I have downloaded the SPICE model for the Fully Differential ADC driver ADA4945-1 and have added the basic components as can be seen in the screenshots. However, the frequency response rolls off after 1MHz whereas the bandwidth in the datasheet mentioned is considerably higher. I tried changing the gains but the issue remains the same. I have attached the screenshots of the basic circuit along with the frequency response. The two curves are of the two differential outputs. Can someone please explain why the SPICE model is not behaving like the datasheet?
Couple of things:
1. You cannot run the ADA4945 off 20V supplies (+/-10V) as you have it configured. Change your supplies to +/-5V instead.
2. You have left the "Mode" tied to ground. You should…
2. You have left the "Mode" tied to ground. You should either tie it V+ or V- depending on the operating mode you like.
See if the attached simulation file works for you?
ADA4945 LTspice simulation 8_8_19.zip
When I use your LTSpice simulation (with no changes) I get a good frequency response on the P output (-3dB @ >100MHz) but a poor one on the N output. (-3dB @ 27MHz)
This also occurs if I update to the latest V1.1 spice model of the ADA4945-1.
What am I doing wrong, or is this a true representation of the device?
I am looking for a single ended to differential driver that is flat to 50MHz (preferably with the +ve and -ve clamp feature)
I'm out of the office right now but I've asked a colleague to take a look at your issue with ADA4945-1 spice model.
Hopefully you'll hear from him tomorrow.
Hi Clystron,The part can handle single-ended to differential operation as seen on the datasheet, the model is not that properly modeled at single ended operation but works well when the inputs are balanced.See screenshot below.