• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 20 subscribers
  • Views 453 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LMX2572: A few questions about the device

Alex OHM
Prodigy 70 points
Part Number: LMX2572

Hi,

I have a few questions about the device from using the LMX2572 Evaluation Board.

1. I seem to have a larger current draw when using the internal channel divider than anticipated

I've tested this on 2 EV boards and seem to get a 20mA current bump (when selecting the CH div) instead of about 10mA in the datasheet

Both boards are clocked externally with a 20MHz Clock and the onboard 100MHz TCXO is disconnected. Charge Pump current is 2.5mA, and the Loop Filter Components give a ~30kHz Bandwidth (changed from the default components that came with the board). Both boards are supplied 3.3V and are configured identically. Output taken from only the positive A channel (B Powered down), negative output terminated 50Ohms.

For the first one:

at 4.5GHz I get 1.23dBm with 67.5mA (vco). at 3GHz I Get 2.2dBm with 87.5mA (div 2)

The second one:

at 4.5GHz I get 1.31dBm with 73.9mA (vco). at 3GHz I Get 2.31dBm with 97.5mA (div 2)

this is about a 20mA bump when the typical current should be about 10mA less with the channel divider on. Is there anything I'm missing about the comfiguration that could warrant the additional current?

+ this is quite a large variance between the cards, and the current specified is only the typical value ~ I was wondering if there is any information for what to expect - max/min current of tested cards etc.

2. In my application I might require to use a 3V regulator instead of the typical 3.3V. If (due to supply voltage accuracy) the voltage were to fall to the 2.9~2.95V range, would there be serious issues with the device's operation? or would you recommend to avoid this scenario at all costs?

How tight of a demand is the voltage requirement?

3. I noticed an issue with the device (oscillation of sorts) when powering down output B, and selecting Output B's source to be High Z

at some frequencies of VCO output A will oscillate/not power on correctly. It usually happened when output A was selected to Channel Divider but it might happen with it selected to VCO too. Changing Output B's source to VCO/Channel Divider fixed this (B is still powered down, didn't matter if it was terminated 50ohm or left open).

I was wondering if this is the reason it is recommended to select VCO for the unused output? or if it is a known issue at all?

Thank you,

Alex

  • 0
    TI__Guru* 88320 points
    Hi Alex,

    (1) you were using different VCO cores, 4.5GHz core and 6GHz core. Please try again using the same VCO core. In addition, please use the same output mux (OUTx_MUX) setting for both outputs even if one of them is not used.
    (2)While I believe our part will work at 3V or even at 2.9V, this is not the supply voltage that we can guarantee to work, as shown in the datasheet. So the short answer is, please use 3.3V supply.
    (3) I think this issue is related to the output mux, we recommend to set OUTx_MUX to the same setting for both outputs if one of them is not used.
  • 0 in reply to Noel Fung
    Prodigy 70 points

    First off,

    Really impressed with the fast and on point response - Thank you!

    1 - The consumption really is just a 10mA increase for enabling the channel divider when using the same VCO core.

    It's just that the current for each VCO core is quite different ~ I will do some more testing for my application.

    However, the current difference between the devices I've tested is still the same, about 8mA. Is there any other information about the device's current consumption other than the typical value?

    It's to understand what's the maximum current to expect out of the device - to have some indication if the device is working properly.

    I'm mostly concerned with one output and typical output power -> 31 in the register, but other conditions are ok too.

    Thanks in advance,

    - Alex

  • 0 in reply to Alex OHM
    TI__Guru* 88320 points
    Hi Alex,

    I guess the two EVM were locking to different VCO core. Both VCO3 and VCO4 can support 4.5GHz.
    You can readback register R110 to see which VCO core is being selected.
    You can also select which VCO core to use. Set R20[10]=1 to enable VCO_SEL_FORCE. Set R30[13:11]=3 or 4 for VCO3 or VCO4.
    I checked the characterization data, for the same VCO core, at the same temperature, part to part current consumption variation is around 4mA.