LPV802: How to simulate the phase margin of transimpedance amplifier with TINA

Hi Deniszmm,

Since the first stage has multiple feedback loops, you'll want to open the loop at the input.  I have included a simulation file below to get you started.  You will need to add your model for the gas sensor.  Then, you should test each amplifier one at a time.  You can use this setup to test the second stage stability as well.  Here is a presentation explaining the method.  See slide 13 for the summary of the setup.

Deniszmm.TSC

Please let me know if you need further assistance.

Regards,

Daniel

Hi Deniszmm,

do you have the equivalent circuit model of gas sensor? Or a datasheet?

Kai

Hi Deniszmm,

I'm not sure if you're question was fully answered.  However, we have not heard back from you in a while.

If you need further assistance, please let me know.  Otherwise, I will consider the issue to be resolved.

Regards,

Daniel

Hi Denis,

I would do that as shown below. The problem is that it's very difficult to find an equivalent circuit model of the gas sensor. The only one I found consisted of a resistance of several hundred Ohm in parallel with a HF capacitance and in parallel with a current source. Not sure whether this model is also valid for the sensor in the appnote.

There are two phase stability analysis to be done, one for the left LPV802 and one for the right LPV802:

denis_lpv802.TSC

Looks like the circuit is instable with CF1=100nF?

CF1=100pF seems to be a better choice:

Now the second LPV802:

denis_lpv802_1.TSC

Doesn't look good with CF1=100nF either:

CF1=100pF improves the phase margin again:

Kai

Hi Daniel Miller56，

Thank you for your reply.

If the TIA  gain resistor RF1 value is greater than 100K，the amplifier output will self oscillate。How to solve the problem of self-excited oscillation.

Regards,

Deniszmm

Hi kai klaas69

Thank you for your reply.

Regards,

Deniszmm

Hey Kai, just FYI, this ckt is coming from this development by Paul Grohe, he may have checked phase margin also at the time, 

http://www.ti.com/lit/an/snoa937/snoa937.pdf?ts=1590935253452

Hi Michael,

there's no problem with a high ohmic gas sensor:

Kai

Hi Denis,

decrease CF1 from 100nF to 100pF and check again.

Kai

Hello Deniszmm,

The problem of self-excited oscillation is caused by instability resulting from a pole that goes too low in frequency.  This pole, or poles, will encroach upon the closed loop response of the amplifier and diminish the phase margin.  Instability results.

The cause of this pole is the interaction of impedances.  For example, the feedback resistor might react with the input capacitance to create a pole.  Or, the output impedance might react with the load capacitance.

In your case, adjusting the value of CF1 or RO1 would help, depending on where the pole is being created.  Kai has suggested adjusting CF1 and I think that is probably the way to go.

However, it would help if you had a model for the gas sensor so we can be sure about the phase margin.  Do you have one you can provide?  How do you know the circuit is unstable?  Did you simulate or build it?

If you have a gas sensor data sheet, I will build a full model in TINA and we can simulate with some more confidence.  Else, I will just do my best to approximate a model, as Kai has done.

Regards,

Daniel

Michael,

Good catch.  I will reach out to Paul and see if he's got any information that might help here.

Regards,

Daniel