots for noninverted use.

works ok for me.

 

Here is what I am getting for simulation. Note all the output is the exact same for the different cases.

 

I see it now, the circuits are wrong, correcting now.

 

Here is what I am getting for simulation. Note all the output is the exact same for the different cases.
View attachment 257995
View attachment 257996
View attachment 257997

Post your .asc files please.

 

I see it now, the circuits are wrong, correcting now.

Your first image states "open loop" but it is not configured for open loop gain test.

 

Pretty much no rolloff to 1 MHz

 

not configured for open loop gain test.

What did I miss?

 

Arghh... Still wrong circuit! Give me a few and be back!

 

What did I miss?

There's an example in the zip file I posted.