TINA/Spice/THS3001: THS3001 Noise

Hi Dinh,

Some questions/comments to better understand your problem:
1. Is this noise captured with an input signal applied, or you are not applying any input signal and still getting this noise?
2. From the noise waveform attached, it does not look like the input being applied is properly referenced to GND and what you are seeing could be a Hum or GND bounce. This is not the case for the output connector (J3) because it seems to be referenced to GND. So, would it be possible for you to make the input TPG3 as an SMA connector similar to J3?
3. You could try to reduce the noise at the output by reducing gain of the THS3001 by using a higher value R6 = 100-ohms. This should indicate whether the noise is actually coming from the device or being picked up elsewhere from the environment.

Best Regards,
Rohit

Thanks Rohit,

I did not apply any input signal, but the noise still happened. 

When reducing the gain, the noise was smaller.

However, I had to connect this amplifier to another THS3001 (also printed by Voltera V-One PCB Printer) to achieve my target gain. And the output (without input) still noised.

I tested with both power supply and 5V batteries, replaced the THS3001 by THS3201 and OPA847, but nothing changed.

Hi Dinh,

Would it be possible for you to make the R5 as 1kohms (or higher) and R6 as 26.7 ohms (scaling R6 accordingly for the gain required), and try to re-check for the output noise? It looks like the noise waveform you attached could be due to instability in the THS3001. The device could be made more stable by increasing the Rf value (or R5 in your case) to 1kohms or higher.

Both the THS3001 and THS3201 require a minimum supply voltage of 9V or +/-4.5V split supply for proper operation. The same goes for the OPA847 which requires a minimum of 10V supply for proper operation. I don't think any of these devices are supposed to operate at single 5V supply. Also, I would refrain from using the THS3201 because it is not recommended for new designs.

The other thing is that would it be possible for you to zoom-out on the oscilloscope screen shot to capture longer duration, and re-attach the scope screen shot?

Best Regards,
Rohit

@Rohit : I connected two inputs of LMH3401EVM to ground via two 1k resistors.

And the noise still happened as follow figure:

The blue and yellow graphs belong to two outputs of EVM board.

I want to amplify 50ns - 1mV pulse to 100mV, but this noise is also 50ns and its amplitude is up to 20mV.

How can I handle this problem?