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.

Need PLL Eval Board with Ramping Capabilities Fast

Other Parts Discussed in Thread: LMX2492, USB2ANY

Hello,

My team needs to put in an order for a frequency synthesizer with ramping/chirping capabilities asap. I've been looking around the web and it seems like TI is the way to go for ramping capabilities at microwave frequencies.

We need a minimum of 3 frequency ranges to ramp over (roughly 2.5 GHz, 6 GHz, 9 GHz). Our ramping duration constraint is pretty relaxed at the moment since no ramping PLL's meet our 1 ns ideal requirement, as that would get into UWB microwave theory as opposed to narrowband.

Facts:

* Used LMX 2492 EVM before. Didn't have much success with it since we had to mock around with the components and decrease the frequency range to less than 1 GHz so we could make measurements with out limited capabilities. Now have access to the required RF equipment to make measurements, but our LMX 2492 board does not seem to work. We have checked the register values, reset them, loaded default settings, traced the voltage waveforms at the tuning voltage pin of the VCO, etc. If you have any suggestions on how to get it to work instead of having to buy another one, it would be very useful.

* HIGHEST PRIORITY: The ability to quickly implement the synthesizer and start testing it with minimal need to change/solder any components. We want to avoid potentially harming it like the last one. 

Thank you

  • Hello,
    Thanks for your question, i have assigned your request to an expert and he will get back to you very soon.
    Best regards
    Puneet
  • Mehrpouya,

    Although the LMX2492 supports ramping over a very wide range, it still requires an external VCO.  The VCO on the LMX2492 tunes 9.4 - 10.8 GHz and has a direct output, divide by 2, and divide by 4 (which is powered down by default).  So this means that you can ramp needs to be contained within these bands:  9.4 - 10.8 GHz, 4.7 - 5.4 GHz, or 2.35 - 2.7 GHz.  Maybe this is what you mean by "decrease the frequency range to less than 1 GHz".

    It sounds like you may have modified your EVM and something is not working now.  Here's what I would do:

    1.  Load the default mode on TICSPro

    2.  Establish communication with the device:

     The purpose of this is to ensure that you are able to communicate with the device. 

    a.  You can toggle the POWERDOWN bit in user controls tab and observe changes in current.

    b. In User controls tab, tou can set the MUXout pin to GND and toggle the MUXout-PP bit from PP to inverted PP to try to blink the LED.

    c.  You can go to USB Communications->Interface and click the "Identify" button.  It should blink the LED on the USB2ANY.  Understand that this proves communcation between the PC and the USB2ANY board, but does not prove communication between the USB2ANY board and the EVM

    3.  Toggle the SWRST bit to reset all registers and then press Ctrl+L to load all registers

    4.  Ensure that you are not trying to tune the VCO out of it's valid range.  If you want to see the full VCO range, program first to 100 MHz and you should see the lower frequency (around 9.4 GHz) and then program to 20 GHz and you should see the upper range (around 10.8 GHz).

    5.  Ensure a valid 100 MHz refernece is attached.  There might be a refenece on the board, but it is likely disconnected.

    Regards,

    Dean

  • I will try these steps as soon as I get back from Spring break. A Follow-up:

    * By decreasing the frequency to less than 1 GHz I meant we soldered our own 500 - 800 MHz custom VCO onto the board and re-routed the resistors that connected the original onboard VCO to the RFout port such that the custom VCO was then routed to the RFout. Also changed some of the resistors routing the PLL's CPout to the Vtune of the original VCO such that it connected to the new VCO's Vtune. Instead of attaching an external VCO again, do you think it would be a good idea to solder a SMA connectors onto the RFout/4 port and power it up?

    * When we loaded the default settings on the TICS PRO, we measured a peak at 4.6 GHz instead of 4.8 GHz at RFout/2. None of the subsequent changes to the register values (e.g. Reset bits) seemed to change the current output level or anything else. We thought there may be something wrong with the USB2ANY connection and the EVM just as you said. We measured the "LE" pin and various others when 5V was applied to the Vcc. The "LE" voltage level was around 1.6V if I remember correctly.

    * Finally: If these steps don't solve our problem and we end up ordering the same EVM again, we're thinking of testing RFout and RFout/2 first before trying to solder a new SMA connector to RFout/4 and "powering" it up. Do you have any other recommendations on starting fresh this way?

    Thank you so much for your fast and thorough response.
  • Mehrpouya,

    My first thought is that if you load the settings and see 4.6 instead of 4.8 GHz is that the VCO is slammed against the lower tuning range (likely around 9.2 GHz and then divide by 2). This could happen if our device is powered down, or forcing the tuning voltage to ground. If there was no 100 MHz input reference provided (as teh 100 MHz XO on the board is default not connected), then this would make sense. Or if the PLL was powered down and the device was not communicating at all,this would make sense.

    So to get through the communication step, try these things:
    1. Blink the LED on the USB2ANY using the USB Communications->USB Interface ->Identify from the menu. Make sure that in this interface window that it shows your USB board and is not in Demo mode. Also, on the bottom of ticspro it shows "connection mode". If you see "device not connected" in red, then this is likely your issue.

    2. If you can verify that you are communicating with the USB2ANY, but the LMX2492 EVM seems unresponsive, then the powerup/powerdown test is next. Aside from checking the current, the DC bias level on the OSCin pin is typically 1.6V when powered up and 0V when powered down. This can be used to check powerup/powerdown status.

    3. In general, it is always the best practice to get the board up and running in default mode before attempting any modificaitons; I have seen far too many times when this was done and it makes debugging harder.

    4. For your modifications, if the intention is to use your lower frequency VCO:
    Disconnect L4 to power down upper VCO. L5 should be disconnected (this powerd 1/4 divider). Connect R14 to power lower VCO. Disconnect R6 and Connect R5 to hook up the feedback. Disconnect R4_LF from main VCO and connect R4_LF to hook up tuning votlage to lower VCO.

    Regards,
    Dean