18LF4620 port D woes!

BLL · Joined: 11 Nov 2006 Posts: 181 Location: Birmingham, UK

Hi,
I am using a 18LF4620 with 10MHz crystal, and H4.
I am trying to use port D2 and D3 as digital inputs.
I have setup_PSP(PSP_DISABLED); at the start of main, as these pins are shown as RD2/PSP2 and RD3/PSP3.

D2 works fine but D3 reads as zero, regardless of the input state.

Can someone explain what I am doing wrong, as the data sheet says that all port D pins can be configured as digital inputs?

Thanks

Brian

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

Those are completely ordinary i/o pins. If pin D3 reads at a logic high
level with a scope and the PIC sees it as low, you probably have a
defective PIC.

asmboy · Joined: 20 Nov 2007 Posts: 2128 Location: albany ny

if a new PIC does not cure it - you need to show the relevant code you created.

how the pins were inited - and where- how you read them

temtronic · Posted: Fri Jul 15, 2016 2:18 pm

might also be a short on your PCB, grounding the pin ?
check with power off, no PIC and ring it out

Jay

BLL · Joined: 11 Nov 2006 Posts: 181 Location: Birmingham, UK

Hi all, Thanks for the replies. The PIC is on a prototype board and there is no short. 2 other, brand new PICs behave the same. Each pin has a 10K resistor to +5V and a connector which allows either pin to be grounded by a sensor switch.
All the code does is to check the logic state of each pin:

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

Your code is overly complex for the problem, and your hardware sounds
overly complex as well.

Switch circuit

Ttelmah · Joined: 11 Mar 2010 Posts: 19539

Take a deep breath, and start again...

First thing is a simple statement of fact. There is nothing special about these pins, except the PSP.

So first thing to do, is to genuinely 'prove' you do not have a short (it's amazingly easy to have whiskers or tiny solder blobs). Assuming the PIC is socketed, unplug it, and test if the pins now go up/down.

Then assuming they do, you come to the problem being in your code. Look for:
1) Any TRIS statements.
2) Any fast_io/fixed_io statements.
3) Any peripheral setup referring to the pins (could be something silly like a missed line in your LCD code). #USE lines etc..

Always the thing to do, is simplify. As with the test for the lines actually going up/down, reduce code till the problem can be isolated. PCM_Programmer gives a good example of a basic 'test' routine to start from

BLL · Joined: 11 Nov 2006 Posts: 181 Location: Birmingham, UK

Hi both, thanks for the input.

I eventually traced it to where, just before reading the pins, I had a strcat statement instead of a strcpy to put the first item in a string!!
Putting that right made everything OK.

Thanks for the help.

Brian