Op Amp Bode Plot 3db Point

Thread Starter

SamR

Joined Mar 19, 2019
5,031
OK, now I really have egg on my face. I completely forgot about the 1+... That makes the result pretty obvious. I'll have to redo it as an inverting instead to eliminate the 1+.

EDIT: Nope, still won't work. A = -(Rfb/Rin) so I can attenuate the negative gain by increasing Rin, but it will never cross over zero. So the Bode plot showing both +&- dB is to cover both conditions of inverting and noninverting. K, think I have it now. Time to breadboard.
 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Inverted results as expected this time.
i10k.pngi30k.pngi100k.pngi500k.png
Essentially 0V flat line output, reading 0.61mV on bench meter. Ran R1 up to 10M with no appreciable difference in output. OK, have it now. Thx for the input guys! The only thing I didn't look at now was roll-off. I'll save that for another day.
 

crutschow

Joined Mar 14, 2008
34,281
OK, I'm not quite up to speed on dBs. I'm still thinking in volts and gain as Vout/Vin...
dB is simply the log10 ratio of the two values times 10, i.e. 10*log(Pout/Pin).
Since power is equal to the voltage squared the voltage dB is 20*log(Vout/Vin) -- (multiplying a log value by 2 is the same as squaring the value).

Thus a voltage gain of 10 equals 20*log(10) = 20dB, a voltage gain of 1 equals 20*log(1) = 0dB, and a voltage gain of 0.1 equals 20*log(0.1) = -20dB.

Make sense?
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
I understand the formula. Maybe I should make a chart of gain/dB. I did a couple of calculations such as 20log(10kΩ/500kΩ)1V))=-13.98dB, 20log(1)=0dB, 20log(10)=20dB, but not sure how it lays out around zero dB. So obviously gain greater than 1 is positive dBs and less than 1 is negative dBs. In other words, it not instinctive for me as yet. I do understand the decades but not intuitively. It will come with use, I'm just not used to thinking in dB.
 

crutschow

Joined Mar 14, 2008
34,281
I do understand the decades but not intuitively.
Well if a gain of 1 is 0dB, and all gains greater than 1 are positive dB, then the only thing left for gains of less than one is the minus dBs.

One nice thing about dB is that it tends to give a straight line on a graph for things like the rolloff of a filter.
Hearing is also logarithmic, so sound power in dB directly relates to how we perceive the sound intensity.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Learned logarithms over 50 years ago and never really used them except for my slide rule back in the day so I'm a bit rusty on them. Hope it gets better with experience. Sound was not one of my favorite physics topics back then. All I remember is speed of sound in air roughly 760MPH @STP and enough conversion factors to get FPS or Meters/S. I was introduced to dBs studying filters and waves and know where to look it up at least.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
I recently saw some LTS tutorials on the Analog Devices site and also have that on the ever-growing todo list.

@ericgibbs I am curious as to where you found the model for the TL081? On the TI site they support their model of PSpice for which most of the links I tried are broken. I have downloaded and installed the Brododynov libraries but it doesn't include the TL line. I don't think I'm quite up to building my own models as yet.

EDIT: I finally found the PSpice model on TIs site. I assume it works in LTS? Now need to figure out how to add it as it's been a while since I added one.
 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
5,031
The PSpice models on the TI site are NOT compatible with LTS! Lesson learned. Now, where can I find the TI OPA models?
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
When I opened up the file, the TL081 model is simply the opamp2 model with the TL081 name attached to it. I'm starting on the section of the handbook dealing with how the various external components affect the open loop gain and would like to model it with an accurate model before breadboarding it for comparison. Someone, Carl I think, gave me a 741 model a year or so ago but sure wish I had the TL models.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
OK, I have both the TL081.301 and TL071.301 files from TI. I don't understand the "set the model path with .include TL082.301" bit. And I assume that is a typo and should be TL081. And where do the .301 files go?

1642275030231.png
 

ericgibbs

Joined Jan 29, 2010
18,766
hi Sam,
The TL081.301 is the name of the Model text, you can read it using a text editor to see the model file text.

The .include TL081.301 tells LTS to use that file as the Model/specification of the TL081
Keep the .301 file in the same folder as the asc files for the TL081

For ref only the F2/Opamps has the 'opamp2' symbol file.

Do you follow, OK.
E
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
.include TL081.301
The .include TL081.301 statement goes on the schematic itself and the file TL081.301 in the same folder as the symbol TL081.asc? Not familiar with .include so a bit confused. So, the PSpice files are used as declarative library defining the symbol?

EDIT: Think I have it now but haven't tested. Text edit schematic as a spice directive on the schematic to point to where the TL071.301 PSpice definition/library file is stored. Which is in the same symbol library as the TL071.asc symbol.
 
Last edited:

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Not sure what happened but edited and built a new TL071 that does not need to call the descriptor and works.
1642309569435.png
 

Thread Starter

SamR

Joined Mar 19, 2019
5,031
Not able to Bode plot.
When I click on the left hand scale range I do not get
1642309835984.png
I get this so I assume it must be something in the config
1642310344263.png
Looking at the control panel, waveforms I find
1642310226726.png
I don't see anything about Bode plotting.

EDIT: I found it came up only after setting the ac frequency sweep. Still puzzling it over a bit. But I am slowly making some headway with it. Lots to learn about the ins and outs of LTS.
 
Last edited:
Top