Refer to application note #32136, good luck finding the code that it references in chapter 7.
↧
Tricore TC1782 GPT Pins
↧
Performance Comparison between single-float and double
I tested the computation performances for float and double on TriCore297, I obtained a strange result, they are so similar.
I let TC297 run dotproduct for float and double, and the length of vector are both 128, executed for 1000 times.
Then, I caculate cycle-count through IfxCpu_resetAndStartCounters and IfxCpu_stopCounters.
At last, I found their performance(cycle count) are very similar!
TC297 just have single-float co-operator, so, the single-float's compuation should be faster than double's computation.
It is so strange, Why?
Is there any one would tell me the reason?
I let TC297 run dotproduct for float and double, and the length of vector are both 128, executed for 1000 times.
Then, I caculate cycle-count through IfxCpu_resetAndStartCounters and IfxCpu_stopCounters.
At last, I found their performance(cycle count) are very similar!
TC297 just have single-float co-operator, so, the single-float's compuation should be faster than double's computation.
It is so strange, Why?
Is there any one would tell me the reason?
↧
↧
Addressing PDO with different size. APPL_InputMapping & APPL_OutputMapping
Yes, padding doesn't help. The problem is 100% the pData in void APPL_InputMapping(UINT16* pData), it's shifted. Now i'm having addressing problem even without the 8entries object. I have 16entries objects and one input object with 4 uint16 entries.
I've set all the entries to 0, but the APPL_InputMapping(UINT16* pData) set the first entries of the 4 uint16 entries object to 1. I've also initialized the aPdInputData[(MAX_PD_INPUT_SIZE>>1)] to 0 in the ecatappl.c file, but nothig has changed.
I don't know how to fix the shifted pointer. I've also tried to use a 8bit pointer, but the problem remains.
I've set all the entries to 0, but the APPL_InputMapping(UINT16* pData) set the first entries of the 4 uint16 entries object to 1. I've also initialized the aPdInputData[(MAX_PD_INPUT_SIZE>>1)] to 0 in the ecatappl.c file, but nothig has changed.
I don't know how to fix the shifted pointer. I've also tried to use a 8bit pointer, but the problem remains.
↧
dependency of short circuit capability with some factors
Hi AnniLe,
some details on SC (e.g. example waveform) you can find in the following Application note:
https://www.infineon.com/dgdl/Infine...61c27fbcda0aae
SC withstand time will be significantly reduced for gate voltages above 15 V. Therefore for applications needed SC capability we are recommending to use 15 V gate voltage, whereas applications not needing SC capability
Lower DC-link voltage will increase SC withstandtime approximately inverse proportional
Lower junction temperature before SC will give a slight increase in SC withstand time.
Best regards,
electricuwe
some details on SC (e.g. example waveform) you can find in the following Application note:
https://www.infineon.com/dgdl/Infine...61c27fbcda0aae
SC withstand time will be significantly reduced for gate voltages above 15 V. Therefore for applications needed SC capability we are recommending to use 15 V gate voltage, whereas applications not needing SC capability
Lower DC-link voltage will increase SC withstandtime approximately inverse proportional
Lower junction temperature before SC will give a slight increase in SC withstand time.
Best regards,
electricuwe
↧
XMC4000 Background Request Source Behavior
The description in the xmc_vadc.h says
So, once you've triggered the sampling on a channel, it goes on
Code:
/**
* @param global_ptr Pointer to the VADC module
* @return
* None
*
* \par<b>Description:</b><br>
* Enables continuous conversion mode.<BR>\n
* Typically for a Background request source to generate conversion request, either a hardware trigger or a software
* request is needed. Using autoscan (continuous conversion)feature it is possible to start the conversion
* once and allow the sequence to repeat without any further triggers. Once all channels belonging to a Background
* request source have been converted, a request source completion event is generated. Generation of this event
* can restart the Background configure sequence. Every request source event will cause a load event to occur.
* A call to this API would access the register bit field BRSMR.SCAN.
