Part Number: LMX2492
Hello,
I have been working on creating ramps on a customized board which has an LMX2492 in the design. Initial experiments with Ramps with 5-6 segments have been working. But now we came across a problem.
First, the good case. This has 5 segments, initially starts at a low frequency of 4950 MHz and stays flat, then moves up to the high frequency of 5130 MHz, stays flat for a while and then back down to the start frequency of 4950 MHz. This works fine. The ramp sequence takes exactly 5 msec as desired. The TICSPro Ramp and PLL settings are attached as a screenshot and also the Time vs Frequency plot of the Ramp (please note: the frequency is scaled down by a factor of 208 and centred around 25 MHz). The frequencies and time measurements are perfect as below.
Now, the bad case. This has 6 segments, initially starts at a high frequency of 5130 MHz and stays flat, then moves down to the low frequency of 4950 MHz, stays flat for a while and then back up to the start frequency of 5130 MHz. Note that the frequencies and total time of the intended ramp pattern are same as before but we start from the high frequency this time. This does not work properly. The overall shape is okay but the frequencies achieved in the ramp are not correct and do not match what we intended. Also, the whole sequence is supposed to take 5 msec but here it is taking more like 6 msec. The TICSPro Ramp and PLL settings are attached as a screenshot and also the Time vs Frequency plot of the Ramp, as with the earlier case.
Sorry for all the details, but this is to illustrate that we have a working ramp but just changing whether we go low -> high -> low versus high -> low -> high is affecting the result: both frequencies and time.
Thanks very much!
Arun
