THS3491: output can not reach 5.5V beyond 300Mhz

Hi,

Can you share a schematic with the supply values the customer is using. Information about the input signal will also be useful. 

Best Regards,

Ignacio

Hi,

  Customer use THS3491DDAEVM, TRF1108-DAC39RF output as input signal.

Hi,

Can you share the supplies? Is there a load that the device is driving or are you just probing at the output? Is the expected output 5.5Vpp?

Best Regards,

Ignacio

Hi,

  Customer use LT3090 and LT3045 to output ± 8V power supply, followed by an oscilloscope. We need to calibrate the oscilloscope module with a sine wave, with an output frequency of 40MHz-600MHz and an output voltage of 5mV-5.5V. The best frequency response is around one percent.

Hi,

Can you confirm if this measurement is being done without any sort of load at the output. We get around ~500MHz of bandwidth at 5Vpp with this device in a similar supply configuration as the one you mentioned. I am wondering if maybe loading is affecting the measurement. If you probe at the input of the device where the 50ohm termination resistor is placed. Is your input signal the same amplitude up to the frequencies of interest?

Best Regards,

Ignacio

Hello,

  Increasing the input voltage can indeed output, but the frequency response is too poor. Do you recommend other better devices? Similar to TRF37C75, peak to peak voltage 5.5V, frequency range 40M-600MHz.

Customer want to make a 5520 oscilloscope correction module, and their peak to peak voltage is up to 5.5V, within 600Mhz,

Hi,

This is our highest performing large signal bandwidth device. Unfortunately, we currently do not have a device that can output 5.5Vpp at 600MHz. I am surprised it is dropping off the so much at 300MHz but it will likely not get you to 600MHz without some level of attenuation as its 3dB bandwidth is 550MHz at 5Vpp.

I can forward your thread to our RF amplifier team to see if they have a device that can output 5.5Vpp at these frequencies.

Best Regards,

Ignacio 

Thanks, Ignacio.

Hi,

Seems the system does not need to support dc (0Hz), f is 40MHz to 600MHz, and TRF1108-DAC39RF is used already.

Based on the details, I'm suspecting the scope load configuration for calibration is 50Ω, and they are looking for device similar to TRF37C75.

Is there a reason TRF37C75 does not work?

We have pin-to-pin compatible gain variants in this TRF37x75, TRF37A75 and TRF37B75, if one of those gain variants is more suitable.

Keyur

Hi,

  The TRF35C75 operates on a 5V power supply and outputs waveforms with a peak-to-peak voltage exceeding 3.5V, all of which are severely distorted, using 50-ohm termination

Hello,

  Any else comment?

Yes, OP1dB of TRF37C75 is close to 19dBm (~5.6Vpp on 50Ω) so if linearity is also a concern, it needs to be backed off, and 3.5Vpp is about right where one will start to see the waveform distort.

For TI options, I think we are back to THS3491.

If source and load are both 50Ω, the amplifier will have to swing 11Vpp right at the output with a total load of 100Ω, to get 5.5Vpp at 50Ω load. They should be able to get about 300MHz, ~5.5Vpp into 50Ω with THS3491, maybe a bit more if inverting config used, but not 600MHz.

Does the load have to be 50Ω or is 1MΩ termination that many scopes have ok, or some lighter load in between?

If it has to be 50Ω load, they would need an RF amplifier (or PA, power amplifier) that supports 40MHz to >600MHz frequency with about 24-25dBm OP1dB, that TI currently does not have.

Keyur