*
* \par<b>Related APIs:</b><br>
* XMC_VADC_GROUP_BackgroundDisableContinuousMode()<BR>
*/
__STATIC_INLINE void XMC_VADC_GLOBAL_BackgroundEnableContinuousMode(XMC_VADC_GLOBAL_t *const global_ptr)
{
XMC_ASSERT("XMC_VADC_GLOBAL_BackgroundEnableContinuousMode:Wrong Module Pointer", (global_ptr == VADC))
global_ptr->BRSMR |= (uint32_t)VADC_BRSMR_SCAN_Msk;
}↧
↧
triboard wth tc1762
Looking for a TriBoard with a TC1762 for purchase
↧
Question about the voltage-class
Also for the Easy modules 0 V can be used in case simplicity of the design has priority vs. switching speed. With a higher turn-on gate resistor the dV/dt can be reduced to a level that also modules - inspite of their longer internal and external gate connections - can be operated with 0 V. In some applications - like drives - dV/dt needs to be limited anyhow to avoid adverse effects on the machine windings and bearings.
↧
Any plans for Cortex M7 ARM chips from Infineon?
I've been asking this same question of our Infineon rep for years now and have gotten nothing but dead silence in response. :(
↧
Tle5012b ssc via tle987x evalboard,
Hello Fiz,
I have not tried 3 wire spi but i will. I have tried with arduino thanks to github arduino library of tle5012b. But it did not work. I do not know what is wrong :)
Kind regards
Murat.
I have not tried 3 wire spi but i will. I have tried with arduino thanks to github arduino library of tle5012b. But it did not work. I do not know what is wrong :)
Kind regards
Murat.
↧
↧
Performance Comparison Between int and float for TC297
Hi Shaquille,
Can you list the sample code for the integer and float sample case? Just so that we get a better understanding of what you're doing.
Best regards,
Henk-Piet Glas
Principal Technical Specialist
Embedded Software
↧
tle987x debugging problem with segger j-link
Hello everyone,
I have a pcb board with tle987x mcu. I have been trying to laod the program with j link segger base. you can seen our errors on the pictures uploaded.We are trying to load our program thanks to SWD with j-link segger base. As you can see on 3th picture uploaded, VDDP connected to 40th pin,
SWDIO connected to 20th pin(TMS), SWDCLK connected to 21th pin,RESET connected to 22th pin. What can our problem be? Any suggestion will be appreciated about loading program to tle987x.
Kind Regards,
Murat.
Attachment 4126Attachment 4127Attachment 4128
I have a pcb board with tle987x mcu. I have been trying to laod the program with j link segger base. you can seen our errors on the pictures uploaded.We are trying to load our program thanks to SWD with j-link segger base. As you can see on 3th picture uploaded, VDDP connected to 40th pin,
SWDIO connected to 20th pin(TMS), SWDCLK connected to 21th pin,RESET connected to 22th pin. What can our problem be? Any suggestion will be appreciated about loading program to tle987x.
Kind Regards,
Murat.
Attachment 4126Attachment 4127Attachment 4128
↧
How can I get
In <TC2xx Core Architecture User Manual (Volume 1)EN>, there are some description as following:
The following additional documents are also available for download from the TriCore Architecture and Core section:
TriCore® DSP Optimization Guide.
TriCore® EABI (Embedded ABI) Users Manual
TriCore® Compiler Writers Guide
But, I cannot find "TriCore® DSP Optimization Guide" and "TriCore® Compiler Writers Guide" in Infineion's website.
where and how to find them?
The following additional documents are also available for download from the TriCore Architecture and Core section:
TriCore® DSP Optimization Guide.
TriCore® EABI (Embedded ABI) Users Manual
TriCore® Compiler Writers Guide
But, I cannot find "TriCore® DSP Optimization Guide" and "TriCore® Compiler Writers Guide" in Infineion's website.
where and how to find them?
↧
Example for temperature sensor
I know this question is old, and I am probably wasting my time posting a response, but maybe I can save somebody a little trouble in the future.
The short answer is, that the XMC_CalcTemperature, or any of its derivatives and variants, does not work correctly because it was never fully implemented by Infineon, which is why it is not included in the #defines in XMC1000_RomFunctionTable.h
The slightly longer answer is that the code as it exists in the XMC1400 boot kit hardware requires four parameters that are supposed to be available in the flash starting at address 0x10000f20, but they are missing (all 0xffffffff). As a result, the calculation always returns 393 since it is the maximum value the sensor was supposed to be good for.
The calculation routine starts at address0x539, so you can set a breakpoint to it in the debugger and overwrite the address that it uses for these values to some place in RAM, and experiment with different values (min, max, gain, offset), but these numbers will be different for each device, so there isn't much point.
The short answer is, that the XMC_CalcTemperature, or any of its derivatives and variants, does not work correctly because it was never fully implemented by Infineon, which is why it is not included in the #defines in XMC1000_RomFunctionTable.h
The slightly longer answer is that the code as it exists in the XMC1400 boot kit hardware requires four parameters that are supposed to be available in the flash starting at address 0x10000f20, but they are missing (all 0xffffffff). As a result, the calculation always returns 393 since it is the maximum value the sensor was supposed to be good for.
