Testing Support for ADL5380

Hi Aishwarya,

Here's some commentary regarding your questions:

1. The conversion gain numbers in the datasheet are voltage gain for a differential pair (for example I+ and I-). You could measure the differential voltages into the RF port and out of one baseband differential pair to confirm gain in your setup, using the equation 20 x log [Vout/Vin].
2. By default the balun is installed to provide single ended configuration (into 50ohm spectrum analyzer), while maintaining 450ohm differential load to the demodulator.
3. The resistors R6x and R7x are optional placeholders for use without the balun. Ideally you want to maintain 450ohm differential, but an acceptable range is 200 to 500 ohms differential. To achieve 50ohms on each baseband output, it's recommended to use the balun, especially for >5MHz outputs.
4. For use without the balun, you will need to know the next stage differential input impedance to compute the appropriate value of R6x and R7x.
5. In theory the baseband ports will support 1kHz-999kHz output, provided you present 450ohms differential load. The conversion gain numbers in the datasheet are expected. However, at these low output frequencies, the balun is likely not needed and could be problematic due to the balun's lower cutoff frequency. In general (many applications, systems, etc), for less than 1MHz operation maintaining a 50ohm system is not necessary. You can safely use high impedances while achieving good power transfer at low frequencies.
6. For reference there are characterization setup diagrams in the datasheet.

Best Regards,

David

Hi David,

Thank you for the response. I have one more confusion. Could you please explain how theoretically, the differential 50 ohm at the output of I/Q demodulator is being transferred as 50 ohm single ended input to spectrum/network analyzer? Is it 50ohm parallel to 450 ohm? 

Hi regarding the gain; 

When we connect the single ended output to spectrum analyzer, what are we supposed to see in terms of power level? Suppose if I provide, -20dbm of RF Power, could you please explain why am I seeing -29dBm of power at QP/IP pin of EVAL board?

Hi Aishwarya,

1. Regarding impedance transformation, I recommend reviewing some literature on transformers, such as a circuits electrical engineering text book or Wikipedia. The transformer used on the ADL5380 evaluation board is the TCM9-1, which has an impedance ratio of 9:1. This means the impedance as seen by the ADL5380 will be 9x the impedance on the other side of the transformer.
2. You might expect about 4 to 5dB of power loss in a 50ohm system with the ADL5380 operating under your conditions, with the transformer. For the observed power on the spectrum analyzer and your calculations, be sure to account for the loss of the transformer, any cabling, and connections.
3. As a sanity check, you could confirm with a power sensor the actual measured power at the end of the cable for the ADL5380 RF input, as well as the observed output power on the ADL5380 outputs. Then connect the outputs to the spectrum and measure the power again.
4. Another possibility is the spectrum analyzer is unable to measure power correctly at low frequencies. Some spectrums have a DC blocking capacitor enabled by default for protection, but at lower frequencies it can be a series high impedance, so be sure to check the input configuration.
5. Alternatively, perform a spectrum analyzer calibration with an external signal generator. To do so, measure the signal generator power at the desired frequency, using the power sensor; then connect the signal generator to the spectrum analyzer to confirm the measured power is as expected / determine any needed offsets.
6. In any case when using the power sensor, prior to measurements, be sure to perform zero/cal of the sensor, and adjust the operating frequency (cal factor) according to the frequency being measured.

Best Regards,

David

Hi Aishwarya,

1. Regarding impedance transformation, I recommend reviewing some literature on transformers, such as a circuits electrical engineering text book or Wikipedia. The transformer used on the ADL5380 evaluation board is the TCM9-1, which has an impedance ratio of 9:1. This means the impedance as seen by the ADL5380 will be 9x the impedance on the other side of the transformer.
2. You might expect about 4 to 5dB of power loss in a 50ohm system with the ADL5380 operating under your conditions, with the transformer. For the observed power on the spectrum analyzer and your calculations, be sure to account for the loss of the transformer, any cabling, and connections.
3. As a sanity check, you could confirm with a power sensor the actual measured power at the end of the cable for the ADL5380 RF input, as well as the observed output power on the ADL5380 outputs. Then connect the outputs to the spectrum and measure the power again.
4. Another possibility is the spectrum analyzer is unable to measure power correctly at low frequencies. Some spectrums have a DC blocking capacitor enabled by default for protection, but at lower frequencies it can be a series high impedance, so be sure to check the input configuration.
5. Alternatively, perform a spectrum analyzer calibration with an external signal generator. To do so, measure the signal generator power at the desired frequency, using the power sensor; then connect the signal generator to the spectrum analyzer to confirm the measured power is as expected / determine any needed offsets.
6. In any case when using the power sensor, prior to measurements, be sure to perform zero/cal of the sensor, and adjust the operating frequency (cal factor) according to the frequency being measured.

Best Regards,

David

Hi Thanks for the clarification. Could you please let me know where can i get IMT Table and S-Parameter file for ADL5380?

Hi Aishwarya,

You're welcome. Unfortunately, at this time there's not an IMT table or S2P files readily available for the ADL5380. The best I can offer is the data that is published in the datasheet. You will note that the return loss is respectable; please see the typical plots.

Best Regards,

David