TI SimpleLink CC2640R2 LaunchPad


#1

I’ve created a part for the new TI SimpleLink CC2640R2 Launchpad. Note that PCB pin alignment might be very slightly off but otherwise hopefully all ok.

TI SimpleLink BLE CC2640R2 LaunchPad.fzpz (160.3 KB)


#2

Looks quite good, a couple of problems though. On pcb your hole size is .036, should be .038 for .1 pins (or .035 for IC pins). It is desirable to have an outline of the board in pcb view so the user can see where the board will be taking up space to place components around it on the pcb. In the schematic svg you have the terminal ids defined (except for pin1 which you need to add) but they aren’t in the fzp file (assuming you are using Parts Editor, which I can never get to work for me and thus don’t use, you haven’t set the north/south/east/west setting on the pins). The lack of the terminal id causes a line in schematic to connect to the middle of the pin definition. I manually added the terminal definition to the fzp file in the part in the sketch below to show how it should look (the top pin) against how yours looks (the second pin). As well connectors should start at 0, not 1 as yours do, but it won’t hurt anything in particular.

fixed.fzz (162.1 KB)

Peter


#3

It’s also better to have silkscreen above copper, because sometimes it hard to select pads for assignment in FZ when it’s in a layer under another layer.


#4

Thanks for all the very useful tips.

Yes, the PCB is a rather ropey. I did not realise pin hole size was out as I thought I had used a standard terminal block as a basis. Yes, did not pay attention to silkscreen layer position. Sorry about that.

Thanks for the schematic review and spotting that pin 1 terminal definition is missing. I will look at that . Say, regarding N/W/S/E settings on the pins, how does one define this within the SVG file or does this have to be done within parts editor.

Yes, I know I was being a little deviant not starting connectors at 0, but was simply easier for me to map with documentation when I started at 1. But then again I’ve seen some Sparkfun parts starting at number 23!


#5

The silkscreen layer position is useful to me, I’m working on a python script to check parts and I wasn’t aware that silkscreen after copper was an issue, so one more thing to check for and complain about :slight_smile: . The easy way to check hole size is to create the gerbers (file->export for production->gerber) and check the filename.drill file:

; NON-PLATED HOLES START AT T1
; THROUGH (PLATED) HOLES START AT T100
M48
INCH
T1C0.039370
T100C0.019370

I don’t remember what part this is (and the holes are odd) but the first is non plated through .039 in (probably a mounting hole) and the next is a small (likely too small unless it is a via) plated through hole. Changing them can be exciting in Inkscape because it changes what xml editor uses as units on some criteria I don’t understand. If it has decided to use thousands of an inch, its fairly easy as it is diameter of the pad - 2 * stroke width. When it uses odd units, I usually copy / paste a pad of the correct size from another svg and then use the radius and stroke width from that (so the x/y coords don’t change) to update the pads in xml editor.

To make the change in the files (which some of the folks here don’t like doing, but I do :slight_smile: ) you need to unzip the part’s fzpz file and edit the fpz file (I use 7zip for this)

part.CC2640R2_f85d57a2103706491299fad882ddd6aa_7.fzp

in the case of your part. This is an xml file that describes the part and the svg files (the format is described in the pat file format document). In this case you need to add a terminalId as I did on connector2 to all the rest:

    <schematicView>
     <p svgId="connector1pin" layer="schematic"/>
    </schematicView>
...

    <schematicView>
       <p svgId="connector2pin" layer="schematic" terminalId="connector2terminal"/>
    </schematicView>

The danger is that if you screw up the xml, there is currently a bug in Fritzing (which I hope to be able to fix once I finish the python script) that will cause Fritzing to hang due to a seg fault. When it does so you need to clear the user directories (which deletes all your sketches so you need to make sure you have backups first!) to recover from it as it has corrupted the parts database and not backed out due to the seg fault.

Yep, it doesn’t appear to be an absolute requirement that parts start at 0, although the parts format document calls for it. As long as the pins are in sequence Fritzing looks to be happy. There is a problem if the pins aren’t in sequence though as if there is a gap the descriptions will screw up (and be wrong when you hover over a pin in the views after the discontinuity). How I typically make parts (without parts editor) is documented here if you haven’t seen it:

Peter