Stellaris® ARM® Microcontrollers

Forum Post: RE: RDK-BLDC current measuring frequency in trapezoid modulation

cb1_mobile — Thu, 23 May 2013 22:26:00 GMT

Yours are - imho - very good questions...

ADC0IntHandler "appears" to be called upon the ADC Sequence Step which includes the parameter, "ADC_CTL_IE." And interestingly (at least to our group) you may interweave ADC_CTL_IE throughout your Step Sequence. I used the weasel word "appears" earlier - as the exact moment of conversion is never really described - or defies my best efforts to discover. Now the PWM Generator "triggers" the ADC - but the critical PSSI Register (which monitors & controls that process) is "write only" - so the exact moment of that "trigger" is unknown by mere mortals. Should you set a break point early w/in ADC0IntHandler() - you'll arrive - but unknown if this is before or after an ADC conversion sequence has occurred. (what I think is needed is a better (i.e. "some") definition of ADC trigger!) That trigger may vector the MCU to the interrupt handler - or may induce the ADC to perform a conversion sequence - and the presence of the ADC_CTL_IE will then cause the ADC0IntHandler to be entered.

Frequency of current readings - look more closely at the "lead code" w/in ADC0IntTrap. You'll note that "only" upon the hall sensor's signalling a "change in low-side drive" will a "phase-current calculation" be enabled! Know that you're doubtful - here's the key code extract:

    // If we have changed phases, calculate the phase current average.
    //
    if (g_ucPhaseCurrentIndex != g_ucPreviousPhaseCurrentIndex)
    {

Now if that ADC's Sequence Step has been directed to the current sense resistor(s) - current will be measured upon "most" entries w/in ADC0IntTrap(). Why "most" - look here:

if(ulPWMEnable != ulLastPWMEnable) // atop ADC0IntTrap() ****

{

and when satisfied - this "if" will "return" - without performing any of the ADC's Sequence Step Conversions. Which is not that "big" of a deal - when ADC0IntTrap() is being hit @ 50uS (20KHz) intervals - and Hall frequency is a pittance of that...

Forum Post: RE: Systick Clock Source

cb1_mobile — Thu, 23 May 2013 22:03:00 GMT

[quote user="Donald Varela"]PIOSC clock is around 16MHZ, I'd like to get the actual frequency.[/quote]

Again - if the MCU Timers will accept PIOSC as their input clock - you can cause that clock signal to be output on a timer pin.

Forum Post: RE: SSI configuration on LM3S6C65

cb1_mobile — Thu, 23 May 2013 21:58:00 GMT

Wise to review SW-DRL-UGxxxx - especially restriction upon GPIOPinConfigure().

Note:
This function is not valid on Sandstorm, Fury, and Dustdevil-class devices. Also, if the same signal is assigned to two different GPIO port pins, the signal is assigned to the port with the lowest letter and the assignment to the higher letter port is ignored.

I know the "class" of many Stellaris - but not yours. Somewhere - on vendor's vast web-site - there is a guide. (I've asked - several times - that this be made more findable)

Again - I'd (temporarily) switch your candidate SSI pins to GPIO Output - and see if you can drive each one high. If not possible - we'll spin wheels forever w/SSI...

Forum Post: RE: I^2C Sending to all registers? "MAX44009"

cb1_mobile — Thu, 23 May 2013 21:50:00 GMT

[quote user="Thomas henriksen"]We have allready tested it with another MCU, which yielded the same results. [/quote]

Sorry - you're so close to the issue - know just what you mean - me - not so much... Does "same results" mean same failed - or same success? If I get a chance this evening will review Slave's data manual - see if anything "pops out." Again - have you ever gotten that Slave device to respond correctly?

Feel your pain - but have no explanation for "each/every" register being loaded by a single I2C write - targeting just one register. (if I fully understood your writing...)

If you could capture & display both SDA & SCL - during the simplest MCU write and answering, Slave response - commentary may improve...

Forum Post: RE: LM3S9D90 Ethernet Phy oscillator not running on a small number of production boards.

Stellaris Dexter — Thu, 23 May 2013 21:07:00 GMT

20 out of how many production?

Are you getting the parts through distribution? Are you in contact with a local distributor or TI FAE?

I'd still like to see some more testing regarding the crystals on the "bad" boards/chips. Make sure they are clean and free of flux especially under the crystal. Also make sure the caps are soldered well and of the right values.

