• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 1105 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

INA333: Cannot simulate INA333 noise properly

Dakotah
Expert 3900 points
Part Number: INA333
Other Parts Discussed in Thread: OPA333

Hi,

I am working through a setup trying to simulate noise in the INA333 using TINA. I have already gone through and calculated the results by hand but wanted to run a simulation for verification. I am not getting the expect output for the noise spectral density or Total noise graphs. 

I did some searching online and found this presentation on the exact topic I was looking for. I worked through the results got a matching answer by hand but when trying to simulate I run into a number of issues.

My Circuit is the following image and I have included my TINA file.

Issues:

1) If I do not have R8 (100K) on Vout then TINA reports issue with U1.Q21 "Operating point not found".

This resistor was not there in the linked ppt presentation but was there in the INA333 Tina file downloaded from the web.

2) Once adding R8 the issue changes to U2.D17 "Operating point not found".

I then removed the OPA333 circuit setting the Vref CM to mid-supply and just used a voltage source.

3) Finally the Noise analysis will run but only if I set "Analysis>>Mode>>"Single". The TINA comments in blue specify to use "analysis>>Mode>>"Parameter Stepping" but will not run in this mode.

Once running in the "single mode" I get a different response from the presentation on slides 30 and 31.

An image of my Total noise graph looks as follows:

The graph for noise spectral density is also very different. I initially thought it had to do with Log scale on axis but once I changed the axis to match I still had different results.

To summarize: I would like to know if it looks like I am missing a step when trying to simulate noise in the ina333. I have an identical circuit to the one in the linked presentation but cannot repeat the results.

ina333noise_removed_opa333.TSCina333noise_with_opa333.TSC

  • 0
    TI__Expert 3900 points
    One of the items I forgot to mention is that the DC nodal analysis doesn't come out as expected. The Vjunk source was added so I could simulated the noise analysis. But this seems to be causing issues because once I remove this source the DC nodal analysis works as expected.

    Maybe it could be a bug with TINA?
  • 0 in reply to Dakotah
    TI__Mastermind 23050 points

    Hi Dakotah,

    The INA333 is an older transistor based model and can tend to be a little more difficult to work with from a convergence standpoint. When you encounter transistor and diode convergence issues, one thing you can try is adjusting the analysis parameters. TINA comes equipped with some pre-canned settings you can switch to based on the kind of circuit you are simulating. First go into Analysis -> Set Analysis Parameters and click on the hand pointing to the right in the lower right corner. For your circuit, you should select Open and the scroll down until you find the .PRM file that says DC Convergence of non-linear circuits. After adjusting this, you should now see a reasonable output voltage regardless of how the Vjunk source is configured. 

    The noise results you are seeing are likely the result of TINA finding a non-linear operating point for the dc sim before it initiates a noise analysis. It manages to converge in this case, but since some of the internal components have saturated the answer it produces is unreliable (this is why it is always important to make sure you have the correct dc operating point before doing any ac sims!)

    After making the adjustment above, you should be able to produce accurate noise curves. One last thing to note is in the presentation you have attached they show the INA333 in a gain of 1001 with RG = 100Ohms, but the noise curves following the schematic are actually with a gain of 101 with RG = 1kOhm. Hope this helps!

  • 0 in reply to Zak Kaye
    TI__Expert 3900 points
    Thanks Zak,

    I was contininuing to work on the issue and was able to resolve the problem.

    Solved the issue by starting a new design and copying in the exact circuit and pasting into a new file and it worked. I had seen similar bugs in TINA where I would rebuild the file and it would work so I tried that with no luck. But copying the entire circuit and pasting into a new file solved it.