The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> LNA I output+Mixer I input.Need match?
https://designers-guide.org/forum/YaBB.pl?num=1094816014

Message started by wccheng on Sep 10^th, 2004, 4:33am

Title: LNA I output+Mixer I input.Need match?
Post by wccheng on Sep 10^th, 2004, 4:33am

Dear all,

The LNA output is current output (output is from the drain of the transistor) and this current output will directly connect to the current input of the mixer (This is the connection point which is connected with the drain of the pmos current source at the upper and the drain of the communtating stage). Actually, I do not do conjungate matching between the LNA and Mixer because it is current conneciton. Now, I don't know how could I explain why do I need not to do conjungate match. PLEASE help me..... :-[ ???

Because traditional LNA is voltage output and it will connect to the gate of the transconductance transistor in the mixer. This is possible to do the matching network between these two blocks and give out the required bias at the same time.

HELP ME, THANKS

wccheng

Title: Re: LNA I output+Mixer I input.Need match?
Post by xuebi on Nov 18^th, 2005, 9:12pm

I'm also confused about interstage matching in RFIC design.

In traditional microwave circuit design, the interstage matching should be conjugate matching, while in RFIC design, is it necessary?

Thanks.

Title: Re: LNA I output+Mixer I input.Need match?
Post by rfzingle on Nov 23^rd, 2005, 12:46pm

For on chip RFIC design , one does not need to impedance match. The reasons are as follows...

1. The dimensions involved for on -chip are not comparable to those off-chip . Hence the transmission line effects are not dominant and a lumped model can be used. In other words what it translates into is microwave power is not reflected across if the on-chip lines are not terminated (as it would have had the dimensions been approaching lambda/10). Thus microwave power is not entirely lost if matching is not done. Under these circumstance what one needs to do is voltage transfer instead of power transfer. Thus if you are passing voltage, you need to make your source a very low impedance voltage source and your target a very high impedance load. And viceversa for current transfer.

2. Another reason which negates requirement for impedance matching on-chip is availibility of MOS devices which have significant power gain for the RF frequencies of operation. Hence microwave power is not entirely a scare quanity given the MOS devices in advanced processes like 0.18u or so. In other words make sure that there is atleast 10-15x ratio between Fmax and frf (the RF frequency of operation). That way even if some microwave power is lost due to impedance mismatch, the device will gain it back.

I hope this adequately answers this. Any more takers on this topic...

rfzingle