Hello,

i want to use the EFM8BB31F16G-B-QFN24 in my devices.

What

- Baseadapter

- Daughter Card

- Programming Adapter

i need?

 

Software: Simplicity_Studio_Production_Programmer

 

Thank you.

Mani

 

I'm a IDE user, and have designed ten or so products using SiLabs uPs.  I just started a 8051F380 project, and the breakpoint won't work.  Both the "run until" and the "set breakpoint" act the same.  Two seconds after starting the code running, I get an "unable to remove breakpoint....." alert, then a "Target failed to respond..."

Single stepping works fine, as does downloading the code. 

Any help?

Hi everyone, 

I have a custom board with a C8051F350 for measuring weight running in two modes: low power mode at 32.768 kHz with an external crystal and normal mode at 24.5MHz with the internal oscillator at 24.5 MHz. The board is powered from two 1.5V AA cells. The voltage reference for the ADC is taken from the battery with some filtering. The device normally is in low power mode and Idle (PCON IDLE = 1) consuming around 40uA (including a few other IC's on the board). It is woken up once a minute by a timer interrupt and switching SYSCLK from the external to the internal oscillator.

The usual sequence starting from normal mode is as follows: enable ADC, take 16 samples  (2mV/V, 350 Ohm Load Cell, Buffers disabled, OWR=10, Gain=128). Disable ADC, switch MCU back to low power mode and turn off internal oscillator. The problem is, that the first 2 or three samples from the ADC slowly raise from zero their final value as can be seen in the attached chart (blue curve).  This does not happen if the ADC stays idle mode but enabled through low power mode as shown with the green curve. Unfortunately power consumption is way to high so this is not really an option.

What's wrong? I have monitored the ADC input and VREF voltages with an oscilloscope and there seems to be no difference with the ADC disabled or disabled. Also, inserting a few delays here and there doesn't seem to help either. 

Here are the relevant code snippets:

main()
{
    // Initialization routines
    // .......
    // .......
   
    Oscillator_External();                // Low power mode at 32.768 kHz
    while(1)
    {
        Oscillator_Internal();            // Switch to normal mode 24.5 MHz
        ADC_Result = ADC_Sample(1, 16);   // Take 16 samples from channel 1
        Oscillator_External();            // Go back to sleep again
        PCON |= PCON_IDLE__IDLE;
    }
}

// ADC_Sample partially from F35x_ADC0_Bufferered.c
ADC_Sample(unsigned char channel, int n)
{
    switch(channel)
    {
       // Turn on load cell switch and select corresponding ADC channel
	   case 1:
			LC1 = 1;
			ADC0MUX = ADC0MUX_ADPSEL__ADC0P0 |
		             ADC0MUX_ADNSEL__ADC0N1;      // Input pin selection:
			break;
		case 2:
			LC2 = 1;
			ADC0MUX = ADC0MUX_ADPSEL__ADC0P2 |
		             ADC0MUX_ADNSEL__ADC0N3;      // Input pin selection:
			break;
		case 3:
			LC3 = 1;
			ADC0MUX = ADC0MUX_ADPSEL__ADC0P4 |
		             ADC0MUX_ADNSEL__ADC0N5;      // Input pin selection:
			break;
		case 4:
			LC4 = 1;
			ADC0MUX = ADC0MUX_ADPSEL__ADC0P6 |
		             ADC0MUX_ADNSEL__ADC0N7;    // Input pin selection:
			break;
	}
    ADC0STA_ADINT = 0;                          // clear pending sample indication
    ADC0MD = ADC0MD_ADEN__ENABLED |
             ADC0MD_ADSM__CONT_CONVERSION;      // Start continuous conversions
    while(1)
    {
        // capture n samples
        for (i = 0; i < n; i++)
        {
            while(!ADC0STA_ADINT);              // wait till conversion complete
            ADC0STA_ADINT = 0;                  // clear AD0 interrupt flag

            // concatenate ADC0 data bytes to form the 24-bit value
            ADC_OutputVal = (char)ADC0H;
            ADC_OutputVal <<= 16;
            ADC_OutputVal += (long)ADC0L + ((long)ADC0M << 8);
            sample_array[i] = ADC_OutputVal;
        } 
    }
    // Turn off Load cell switches
    LC1 = 0;
	LC2 = 0;
	LC3 = 0;
	LC4 = 0;

    // This works as expected --> Green curve
	ADC0MD &= ~(ADC0MD_ADSM__FMASK);         // clear bits (put ADC0 in IDLE)
    
    // This doesn't work --> Blue curve 
    // ADC0MD &= ~(ADC0MD_ADSM__FMASK | ADC0MD_ADEN__ENABLED);
}

 

 

With the new batch of chips, the AN127SW programming example stopped working.
Function C2_Discover -> C2_GetDevID returns the correct value 0x12
C2_Discover -> C2_WriteAR  ->  OK
C2_Discover -> C2_WriteCommand -> C2_WriteDR in the loop where the byte of the command is transmitted the MCU starts to transmit something to me! It puts C2D in a high state!

