--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/pirelli/undersim	Fri Apr 19 21:31:47 2013 +0000
@@ -0,0 +1,36 @@
+There are 4 test points accessible from the battery compartment under the
+SIM card slot.  I shall number them as follows:
+
+	B
+	A (battery connector)
+	T
+
++-----+
+|     |
+| SIM |
+|     |
++-----+
+1 2 3 4
+
+i.e., the test point closest to the headset jack shall be numbered 4.
+
+TP 1 and TP 2 have tiny vias inside the TP pads themselves; these vias
+appear to go no deeper than L2.
+
+TP 1: on L2 it crosses over the tiny via coming from L1 to a larger via
+that penetrates almost the entire layer stack.  On L7 the latter via
+connects to a trace.  The latter trace runs at normal signal thickness
+until it hits yet another via.  On the other side of that via, still on
+the same L7, there is a dead-end arm of what appears to be the same net,
+and the latter arm is thick - almost as if it acts as an antenna of
+some kind.  From there the connections are unclear, but it appears to
+connect a bunch of fat traces and possibly even copper fill on other
+layers, so it's unlikely to be an interesting signal.
+
+TP 2: on L2 it goes to a trace, that trace goes to another via at image
+coords (4378,1746).
+
+TP 3 connects to a via; that via appears to make a solid connection to
+the copper flood-fill on L6.
+
+TP 4 is connected to the L1 GND copper fill with thermals.

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/pirelli/usb	Fri Apr 19 21:31:47 2013 +0000
@@ -0,0 +1,7 @@
+All PCB trace connections to the USB connector are made on the outer layer
+on the battery/SIM side of the PCB; none on the display/keypad side or on any
+of the inner layers.
+
+The L1 copper image from steve-m shows that the ID pin appears to be totally
+unconnected.  The GND, VBUS and data pair connections appear the way one would
+expect.