Hi everyone,
For a product we want to design with LTC6268 as a trans-impedance amplifiers. When I use the built-in model in LTspice to simulate the stability, I find the open-loop gain characteristic of the model did not match the datasheet. The same problem is also encountered when simulating with the LTC6268-10.
According to the LTC6268 gain curve in the datasheet, the slope is essentially 20 dB/decade between 1 MHz and 10 MHz. In other words, the open-loop gain has a pole before 1 MHz. Since a single pole corresponds to a phase shift of -45°, based on the curve this pole should be around 400 kHz. The part not shown on the left side of the figure does not exhibit a -45° phase shift, so it can be inferred that there is no pole within 0.1 MHz.
However, the built-in model in LTspice already has a pole with a -45° phase shift at 2.4 kHz. No matter whether the circuit is unity gain or has 1000 gain, the pole is all around 2.4 kHz. The simulation circuit is shown below. This simulation result does not match the datasheet. As a result, when we simulate the circuit stability, the phase margin at 500 kHz is only a little over 20°. Theoretically, oscillation would occur in our TIA circui, but none is observed in actual measurements.
So is this a problem with my loop simulation, or a problem with the LTspice model?
