bit bashing with SUART module-not reading port - SOLVED

[RadioT]

Hello,

I decided to give using a free port pin a go on my 18F87j11, since the serial ports are used up. I am putting out a GPS signal to port J7 directly, with no RS232 level shifter. I have run the same GPS output for several years on a EUSART port.

I'm running the sample SUART demo program that comes with SF. For whatever reason, I just can't seem to get any data into the port. Set up RX as j.7 and TX as j.6. The signal is 2V from low to high, although I should mention that with this GPS module I have to use a pullup resistor. I have the trace on a scope, though, and the transisions look good.

I don't suppose there have been any updates to this module? On a more complex version that is able to compile in MPLab (the small version can't for whatever reason) I see the "read" function enters the suart module and never is happy it has seen the FTerminator byte.

Anybody run into this problem before?

73's,

de Tom

[RadioT]

Now SOLVED

Things I have found as I worked through examples:

1. don't forget to set OSCCON to vary speed...."clock = xx" only tells the compiler what speed to use in the rate calculations.

2. transmitting a string works fine.

So far, have code:

Code: Select all

Device = 18F87J11
Clock = 32

Include "SUART.bas"
Include "convert.bas"

// Set speed to internal 8 MHz clock
OSCCON.6 = 1
OSCCON.5 = 1
OSCCON.4 = 1

// Multiply speed x 4 = 32 MHz
OSCTUNE.6 = 1

//Set input and output I/O pins
SetRX(PORTJ.7)
SetTX(PORTE.1)
SetMode(umTrue)

// clear off anything in a pin register, to be sure, as per Microchip datasheet I/O section
Asm
clrf PORTJ
clrf LATJ
End Asm

//Set up a Test Loop for transmit

Whle true

   UART.Write("Bitbanging",13,10)

Wend

So all this gives a nice output waveform.

To see if the Schmidt Trigger input port I'm using is fine, substituted the following in the While loop:

Code: Select all

Whle true

   If PORTJ.7 = 1 then PORTE.0 = 1 
   ElseIf PORTJ.7 = 0 then PORT.0 = 0 
   Endif

Wend

The result is what appears to be a perfect duplicate waveform on the output, as expected. So the Schmidt Trigger is not a problem.

So now the While loop is what it was in the original Sample example:

Code: Select all

Dim Value as string //added the string array for holding the data read

Whle true

   UART.Read(Value)
   UART.Write(Value,13,10)

Wend

What did, in the end, fix the problem was creating a new SUART.BAS library file. SF changed the ASM directives to uppercase in this process. That is the only difference between the old and the new version of SUART.BAS that I made. Perplexing but true. Now the I/O port (in this case, PORTJ.7) reads the GPS strings perfectly.

So I have no idea why case would make a difference in a library file. Maybe it is just a quirk in my system. It is a brand-new Windows 7 box with the latest-and-greatest SF version 2.2.1.4, ICC 1.1.5.7 that I bought to ensure it worked in the shiny new box (time is money - why spend hours tweaking Windows 7 to use an old SF version when the new one installs in 2 minutes??). In this version, the executables are split from the user data and library files, with the executables in C:/Program Files (x86)/Mecanique/Swordfish and the user files in the linked directory Libraries/Documetns/Swordfish, with the actual path being C:/Users/Tom/Documents/Swordfish.

73's,

de Tom