From: C Deibele (email@example.com)
Date: Thu Feb 24 2000 - 07:06:31 PST
Dan Swanson wrote:
> Once you get down to 1 or 2% error, it can
> be really tough to sort out source of center
> freq error. You have a measurement of Eps_r,
> how about a cross section through one of
> the coupled sections. Are the dimensions
> what you think they are?
> You may also have a small error in one of
> the analytical models in Touchstone.
I have made 8 different versions of these filters - using a 95 mil
ground spacing for the stripline, and also with a 67 mil ground spacing,
and having different center frequencies. the 95 mil ground spacing has
a consistent center frequency error, while the 67 mil spacing has a
bigger center frequency error!
And the error is consistent - that is to say, that the error is always
too high by about 70 MHz for the 95 mil boards, and is around 100 MHz
for the 67 mil boards.
It isn't fair to say that the boards warp since it is teflon....floppy
flimsy teflon. The copper is 1/2 oz, and I can accurately measure up to
around 350 mils. All dimension below 350 mils seem to be right on. I
will go out and measure the lengths at a lab this afternoon.
I've heard that some heat treatment of the material can solve this
Additionally, I am hesitant to say that any trapezoidal effects are
coming into play here. I kept each trace relatively far away from other
traces (the closest is 30 mils on the 67 mil board....50 mils on the 95
Anyway, I appreciate everyone's input on this problem.
**** To unsubscribe from si-list or si-list-digest: send e-mail to firstname.lastname@example.org. In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:35:09 PDT