The calculation routine starts at address0x539, so you can set a breakpoint to it in the debugger and overwrite the address that it uses for these values to some place in RAM, and experiment with different values (min, max, gain, offset), but these numbers will be different for each device, so there isn't much point.
↧
↧
FAQ for XMC4800 EtherCAT
Hi,all:
Is there any example on how ethercat nodes can be linked like daisy chain and pass message in relay to the master? And how can I monitor the chain is working fine with TwinCat?
Regards & thanks .
Is there any example on how ethercat nodes can be linked like daisy chain and pass message in relay to the master? And how can I monitor the chain is working fine with TwinCat?
Regards & thanks .
↧
Replacement guide for NovalithIC products to be phased out
Replacement guide for NovalithIC and Trilith IC products to be phased out over the next years
Attachment 4134
Attachment 4134
↧
Replacement guide for TrilithIC products to be phased out
↧
Aurix MotorControl Application Kit - Salesname Kit_Aurix_TC234_MotorCtr"
I downloaded the available files from MyInfineon Aurix Customer Documentation including the EvalBoard software example "Ap32298_Application_Example_11".
However, I can not find the file Cfg_PosIf_RdcDs.c which is required to run the application board with a resolver position sensor.
You can find the reference to this file in the PDF eMotor application example (PMSM with FOC) page 13.
Thanks for your help.
However, I can not find the file Cfg_PosIf_RdcDs.c which is required to run the application board with a resolver position sensor.
You can find the reference to this file in the PDF eMotor application example (PMSM with FOC) page 13.
Thanks for your help.
↧
↧
HSM lock
Hi everyone,
I have problem after flashing HSM from examples on TC297 devkit board:
1. flashed BootBasic - (i used percompiled file from example)
2. flashed HsmDemo (i compiled it with HIGHTEC ARM toolchain)
3. flashed ProconHsmCotp for configuration UCB
I checked code in memory on every step, everything flashed. I flashed it with memtool, convert ELF file to HEX with objcopy -O ihex file.elf file.hex
After restart i cann't connect to HSM. Also i have red led ESR0 (RESET state) What does it mean?
I try check board with DAS device scanner, it sees board, but i have strange status "DAS Error: DEVICE_LOCKED". But i don't set any protection, just flashed examples HSM. Why have i this error?
Attachment 4137
Give me any advice.
I have problem after flashing HSM from examples on TC297 devkit board:
1. flashed BootBasic - (i used percompiled file from example)
2. flashed HsmDemo (i compiled it with HIGHTEC ARM toolchain)
3. flashed ProconHsmCotp for configuration UCB
I checked code in memory on every step, everything flashed. I flashed it with memtool, convert ELF file to HEX with objcopy -O ihex file.elf file.hex
After restart i cann't connect to HSM. Also i have red led ESR0 (RESET state) What does it mean?
I try check board with DAS device scanner, it sees board, but i have strange status "DAS Error: DEVICE_LOCKED". But i don't set any protection, just flashed examples HSM. Why have i this error?
Attachment 4137
Give me any advice.
↧
Addressing PDO with different size. APPL_InputMapping & APPL_OutputMapping
Finally i've solved the problem! pData pointer wasn't shifted, i just had to read/write the 8 MSB bits or the 8 LSB bits according to the mapped object
↧
SDMMC004 without card detect signal
I'm using FATFS App which uses SDMMC_BLOCK App and this behavior cost me a few hours
@Infineon: can you please fix that, so that when no card detection is used, the bits are set by the app.
I'm not sure if this is the correct/best place, and it now seems OK.
I put
@Infineon: can you please fix that, so that when no card detection is used, the bits are set by the app.
Quote:
I'm not sure if this is the correct/best place, and it now seems OK.
SET_BIT(SDMMC->HOST_CTRL, SDMMC_HOST_CTRL_CARD_DET_SIGNAL_DETECT_Pos);
SET_BIT(SDMMC->HOST_CTRL, SDMMC_HOST_CTRL_CARD_DETECT_TEST_LEVEL_Pos);
after DAVE_Init(), so that a code generation doesn't remove itSET_BIT(SDMMC->HOST_CTRL, SDMMC_HOST_CTRL_CARD_DETECT_TEST_LEVEL_Pos);
↧