I am trying to eliminate the chance that something else on the board is slightly outside of margin. Then combined with normal margin on our parts the system fails. It's possible that are parts are within margin but some parts work and some don't given the margin and variation in the rest of the system.

What are the caps on the crystal and what is the crystal part number?

Dexter

Forum Post: RE: HID + HID Interface on DK-LM3S9B96 Stellaris

Vinicius Hubner — Thu, 23 May 2013 20:45:00 GMT

I am trying to modify the example USB_DEV_CAUDIOHID, but I can not recognize 2 keyboards when I connect to the computer.
The example consists of 1 audio interface and 1 HID keyboard..

i need to recognize 2 keyboards.. I'm already make changes on usb_structs.h and code, but when DK connects on computer recognize 1 Keyboard and other uknow audio device..

I have sucess with 2 HID mouses, with ek-lm3s9d92 usb_dev_chidcdc example, changing the CDC interface to a HID mouse.

Any1 can help me with this?

Thanx

Forum Post: RE: Header file (rom.h) not working appropriately

Jo Chen — Thu, 23 May 2013 20:23:00 GMT

Hello Udit Daga,

Open Rom.h file and you will find that the functions you called are not defined because the conditions of #if statements are not satisfied.

Right click your project, and choose property (alt + Enter), switch to predefined symbol tag:

You can check your predefined symbol on that and whether it matches Rom.h or not.

Regards,

Forum Post: Hardware Interrupt: Interrupt number

Suril Shah — Thu, 23 May 2013 20:16:00 GMT

Hello,

I am trying to make a hardware interrupt in a sys-bios project, however, I don't know what to enter into the Interrupt number field when setting up the hwi via the .cfg file.

After playing around with it, I found that the Interrupt number has to be >= 15, but the interrupt number according to the datasheet for the peripheral I want to hook up is 4.

Going through the data sheet for my specific micro controller, there are two columns, Interrupt Number and Vector Number, but I am not sure which one to look at.

Forum Post: Question about SYSCTL_SYSDIV_1

Jin Seo — Thu, 23 May 2013 20:10:00 GMT

Hi there,

I recently got involved a project based on LM3S9B96. Now I am looking at the datasheet and examples using a timer.

There is a certain code for setting up a system clock

//Configure the system clock to run directly from the Main Oscillator at 8 MHz

SysCtlClockSet(SYSCTL_SYSDIV_1 | SYSCTL_USE_OSC | SYSCTL_OSC_MAIN | SYSCTL_XTAL_8MHZ);

In sysctl.h, SYSCTL_SYSDIV_1 is defined like that:

#define SYSCTL_SYSDIV_1 0x07800000 // Processor clock is osc/pll /1

I guess the SYSCTL_SYSDIV_1 is supposed to write in RCC register, and RCC configuration tells SYSDIV bit fields[23:26] are used for system clock divisor.

In order to set /1, SYSDIV fields need to set up 0x0 according table 5-5 in the datasheet, but in sysctl.h the value of SYSCTL_SYSDIV_1 is 0x07800000, which coverts to

0111 1000 0000 0000 0000 0000 0000 as a binary.

SYSDIV bits are set up 1111, which means SYSCTL_SYSDIV_16 in Table 5-5 of the datasheet.

I am a little confuse how the datasheet and actual codes use a different value to set up divisor 1(/1). I might misunderstand but I don't get it between the example and the datasheet.

Forum Post: RE: LM3S Change to NRND

JohnA — Thu, 23 May 2013 19:07:00 GMT

That's what happened here also. Last Jan TI said there would be LM3S parts in May / June. So here we are in May / June and the guy on the phone at TI says it may be Oct 2013" or "Never" for our orders. My other client gave up: we re-designed and ported to Freescale and are in production now. If we had waited for TI we'd still be waiting.

It seems all that's going on here is TI saying "Its OK, We'll build the parts" except all that happens is the production / leadtime dates keeps moving forward. 16 weeks? Baloney. Show me the parts.

Like: "I'll pay for the hamburgers next Tuesday"

Another client is tired of waiting and for them we'll probably re-port to ST32 or Renesas. Both companies have parts in stock, and don't try to do -all- tech support / customer service on an open web forum. At least you get to talk to a real person (sometimes) and when they say they will ship your stuff tomorrow, it actually happens. At least so far. Not once have they told us "Wait 16 weeks". Somehow the other guys have figured out how to make M3's work at the fab.

