Steven Nickels

MC13213 single-port, internal T/R switch - does it work?

Discussion created by Steven Nickels on Jun 5, 2007
Latest reply on Aug 1, 2007 by David Horst
Anyone successfully implement the single-port, internal T/R switch for the MC13213 as described in AN3248?  I saw a related post "BeeKit User Defined Target Question", but the discussion seemed to wander.
My code, based on the WirelessUart example (in BeeKit), works on the MC13213SRB, but it won't run on my custom board - uses the single-port RF with internal T/R switch design.  I have modified the MC21XSRB.c file (per AN3248) so that register 0x07 is loaded with 0x5C00 instead.  This should be the only change.
I don't have a spectrum analyzer to test if RF is being emitted.  My circuit isn't exactly as described in the app. note.  It has a single 6.8nH inductor across the RF lines to the left of the balun, and then an 8.2nH inductor on the right side.  I've attached the schematic.  It's based on some other reference design (someone else created board for me).  The actual part on the board is "PC13213", so perhaps this pre-production part had an issue?  I have looked at the CT_BIAS pin using a scope and it is correctly goes to VDDA during transmit.  I see some pulses on the RFIN_P and RFIN_M pins at the time of the transmit.  For kicks I placed the scope on the othe side of the balun, but didn't see anything.  There's no mention of RF problems in the errata. 
All of Freescale's development boards for the MC1321X use differential output, so I would like a little proof that it works on real hardware.  My next step is to hotwire it to exactly match AN3248.  That'll involve some microscope work so it'd be good to hear "it worked for me" before I go through the effort.