The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Phase shift of mixer
https://designers-guide.org/forum/YaBB.pl?num=1377523524

Message started by Chuck on Aug 26^th, 2013, 6:25am

Title: Phase shift of mixer
Post by Chuck on Aug 26^th, 2013, 6:25am

Hi,

I am going to design a circuit and it's important for me to know the relation between the phase of the transmitted signal, and the received one.
I know what the phase of the original transmitted signal is, but after using mixer and down converting the received signal to an IF frequency, I think the phase of the signal would be changed due to the filter used in the mixer. I wonder if what I think is correct, and if so, is there any way to solve this problem?

Title: Re: Phase shift of mixer
Post by aaron_do on Aug 26^th, 2013, 8:35am

Hi,

what do you mean by "the filter used in the mixer"? Are you deliberately adding a filter, or do you mean just some parasitic RC? Anyway a bandpass filter can be designed to have zero phase shift at the center frequency. Likewise if you mix the signal down to zero IF, the phase shift is zero at DC. I guess its possible (in theory anyway) to design the receiver with zero phase shift at one frequency. The wider your bandwidth, the less the phase variation over frequency.

regards,
Aaron

Title: Re: Phase shift of mixer
Post by Chuck on Aug 26^th, 2013, 8:58am

Thanks for the reply. Say the RF frequency is at 200MHz and the local oscillator is at 250MHz. So the down converted signal would be at 50M, now I need to use a low pass filter to get rid of the higher frequencies. What I did was simulating this system in Matlab and I saw that due to the filter response, there was phase shift between the filtered signal at IF frequency and the original signal. But I need to know the exact phase of the original signal, and I wonder whether phase of the mixed signal would be preserved or not.

Title: Re: Phase shift of mixer
Post by aaron_do on Aug 26^th, 2013, 5:49pm

Hi,

it depends on the bandwidth of your low-pass filter. Definitely the phase shift is not going to be zero, but if the bandwidth is wide enough, then the phase shift will be minimal.

In your case, with a 50 MHz IF, your LPF is probably going to cause considerable phase shift. Off the top of my head you could:

1. Use a bandpass filter instead.

2. Digitize the signal first and use a digital filter. You will need a pretty high sampling rate so that your anti-aliasing filter doesn't cause any phase shift.

3. Choose a lower IF?

When you say you need the exact phase, I assume you mean w.r.t the transmitted signal right? If you describe your application a bit you will probably get better responses.

regards,
Aaron