The TI concept of putting the PHY on the chip is neat - as far as I know Freescale is the only other mfr that does that on their Coldfires - but they don't offer it in BGA.

But even if TI's concept is a good one, its useless as vaporware. The end users couldn't care less if there's an extra PHY chip or not, they just want to plug it in and see the gadget run.

Forum Post: Input Edge Time

david Nyami — Thu, 23 May 2013 16:55:00 GMT

Hello,

I am completely new to the ARM world. I would like to have the input edge capture mode running. so I have a signal generator connected to my microcontroller( LM3S9B92). I would like to capture the positive edge of the signal currently running at 1 khz. but the timer never captures the interrupt, it times out. I cannot figure out why.Here is what my code looks like :

#define

PART_LM3S9B92

#include

"inc/lm3s9b92.h"

#include

"inc/hw_memmap.h"

#include

"inc/hw_types.h"

#include

"inc/hw_timer.h"

#include

"inc/hw_ints.h"

#include

"driverlib/timer.h"

#include

"driverlib/interrupt.h"

#include

"driverlib/sysctl.h"

#include

"driverlib/gpio.h"

#include

"driverlib/pin_map_LM3S9B92.h"

#include

"utils/uartstdio.h"

unsigned

longg_ulCounter = 0;

void

InitConsole

(void)

{

//Enable input for capture

SysCtlPeripheralEnable(SYSCTL_PERIPH_GPIOD);

// Enable GPIO port A which is used for UART0 pins.

// TODO: change this to whichever GPIO port you are using.

SysCtlPeripheralEnable(SYSCTL_PERIPH_GPIOA);

// Configure the pin muxing for UART0 functions on port A0 and A1.

// This step is not necessary if your part does not support pin muxing.

// TODO: change this to select the port/pin you are using.

GPIOPinConfigure(GPIO_PA0_U0RX);

GPIOPinConfigure(GPIO_PA1_U0TX);

// Select the alternate (UART) function for these pins.

// TODO: change this to select the port/pin you are using.

GPIOPinTypeUART(GPIO_PORTA_BASE, GPIO_PIN_0 | GPIO_PIN_1);

// Initialize the UART for console I/O.

UARTStdioInit(0);

}

void

PWMinput(void)

{

static unsigned long curr,prev;

unsigned long

period=0;

TimerIntClear(TIMER0_BASE,TIMER_CAPA_EVENT|TIMER_TIMA_TIMEOUT);

//reset capture A interrupt flag

curr =

TimerValueGet(TIMER0_BASE,TIMER_A); //read the capture value

UARTprintf("Here is the period %x\n", curr);

//TimerIntDisable(TIMER0_BASE, )

}

void

TimerInit(void)

{

//Enable input for capture

SysCtlPeripheralEnable(SYSCTL_PERIPH_GPIOD);

SysCtlPeripheralEnable(SYSCTL_PERIPH_TIMER0);

GPIO_PORTD_DEN_R = 0x00;

GPIOPinConfigure(GPIO_PD4_CCP0);

GPIOPinTypeTimer(GPIO_PORTD_BASE, GPIO_PIN_4); //PortD.4 as CCP0

TimerConfigure(TIMER0_BASE,TIMER_CFG_16_BIT_PAIR | TIMER_CFG_A_PERIODIC |TIMER_CFG_A_CAP_TIME); //16bit capture

TimerControlEvent(TIMER0_BASE,TIMER_A,TIMER_EVENT_POS_EDGE);

TimerLoadSet(TIMER0_BASE, TIMER_A, SysCtlClockGet()/100000);

TimerIntEnable(TIMER0_BASE,TIMER_CAPA_EVENT|TIMER_TIMA_TIMEOUT); //enable capture A interrupts

TimerIntRegister(TIMER0_BASE,TIMER_A,PWMinput);

IntEnable(INT_TIMER0A);

TimerEnable(TIMER0_BASE,TIMER_A);

}

void

myPeriod(void)

{

TimerIntClear(TIMER0_BASE,TIMER_CAPA_EVENT); //reset capture A interrupt flag

g_ulCounter =

TimerValueGet(TIMER0_BASE,TIMER_A); //read the capture value

UARTprintf("MP-Labor-Uart0 ");

}

//*****************************************************************************

// Configure Timer0B as a 16-bit periodic counter with an interrupt

// every 1ms.

