THS3091: Output attenuation observed above 30MHz

Ryan Morrison

Part Number: THS3091

Hi,

I am trying to amplify a DDS sine wave generator using the THS3901. The DDS can produces a sine wave from 50kHz to 70MHz with an amplitude of 50mVpp. As a test, I set up the THS3901 as a non-inverting amplifier and tested gains of 1, 2, and 10 using figure 60 as my reference schematic.

I am seeing significant attenuation of the output starting at 30MHz at all gain levels, which gets worse with increasing frequency. There is almost no output above 50MHz. From the datasheet, I was assuming I would be able to use the THS3901 with a gain of 10 to consistently amplify my sine wave generator output over its entire frequency range. Am I missing something?

Thank you,

Ryan Morrison

over 5 years ago

0 Micah B Brouwer over 5 years ago

TI__Intellectual 1910 points

Hi Ryan,

Thanks for reaching out and welcome to E2E. Could you provide your schematic and scope shots of the input and output signals?

0 kai klaas69 over 5 years ago

Guru 140200 points

Hi Ryan,

does the DDS have a restauration filter at its output? How do you connect the scope to the output of THS3901?

Kai

0 Ryan Morrison over 5 years ago in reply to kai klaas69

Prodigy 60 points

Yes, I do have a LPF on the output. Please see the attached schematic. I was testing the 10x gain stage of my planned amplifier. I had a BNC cable going to a oscilloscope with the 50 ohm load setting.

impedance measurement.pdf

0 Ryan Morrison over 5 years ago in reply to Ryan Morrison

Prodigy 60 points

Unfortunately the scope I was using is gone for calibration for a few days, but I will post scope shots when I get it back.

0 kai klaas69 over 5 years ago in reply to Ryan Morrison

Guru 140200 points

Hi Ryan,

the frequency response of the LPF looks a bit strange. Is it 3.9p or 39p in the LPF? Where do you have this LPF from?

Is the input of LPF correctly driven by a 50R source resistance? Do you trust the transfomer?

Driving the LPF by a transformer might not be a good idea. All these parasitics can compound...

Kai

0 Ryan Morrison over 5 years ago in reply to kai klaas69

Prodigy 60 points

This is the reference design from Analog Devices's AD9913 evaluation board that I purchased.

AD9913_RevC_Schematic.pdf

0 kai klaas69 over 5 years ago in reply to Ryan Morrison

Guru 140200 points

Hi Ryan,

ok, divide and conquer! I would first check the signal behind the transformer, with disconnected LPF. Use the 50R input of your scope. Then I would check the signal behind the LPF, with disconnected THS3901. Connect the 50R input of your scope directly to the output of LPF. Disconnect everything on the right (THS3901, including the 50R input resistor). This will demonstrate the response of transformer and LPF without the THS3901.

Kai

0 Ryan Morrison over 5 years ago in reply to kai klaas69

Prodigy 60 points

I apologize that I also forgot to mention that I replaced the DDS and LPF with a lab bench function generator to see if the DDS and LPF were the issue. I am seeing the exact same 30MHz and beyond attenuation problem with the THS3901. I wish I had my scope right now to show you waveforms. I will send those when I get it back.

0 Micah B Brouwer over 5 years ago in reply to Ryan Morrison

TI__Intellectual 1910 points

Hi Ryan,

Any update on this?

0 Ryan Morrison over 5 years ago in reply to Micah B Brouwer

Prodigy 60 points

Hi everyone,

I very much appreciate the feedback and quick response time. I will have my scope back by this Friday. I will do my best to get some scope captures and pictures early next week.

Sincerely,

Ryan

Amplifiers

Amplifiers forum

THS3091: Output attenuation observed above 30MHz