Other Parts Discussed in Thread: OPA197, LF353
I'm using the OPA2197 dual opamp in an inverting opamp summing junction with 3 input signals. The device is powered from +/-12Vdc. Each input signal consists of an 25Khz to 60Khz 2Vpp sinewave with a DC component of around +1.5VDC. (Thus the signals swing from 0.5 to 2.5Vpp). I put a DC blocking cap on each input, in order to sum only the AC components. Thus the opamp output should consist of only of the sum of the AC signals, which should then swing between +/- (3 x 1V) = +/-3V or 6Vpp, with 0Vdc.
So far the circuits works very well,
I was trying to estimate gain and phase margin of this circuit, using the OPA2197 datasheet.
When looking at the open loop Bode plots of the OPA2197 (figure 10, page 13) I noticed the following
->Open loop gain A0 (gain at 0Hz)=~ +135dB
->Open Loop Phase shift starts at around +65degrees at 0Hz,and drops to around +20degrees at 40Mhz.
This looks atypical of opamp phase shifts, which typically start at 0degrees at 0 Hz.
Also for this estimate, I tried to simulate the circuit using the TINA and also Multisim analog simulators. In each case, the 'closest' spice model is the OPA197. I tried to simulate the combined ‘open loop’ response of the opamp + the R1/R2/C feedback circuit (R1= feedback resistor = 10K to 40K, R2= gain setting = 10K, resistor C = 100nF DC blocking cap) .
I wanted to estimate the phase margin using R2 (feedback res) with values ranging from 10K (AC gain = 10K/10K=1) to 40K (AC gain = 40K/10K=4).
So, as a first step I tried to simulate the open loop Bode plot of the OPA197, to see if it would be close to the datasheet. Well, the simulated open loop bode blots was totally different from the datasheet graphs. Moreover I tried to simulate a older ‘classic’ opamp, using the LF353. This time the simulated open loop results were close to the LF353 datasheet in one tool (Multisim) and not as close with the TINA tool.
Here are the open loop simulated vs datasheet results:
DC open loop gains Phase shifts at 1Hz
Device Datasheet TINA Multisim Datasheet TINA Multism
OPA197 +135dB -34dB -26dB +65° 0° 0°
LF353 +110dB +14dB +100dB N/A 0° 0°
Note that when simulating with a ‘normal’ closed loop opamp circuit set to gain = +20dB , the results using both simulation tools was correct for the DC gain (20dB as expected).
1) I’m wondering if the open loop positive phase shift graph in the OPA2197 datasheet is correct or not?
2) If it is correct, then would this +ive phase shift explain why the OAP2197 is able to drive high capacitive loads without oscillating?
3) Can you provide the estimated gain & phase margin for the following circuit values in my OPA2197 circuit?
RFdbk R1(gain setting) C(DC block) AC gain Est Gain Margin Est Phase Margin
-------- ---------------- ------------- -------- ------------------ --------------------
40K 10K 100nF 4 ? ?
30K 10K 100nF 3 ? ?
20K 10K 100nF 2 ? ?
10K 10K 100nF 1 ? ?
4) Can we trust the spice model for the OPA197 or not, using TINA and Multsim?
I have the following 10 files available as attachments if needed. (It seems I cannot add files in this post..)
1_opa2197 summing circuit with DC blocking cap expected phase and gain margins.pdf
2_opa2197 datasheet open loop gain and phase shift.pdf
3_multisim_opa197 vs lf357 open loop bode plots.pdf
4_multisim_opa197 vs lf357 20dB closed loop bode plots.pdf
5_TINA op197 vs LF353 open loop bode plots.pdf
6_TINA op197 vs LF353 20dB closed loop.pdf
opa197 vs LF353 open loop gains only.ms14 (MultiSim open loop circuit simulation using the OPA197 and LF353)
TINA op197 vs LF353 open loop.TSC (TINA open loop circuit simulation using the OPA197 and LF353)
opa197 vs LF353 20dB closed loop gains only.ms14 (MultiSim closed loop 20 db Gain simulation using the OPA197 and LF353)
TINA op197 vs LF353 20dB closed loop.TSC (TINA closed loop 20 db Gain simulation using the OPA197 and LF353)