boon Post subject: Microstip/CPW question Posted: Thu Apr 13, 2006
7:21 pm Captain Joined: Tue Feb 21, 2006 5:19 pm Posts:
17 hi which one is less lossy: microstrips or CPW's given
both are on the same substrate and the conductor metal is at the same
distance away from the substrate? some discussion will be nice. thanks
Top kpainter Post subject: Posted: Thu Apr 13, 2006
10:56 pm Colonel Joined: Wed Aug 20, 2003 11:47 am
Posts: 47 Location: Santa Barbara, CA IMHO and experience, the
CPW is the less lossy of the two. This has more to do with the fact
that the fields tend to be more contained than with Microstrip which
go to infinity on the air side. The losses are due to radiation. Dielectric
losses should be the same. Because the radiation is reduced, coupling
between lines is also reduced. If you are worried about spurs, CPW is
the way to go - that is unless you can do stripline. Microstrip
kinda sucks, otherwise folks wouldn't make antennas out of a similar
structure. Top boon Post subject: Posted: Mon
Apr 17, 2006 6:17 pm Captain Joined: Tue Feb 21, 2006
5:19 pm Posts: 17 thanks kpainter....but the field coupling to
substrate is a major issue in on-chip transmission lines and wouldn't
microstrip be better because the substrate is shielded by a ground plane
whereas in CPW the conductor plane and ground plane are all on the same
plane. boon Top kpainter Post subject:
Posted: Tue Apr 18, 2006 1:25 am Colonel Joined:
Wed Aug 20, 2003 11:47 am Posts: 47 Location: Santa Barbara, CA
Hi Boon, I assumed you were talking about grounded CPW. In this
case, there is a ground below the substrate just like microstrip. It
turns out, according to AWR's Txline program, that the calculated loss
is approximately the same for grounded CPW and microstrip when the gap
on the CPWG is the same as the transmission line width. Otherwise, the
loss of CPWG is higher - which makes sense. I am saying that in my experience,
the actual loss of microstrip is somewhat higher due to radiation. But
then, I don't recall what gap I was using but I think it was equal to
the line width. In both structures, the primary propagation path
is in the substrate, not in the air - otherwise Er would 1. I don't
know how you could build the ungrounded version of CPW in a IC??
Then again, your results may vary.
Posted 11/12/2012
|