We currently have a customer using a C8051F970 Capsense  module. Currently the firmware we supply to the customer allows them to control sensitivity, and thresholds to match their device and sensors.

In some cases we find we need to alter the Hardcoded Capsense gain.

#define GAIN_VALUE_ARRAY 5, 5, 5, 5, 5, 5, 5

SI_SEGMENT_VARIABLE( CSLIB_maxValues[DEF_NUMN_SENSORS], uint8_t, SI_SEG_CODE ) = { GAIN_VALUE_ARRAY }; 

At the moment this is set at start up.

Is it possible to change this at run time. If I try to change the variables in a function, the compiler returns an error indicating it is read only.

Or more precisely modifiable lvalue.

 

We are using C8051F970 Capsense  modules. We received a new sensor sheet from a customer. The sheet is to be used for measuring water level. It is highly sensitive and causes the sensor input to overflow/saturate. We reduced the sensor gain to 1 which reduced its sensitivity. However it still saturates far earlier than required (After a water height of 70cm).

The obvious approach is to modify the sensor sheet to lower the sensitivity.

However if this is not possible is there anyway to reduce the sensor sensitivity further via firmware/software.

The current sensor output goes from 20000->65520.

One thought process, is this 20,000 could be considered 0.  The idea is it would start measuring from 20,000 onwards. Rather than 20,000->65520, it could do 0->65,520 (+20,000). I guess this is all dependent on hardware, register design.

I think the answer to this is already no, but I have been asked to confirm. Is it possible to change sensitivity at run time.

 

 

 

I'm fairly new to the profession and was wondering what people's general perception of the major MCU players is in terms of things like: price, performance, ease-of-use, manufacturer support etc.

I guess examples of the major players would include companies like ST, NXP, TI, Cypress, Renesas etc.

Or you could just talk about what you like about a certain family of MCU like STM32 or MSP430 or whatever. I'm just looking for general thoughts.

I like AVR for 8-bits mostly because its what I've used in that field. Getting up and running is pretty easy and theres so many different families and price-breaks. I would like to try out PICs although.

I'm a fan of STM32s in the ARM / 32bit field. Their low power line of ARM processors are amazing. The HAL and LL libraries are pretty well documented and have improved significantly compared to the past. I would like to try out silab's chips although since I've heard good things from them.

Hi Tech support team,

We are using C8051F580-IQ, MCU in our design for burner controller.

Few I/O are used for processing of 50/60HZ  , (DUTY CYCLE ~50%) TTL signals. These signals are the O/P of Optocouplers. These pins are configured as Input, weak pull up are enabled.

As per data sheet this part  should support up to -55 Degree centigrade.

But  when we are soaking the product at -40 degree centigrade  for around two hrs.  and restarting then sometime we observed that MCU sees static 5V signals instead of 50hz dynamic signals. To find out the root cause of the issue we  checked optocouplers Behaviors independently  up to -50 Degree centigrade, and observed that every time Optocopulers are working as expected. This is how we are suspecting that for some reason MCU is pulling these dynamic signals to High level .

This issue causing showstopper for starting the formal product verification 

Can you pls. help us on this on high Priority?

 

Thanks and Regards,

Aditya

+91-8374006891

 

Hi Tech support team,

We are using C8051F580-IQ, MCU in our design for burner controller.

Few I/O are used for processing of 50/60HZ  , (DUTY CYCLE ~50%) TTL signals. These signals are the O/P of Optocouplers. These pins are configured as Input, weak pull up are enabled.

As per data sheet this part  should support up to -55 Degree centigrade.

But  when we are soaking the product at -40 degree centigrade  for around two hrs.  and restarting then sometime we observed that MCU sees static 5V signals instead of 50hz dynamic signals. To find out the root cause of the issue we  checked optocouplers Behaviors independently  up to -50 Degree centigrade, and observed that every time Optocopulers are working as expected. This is how we are suspecting that for some reason MCU is pulling these dynamic signals to High level .