//*****************************************************************************

int

main

(void)

{

unsigned long ul_count;

// Set the clocking to run directly from the external crystal/oscillator.

// TODO: The SYSCTL_XTAL_ value must be changed to match the value of the

// crystal on your board.

SysCtlClockSet(SYSCTL_SYSDIV_2_5 | SYSCTL_USE_PLL | SYSCTL_OSC_MAIN |

SYSCTL_XTAL_16MHZ);

// Set up the serial console to use for displaying messages.

InitConsole();

//start the timer

TimerInit();

ul_count =

SysCtlClockGet();

UARTprintf("Here is the system clock %d", ul_count);

while(1)

{

}

Your help will be much appreciated , thank you

Forum Post: RE: Keyboard and Mouse composite USB device

Vinicius Hubner — Thu, 23 May 2013 13:25:00 GMT

Hi Andrea,

i'm trying to do a HID combo with mouse and keyboard, but i can´t do this yet.

Can you help me with something?

Thx

Forum Post: RE: Debugging RDK-BLDC with Wireshark

Alexander Developer — Thu, 23 May 2013 12:28:00 GMT

Thank you. I found this documentation.

Forum Post: Sleep mode and wake up

Burak DERYA — Thu, 23 May 2013 08:20:00 GMT

Hi,

I am using lm3s9b96 board. I and I want to use it in low consumption mode.

I have configured the sleep mode as in errata;

void sleepConfig(void)
{
SysCtlPeripheralSleepEnable(SYSCTL_PERIPH_GPIOJ);
SysCtlPeripheralDeepSleepEnable(SYSCTL_PERIPH_GPIOJ);
SysCtlPeripheralClockGating(true);
SysCtlSleep();
}

i can achive make peripheral slept, but i can not wake it up. Is there any way to wake it up without interreupt. If there is no way without interreupt, for example can i wake it up with uart interrupt ?

Thanks

Forum Post: RE: cc2530

mohamed moyida — Thu, 23 May 2013 06:26:00 GMT

Dear mohammed

i am also new with cc2530 , can you pleas send to me the codes and helping material(example )

thank u

Regards

Forum Post: RE: RDK-IDM mirror image problem

Markel Robregado — Thu, 23 May 2013 02:26:00 GMT

[quote user="cb1_mobile "]As poster appears to have loaded his code w/out incident - hard for me to justify the presence of, "error message."[/quote]

I guessed wrong this time. I misunderstood the first post.

Forum Post: RE: Stellaris LM3S to Tiva C Series Migration_

miguel2667 — Wed, 22 May 2013 17:49:00 GMT

Roberto,

Tiva devices are seeing STRONG investment from TI across the board. Given they just released to production, we will be into the summer months when we get to the targe 8 week lead times that TI strives to achieve and the parts are fully stocked across online distributors.

Paul,

Regarding IEC 60730 we are looking into when we can make a port available, but I do not have a concrete commitment at this time. What kind of application are you developing that would benefit from its port?

Forum Post: RE: LM3s9b96 ethernet

Sue Cozart — Wed, 22 May 2013 14:35:00 GMT

Hi Jay,

Sorry the link broke when we updated it. I fixed the link, but the App Note is System Design Guidelines for Stellaris® Microcontrollers and its literature number is SPMA036B.

Regards,

Sue

Forum Post: RE: Error connecting to target

cb1_mobile — Wed, 22 May 2013 14:06:00 GMT

[quote user="Makarand Deshmukh"]NO writes to PORTC at all, but I am using pins 5 and 6 as inputs for QEI[/quote]

While I don't know if "pins 5/6" are w/in Port_C - and if we can assume those pins do not "default" into QEI mode - then it may well be that you've "written" to Port_C (either directly or indirectly) - to order those pins into QEI service.

And - done incorrectly - this can cause dreaded JTAG lockout... W/out cause/effect analysis/safeguard - issue may reappear... (i.e. "you've stopped the bleeding" - yet sharp knife remains, "at large!"

Forum Post: RE: LM3S9D96 evaluation board - can't use USB port as serial

Dennis Begun — Wed, 22 May 2013 12:54:00 GMT

I found the windows_drivers directory in StellarisWare, and installing the appropriate driver from it solved the issue.