Post Go back to editing

ADRV9029 EVM and ACPR does not meet typical measurements in datasheet

Category: Hardware
Product Number: adrv9029

Hi AD support,

I'm trying to have some tests with ADS9-V2EBZ + EVAL-ADRV9029-HB on TES-GUI-6.3.0.5. I'm using profile 50 Link sharing profile. When transmitting 20MHz LTE signal from TX1 port, I got the following VSA decode info: EVM at -40dB and ACPR at -61dBc. However, in comparison with ADRV9029 datasheet, the typical values are -46dB and -65dBc, respectively.

I only use the Tx1 port for measurements, and I enabled all calibrations already. Do I need to run any other optimization process?

  • At what output power, are you checking the EVM and ACPR?

    Is the Ref Clock phase noise in line with the desired phase noise specs?

    What is the digital backoff of the signal or in other words, the PAR of the signal? You will need to back off by the PAR at least from the full scale power so the peaks wont get clipped.

    Try to optimise the spectrum analyzer settings as well.

  • I'm using LTE signal, increasing the channel power to -9dBm, PAPR 11.3dB resulting 2.3dB peak, still lower than the maximum power on Tx ports according to the datasheet. So far, I can see that ACPR is better (-66dBc), but EVM is still far from the datasheet (-41dB versus -46dB).

    The Ref clock is Keysight N5182 Sig gen, which is in line with the desired specs.

    What do you mean by optimize the spectrum analyzer?

  • Can you decrease the TX attenuation which means the output power will be more and the EVM can be better (as the SNR will be better)?

    Can you check the EVM by doing Auto-Range (The 4th button in the top left corner) on the VSA?

  • So far, I tried both: increasing Tx power as far as the peak is not clipped, as well as set the SA to lowest sensitivity before over-loaded; nothing resolves that.

    Can the quality of ADP5056 affects the EVM? Previously, I have a defective ADP5056 board and EVM is really bad. Right now, the new ADP5056 makes the EVM better; so is there any chance that an even better ADP5056 can further improves EVM.

  • You just replaced the ADP5056 part or replaced the power board with a new power board?

    This could be problem with the power supply rail bypassing capacitors. 

    Can you check with the local representative to get a new power board and test with that once?