For problems encountered when testing power loop response with the ADALM2000

Hello,

So you were able to reproduce successfully the hardware setup  / measurements from `ADALM2000 Measuring Loop Gain`?

Can you please provide the exact hardware setup that you are using in your custom experiment?

Also, you might find some useful information about power supply loops tests/designs here: https://www.analog.com/en/technical-articles/power-supply-loop-stability-loop-compensation.html

Regards,

Antoniu

Thanks for getting back to me.

There's a photo of the test equipment.

As you can see, our hardware equipment includes an ADP5051 DEMO board, the main changes are the addition of a 10 ohm injection resistor in the second channel of 3.3V output voltage, and a WB1010-1 transformer by way of flying wire.

This is a picture of the circuit we tested.

We believe that the circuit diagram is the same as that required by the specification for testing loop response. Attached is a sheet of the circuit used for testing in the manual.

Regards,

Wang

Hello,

Thanks for the additional information. From the picture you sent with the hardware setup I can see 3 wires connected from the ADALM2000 which I suppose that represent W1, 1+ and 2+. Do you have a common ground between ADALM2000 and ADP5051 demo board?

Another way to debug things is to try and simulate your circuit with: www.analog.com/.../ltspice-simulator.html

This might narrow down the possible issues.

Regards,

Antoniu

Thanks for the quick reply.

For your doubts, we make a common ground between ADALM2000 and demo board.  We now suspect the PCB trace parasitic inductance or capacitance can cause additional errors to loop Bode models. Or the problem is located in the transformer?  But that's just a guess. 

Meanwhile, I noticed ADALM2000 have AD-M2KBNC-EBZ, a BNC adapter board of ADALM2000. Whether this adapter is required when using the ADALM2000?

Is it because the amplitude of the injected signal through the fly-line test is only 50mV when it is disturbed by noise? However, when we tried to increase the injection voltage to 5V, the waveform still did not appear. Is it necessary to redraw the PCB with the ADP5051 power chip and make a new transformer for testing?

As for the use of LTpowerCAD, I believe the products designed by your company are in line with the test. But due to external component value inaccuracy and variations, there will be some difference between the actual test results and the simulation results. And because there are different power chips including switching power supply or DCDC power supply chip, the ADALM2000 method may be more convenient and quick in the future.

Looking forward to your insight.

Best wish,

Wang