LMX2595: What is the maximum slew rate of a calibration-free automatic ramp?

JackM

Part Number: LMX2595

Tool/software:

I am trying to generate an automatic ramp signal between 10.2 and 10.25 GHz over 5 microseconds with the LMX2595EVM. However, when I configure the ramp parameters for this waveform, I am not able to generate the desired ramp signal. However, if I change the ramp time from 5 us to 20 us, I am able to generate the desired signal. What is the maximum ramp slew rate for the LMX2595? It looks like 50 MHz over 5 us (10 MHz/us) exceeds this limit, while 50 MHz over 20 us (2.5 MHz/us) works.

Would it be possible to achieve a 50 MHz over 5 us (10MHz/us) ramp with the LMX2595 if I increase the bandwidth of the loop filter?

10 months ago

0 Connor Lewis 10 months ago

TI__Genius 13180 points

Hello,

Our expert for this device is out of office today. I've assigned this thread to them, please expect a response tomorrow.

Regards,

Connor

0 Noel Fung 10 months ago in reply to Connor Lewis

TI__Guru** 110700 points

Hi Jack,

What is your reference clock and fpd frequency? What is your loop bandwidth?

0 JackM 10 months ago in reply to Noel Fung

Prodigy 10 points

I was doing this with the LMX2595 evaluation board. My ref clock was 100 MHz and my fpd frequency was also 100 MHz. The loop bandwidth is whatever the default is for the LMX2595EVM.

Here are the PLL and Ramp settings I used.

0 Kadeem Samuel 10 months ago in reply to JackM

TI__Mastermind 23000 points

Hello,

The team is out for the holidays. Please expect a response by Jan 2 at the earliest.

Thanks,

Kadeem

0 Noel Fung 10 months ago in reply to Kadeem Samuel

TI__Guru** 110700 points

Hi Jack,

Looks to me that this could be a loop bandwidth/response issue.

I am not able to make successful ramp up and down when the ramp duration are both 5µs.

However, if I set ramp up time = 10µs while keeping the ramp down time to 5µs, I can make it ramp robustly.

From the plot, we can see that even though ramp up time is 10µs, the trace is not a straight line. Compared with the ramp down trace, it is more or less a straight line and the duration is only 5µs. This behavior tells us that the synthesizer's response time is not fast enough to ramp up in 5µs.

I believe, if we increase the loop bandwidth, we may be able to reduce ramp up response time. You may also need to use C0G/NP0 or film capacitor in the loop filter in order to get a faster charging/discharging time.

0 Dean Banerjee 10 months ago in reply to Noel Fung

TI__Mastermind 28021 points

Jack,

We also have this article app note discussing the slew rate limitations with loop bandwidth

Clock & timing

Clock & timing forum

LMX2595: What is the maximum slew rate of a calibration-free automatic ramp?