The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> current-mode versus voltage-mode passive mixer
https://designers-guide.org/forum/YaBB.pl?num=1326110461

Message started by aaron_do on Jan 9^th, 2012, 4:01am

Title: current-mode versus voltage-mode passive mixer
Post by aaron_do on Jan 9^th, 2012, 4:01am

Hi all,

I would like some opinions on which is better: voltage-mode or current-mode passive mixers. For voltage-mode passive mixers, the switching core is followed by a high impedance, while for a current-mode passive mixer, the switching core is followed by a low impedance. I will start with a few.

voltage mode
=========
1) Smaller devices means less driving strength needed by frequency synthesizers.
2) Higher gain before baseband amplification. Potentially less noisy?
3) Easily coupled with Gm-C type filtering for low noise, low power, and wide bandwidth.

current-mode
-=========
1) Better linearity due to low voltage across the switches and the LNA.
2) Easier to get a well defined gain as the load impedance is easier to define at baseband.
3) Smaller resistance in series with the signal path. Potentially less noisy?
4) Easily coupled with op-amp based filtering for high linearity.

Overall, my opinion is that a current-mode passive mixer will get you a more robust system, however, I believe potentially the voltage-mode passive mixer has better NF. I have been assured by a very experienced designer that all real systems use current-mode passive mixing. What are your thoughts?

thanks,
Aaron

Title: Re: current-mode versus voltage-mode passive mixer
Post by RFICDUDE on Jan 10^th, 2012, 4:30am

Hi Aaron,

I concur that the majority of passive mixer implementations are current mode; although, I've seen a few (very few) voltage mode implementations too.

The noise performance comparison is a bit fuzzy because high voltage gain before the mixer means there is high voltage gain at high frequency and that causes design issues for the amplifier driving the mixer (head room, current consumption and maybe stability). For current mode, you can still have fairly high transconductance before the mixer and the voltage swings are smaller which makes the amplifier and mixer interface more straight forward to deal with. The noise of the mixer (I suppose) looks like another noise current source added to the transconductance noise, so as long as the mixer resistance is fairly high compared to 1/gm then the added noise should be small (I think).

Linearity appears to be more difficult to deal with in voltage mode (in my opinion) than current mode because of the different voltage swings on both sides of the mixer devices.

I'm interest in what others have to say.

Title: Re: current-mode versus voltage-mode passive mixer
Post by aaron_do on Jan 10^th, 2012, 5:43pm

Hi RFICDUDE,

thanks for the reply. I have a feeling that you are mistaken in the way you model the noise of the current-mode mixer. I believe that the mixer stage is treated as a resistor in series with the signal path. Therefore, the smaller the resistance, the better. Correct me if I'm wrong here.

Actually, I've beed doing a little analysis and I'm starting to have doubts about the voltage-mode passive mixer. I will post it if I have the time.

regards,
Aaron

Title: Re: current-mode versus voltage-mode passive mixer
Post by RFICDUDE on Jan 11^th, 2012, 5:53pm

I was thinking that if this is a current mode receive mixer then the output side of the mixer is connected to a low impedance to sink the current. If that is the case then the noise contribution of the mixer switch is primarily current into the sink. Of course, it doesn't really matter which way you look at it, the answer should be the same in the final analytical solution. I didn't go through the analysis, so my guess at impact may be in error.