Problem with comparator

Hello Dorene,

PF0 is a locked pin. The following post will help you unlock it and use the comparator output

http://e2e.ti.com/support/microcontrollers/tiva_arm/f/908/t/284566.aspx

http://e2e.ti.com/support/microcontrollers/tiva_arm/f/908/t/351475.aspx

Regards

Amit

Hi,

    See, the link at my signature below "TM4C123 NMI Unlock. Also, this sticky below by Sue.

    Common GPIO Problem

-kel

I tried to run my codes  in TIVABOARD TM4C123GH6PM but seem nothing show at the output PF0. Anyone can guide me where is my problem?

Here is my code :

Hi Dorene,

dorene jong said:

   SysCtlPeripheralEnable(SYSCTL_PERIPH_GPIOF);

   GPIODirModeSet(GPIO_PORTF_BASE, 0x01, GPIO_DIR_MODE_OUT);

   GPIOPadConfigSet(GPIO_PORTF_BASE, 0x01,GPIO_STRENGTH_2MA,GPIO_PIN_TYPE_STD);

   HWREG(GPIO_PORTF_BASE + GPIO_O_LOCK) = GPIO_LOCK_KEY;

   HWREG(GPIO_PORTF_BASE + GPIO_O_CR) |= 0x01;

   HWREG(GPIO_PORTF_BASE + GPIO_O_AFSEL) |= 0x400;

   HWREG(GPIO_PORTF_BASE + GPIO_O_DEN) |= 0x01;

   HWREG(GPIO_PORTF_BASE + GPIO_O_LOCK) = 0;

  GPIOPinTypeGPIOOutput (GPIO_PORTF_BASE,GPIO_PIN_1|GPIO_PIN_2|GPIO_PIN_3);

     Do you only want to output at PF0? What do you need PF1, PF2, and PF3. Try first outputting at PF0 first without the other code.If you are not familiar how to output at GPIO pins, see GPIO example program at Tivaware Peripheral Library Users Guide.

     Regarding your PF0 unlocking code. Try this below. I have not tested this code below.

SysCtlPeripheralEnable(SYSCTL_PERIPH_GPIOF);

HWREG(GPIO_PORTF_BASE + GPIO_O_LOCK) = GPIO_LOCK_KEY;
HWREG(GPIO_PORTF_BASE + GPIO_O_CR) |= 0x01;
HWREG(GPIO_PORTF_BASE + GPIO_O_LOCK) = 0;

GPIOPinTypeGPIOOutput (GPIO_PORTF_BASE, GPIO_PIN_0);

GPIOPinWrite(GPIO_PORTF_BASE, GPIO_PIN_0, GPIO_PIN_0); // This will output a logic 1 to PF0

-kel

Poster stated her issue was deficient Comparator Output.  Thus - forcing PF0 into gpio_output is not the answer.

Instead - PF0 must be configured as an analog comparator output.  Code below details:  (our usage is more general purpose {i.e. more duplicative of past LM339 comparator's operation - i.e. no internal Vref})

As helper states - that pin must be first unlocked & poster appears to have understood/complied.  Yet - set-up/config of PF0 as comparator output does not agree w/our coding.  (ours is earlier LX4F MCU)  Below are key elements of our code - but not showing the full set-up/config of PF0.  (manual review indicated there)

ComparatorConfigure(COMP_BASE, 0, (COMP_TRIG_NONE |    // we feed analog signals to both comp. inputs                     COMP_ASRCP_PIN0 | COMP_OUTPUT_NORMAL));

GPIOPinConfigure(GPIO_PF0_C0O); //  comparator Output pin

GPIOPinTypeComparator(GPIO_PORTC_BASE, GPIO_PIN_6 | GPIO_PIN_7); // comparator input pins

With the code shown (above, here) the analog comparator will drive its output (PF0) "high" when the voltage upon the Comp + pin exceeds the voltage upon the Comp - pin.  We did not employ the internal reference settings as they were too coarse for our usage - and we wanted to be able to "dynamically" change the comparator's set-point via SW - on the fly.

Code the poster provides (latest version) does not detail all of the set-up/config of the comparator pins.

Amit Ashara said:

The issue is in the AFSEL and missing PCTL

Let the record show that the inclusion of function: "GPIOPinConfigure()" (as earlier suggested by this reporter) nicely encapsulates the "GPIO_O_PCTL" requirement.  Following is a true copy of the ending code w/in GPIOPinConfigure().

// Write the requested pin muxing value for this GPIO pin.
   
HWREG(ulBase + GPIO_O_PCTL) = ((HWREG(ulBase + GPIO_O_PCTL) &
                                    ~(0xf << ulShift)) |
                                   ((ulPinConfig & 0xf) << ulShift));

In our long working, analog comparator code, we set both DEN and AFSEL = 1, too.

One would think (surely would hope) that the "normal" library functions could manage such fine detail.  (otherwise - direct register may return to, "rule the earth.")

Might the "real issue" here be the unexplained/unexpected absence of a "real/valid" GPIOPinTypeComparatorOutput() function?  One notes that, "GPIOPinTypeComparator()" is present - and strangely - may only be applied to those comparator pins serving as Inputs.   Strange that, indeed!  And necessitates the "cobble" of GPIO__PCTL & friends..."

Hello cb1,

Yes the function encapsulates the PCTL and for the record, GPIOPinTypeComparatorOutput is what is needed.

Regards

Amit

cb1- said:

unexplained/unexpected absence of a "real/valid" GPIOPinTypeComparatorOutput() function

Thanks & pleased to see that we're in agreement, Amit.  You of course are, "blameless."

I sought - via my post - to be first to raise the flag so that future lib. revisions could add this, "strangely absent" (yet fairly complex),  "Analog Comparator Output" missing piece.  Trust you'll, "bang the drum" so that this moves upstream...