This issue causing showstopper for starting the formal product verification 

Can you pls. help us on this on high Priority?

 

Thanks and Regards,

Aditya

+91-8374006891

 

The bootloader configures the clk source. The bootloader then transfers control to the firmware. 
If the firmware uses a quartz resonator rather than an internal oscillator, the firmware does not work.
How to change the clock source correctly?

I'm working on a project that uses the EFM8BB10F2I - QFN20. However when I was programming it using the MCU production programmer, I noticed that it is not in the list of supported devices. How should I proceed? Thank you!

I couldn't find another SI forum to post this on so I'm putting it here since this is in regards to SI 8051.

We are currently attempting to move an old C8051F320 project from the SI IDE to the supported Keil uVision5 tools as the older SI IDE seems to no longer be supported fully.  However I have not found any documentation on if the SI Debug Adapters can work with the uVision 5 SI tools.  Is this supported or must we move to the Keil ULINK adapters?  Or are there other options to support the C8051F320 that are recommended options?  Or should we stick to programming with the last version of the SI IDE with the new tool chain deployed with the SI Keil uVision 5 deployment?

 

 

One week ago on some point I faced  the problem that the connection between j-link and the uC could't be established any more.

That happened when I've tried to start a debugging session after the SYSCLK configuration but here I wasn't 100% sure about but commented this part in my firmware.

I proceeded the work with a new ordered evaluation board until I tried to config the clock again.

void SiLabs_Startup (void) {

	//#########################################################################
	// Disable the watchdog
	//#########################################################################
	SFRPAGE = 0x00;
	//Watchdog Disable  sequence (always disable, enabled by default)
	WDTCN = 0xDE; //First key
	WDTCN = 0xAD; //Second key
	//#########################################################################
	// System Clock Config
	//#########################################################################
	//- Clock derived from the Internal High-Frequency Oscillator
	// SYSCLK is equal to selected clock source divided by 1
	SFRPAGE = 0x00;
	//Use internal 24.5MHz RC Oscillator, and no clock division
	HFOCN |=  HFOCN_HFO0EN__ENABLED;
	CLKSEL = CLKSEL_CLKSL__HFOSC0 | CLKSEL_CLKDIV__SYSCLK_DIV_1;
	CLKSEL = CLKSEL_CLKSL__HFOSC0 | CLKSEL_CLKDIV__SYSCLK_DIV_1;
	while (CLKSEL & CLKSEL_DIVRDY__BMASK == CLKSEL_DIVRDY__NOT_READY);

}

Now I can't communicate with the second evaluation board too....

 Simplicity studio is showing following message (seems to be ok) when i connect the board: 

deviceCount=1

device(440143631) {
  serialNumber=440143631
Connecting to device with serial number 440143631 .. ok.
FirmwareString is Silicon Labs J-Link OB compiled Jun 28 2017 16:11:16
EMDLL version: 0.17.7b519
J-Link DLL Version: 6.44h
  AdapterType=JLink
  adapterLabel=J-Link Silicon Labs
  adapterId=5300A
  adapterName=BRD5300A Rev. A04
  adapterDescription=EFM8LB1 Laser Bee Starter Kit Board
  adapterRevision=A04
  adapterSerialNumber=184600134
  adapterDate=0/0/0
  adapterFirmware=0v15p4b785
  adapterFpga=15v15p255b65535
  adapterInBootloader=no
  supportedDebugModes=OFF,MCU,IN,OUT
  debugMode=MCU
  inferPart=yes
  supportsAEM=yes
  supportsVCom=yes
  supportsEmucom=yes
  supportsIP=no
  boardCount=1
  boardId[0]=5300A
  boardName[0]=BRD5300A Rev. A04
  boardDescription[0]=EFM8LB1 Laser Bee Starter Kit Board
  boardRevision[0]=A04
  boardSerial[0]=184600134
  boardDate[0]=2018/12/3
  inferPart[0]=yes
}

Trying to debug or program the firmware results in a timeout (see attached picture). And j-link disappears from Debug-Adapters list.

Any suggestions to bring the evaluation board back to the life???

 

Best Regards

 

 

I am trying to set up aC8051F970 for I2C fast mode (400kbps).  The datasheet tells that it is possible on pg. 28t:

 

However, I can find no example software, application notes, or information in the datasheet on how to do this.  All example software supports only 10k-100k.  If someone could tell me how to set up the Fast Mode, it would be greatly appreciated.

 

Thank you.