Op-amp Oscillation

well before someone runs a LG phase margin test, how are you measuring the output? Needs to be low capacitance sense. 

Hello Mohammad,

Your circuit configuration is unstable, measuring at around 31 degrees of phase margin in sim. Not sure what your exact design requirements are, but if your goal is to subtract your inputs and achieve a differential output why not use the FDA in a typical configuration? 

THS4151_Stability.TSC

Best,

Hasan Babiker

31deg is not unstable, but, with a bad load it might get there, Load???

Hi, 

I'm sorry for the late reply. Thank you for taking the time to reply to my post. Below I'm sharing the full schematics for what I'm working on:

From the beginning, I was having trouble with the THS4151 it is over heating too. I think it's also effecting the other op-amps in the circuit. The output of the THS is two 100 ohms resistors which then it's connected to AD8330 VGA to gain the signal. Now, I did remove R34 and R30 and I did put in a 1 MHz sine wave from a function generator. I grounded the side of the R30 resistor so I know that it's 0 volts. This will allow to only worry on one side. Now, Normally, the two signals coming to R34 and R30 are the same unless the device is making a measurement in space, then the signal on the R34 side will be slightly higher. If there is any suggestion on how I can make the THS more stable would be appreciated.

Thank you

Hi, 

I'm sorry for the late reply. Thank you for taking the time to reply to my post. Below I'm sharing the full schematics for what I'm working on:

From the beginning, I was having trouble with the THS4151 it is over heating too. I think it's also effecting the other op-amps in the circuit. The output of the THS is two 100 ohms resistors which then it's connected to AD8330 VGA to gain the signal. Now, I did remove R34 and R30 and I did put in a 1 MHz sine wave from a function generator. I grounded the side of the R30 resistor so I know that it's 0 volts. This will allow to only worry on one side. Now, Normally, the two signals coming to R34 and R30 are the same unless the device is making a measurement in space, then the signal on the R34 side will be slightly higher. If there is any suggestion on how I can make the THS more stable would be appreciated.

Thank you

But when you are measuring that oscillation where and how are you making that measurement?

I'm not sure I understand your question correctly, but I'm measuring with the oscilloscope probe across the two 100 ohms resistors. I used a function generator to input the signal after removing R34 and R30 from the printed circuit board. The THS is very hot to the touch which tells me that it's oscillating without even measuring it. 

Hello Mohammad, 

Can you share your requirements for the THS4151 stage of your circuit? Gain, bandwidth, noise, etc.

Best,

Hasan Babiker

As far as the op amp bandwidth, I would be happy if I can get between 1-10 MHz. As far as the gain, it should be one. The signal will be gained using the AD8330 VGA. As far as the noise, I'm really not sure. As low as I can get it to be.

Thank you for taking the time to help me.  

Hello Mohammad,

Below is a configuration for a 10.3MHz filter. Note that the 1kohm & 4pF load is meant to represent the AD8330 input and isn't a part of the circuit:

10,3Filter.TSC

If you don't have a slot on your board for a differential input cap, you can use the following:

10,3filter_alt.TSC

Best,

Hasan Babiker

Well Hasan, I think he still has an oscillation,

Since the nominal sim shows fine phase margin, there must be some parasitic not in the simulation. Next would be supply decoupling and layout capacitance questions. 

Good point here Michael,

Mohammad, have you tried measuring the outputs of the device when removing the load (removing 100-ohm resistors)? Were layout recommendations followed for THS4151 board design? (note removal of gnd and power planes under input/output of device). 

A higher phase margin design may still cause issues if proper layout techniques weren't used here.

Best,

Hasan Babiker

Thank you, I will try this circuit configuration and I will let you know.

I didn't try measuring without the load. I will try that and let you know. As far as the layout, I followed that correctly. Thank you