Solved

Hi, Unable to import projects from Example selector, I have installed latest CubeIDE  It shows FW_WBA_1.4.0, But I can see only FW_WBA_1.3.1 in github or st.com   Thanks KarthiG

Hi there,i am currently working on a BLE Customboard using a STM32WB55.The current setup allows for a 1 second long BLE Advertising, to transmit sensor data.After this second the device is going into STANDBY MODE using the tiny_lpm class: // from app_ble.c -> Callback when the HW_TIM has timedout after 1 second. static void Adv_Mgr(void) { aci_gap_set_non_discoverable(); UTIL_LPM_EnterLowPower(); } In the stm32_lpm_if.c i am doing the following to actually stretch the wakeup time to be several minutes / hours: void PWR_EnterOffMode(void) { /* USER CODE BEGIN PWR_EnterOffMode_1 */ HAL_RTCEx_DeactivateWakeUpTimer(&hrtc); if (HAL_RTCEx_SetWakeUpTimer_IT(&hrtc, RTC_WAKEUP_COUNTER, RTC_WAKEUPCLOCK_CK_SPRE_16BITS) != HAL_OK) { Error_Handler(); } ..... The rest of the function as it is. Overall this seems to work quite nice since i am ending up with a power consumption of 0.6-0.7 uA which is matching with the datasheet.I wanted to clarify if the spikes

I'm new to STM parts and toolchain and I need to send input capture values from TIM1 to the DAC output via DMA. Cube doesn't support peripheral to peripheral DMA, but its been suggested I can accomplish it with a peripheral to memory DMA transfer, triggered by TIM1\CH4 if I supply the address of the DHR12RD register as a destination address. The user guide says this register is not memory mapped, but I can see what address is from the generated code. What happens if I attempt this? Is there another way to do it?Also, the DAC is 12 bit and TIM1 is 16 bit, so I'm hoping the top four bits are simply ignored. Since the above method is non-standard, I'm not sure if this could make things not work. Under normal functioning, the capture registers will not record values above 4096, so overflows are not anticipated.

I need to use DMA to forward input capture values from the timers to the DAC output register. The input capture values represent the period of the input signal, which makes it work as an FM demodulator. If I can route the captured values to DACs and other peripherals, I can process more channels and use higher carrier frequencies. I’ve tried to do this before, but couldn’t make it work.  This process may end up gating the application so I'm trying again.  I've received bits of advice, but nothing worked.  I still don't know if is even possible. I'd basically need working code to confirm since I can't figure it out. Also, I attempted this process on an STM32L4...and I'm not sure if behavior might differ from unit to unit.  I'm considering switching to a unit with more timers, so I'm not sure if DMA behavior will be substantively different.  

I'm using CubeMX. With the new unit, I had to set ENVR and clear HIZ in VREFBUF_CSR in user code in order to use the internal reference.  In this config, I can see the output value change DC level, but with a 700khz noise tone on it that amplifies along with the DC level. It reaches 1v peak to peak when the DAC is outputting the max value.Are there settings that could end up causing this behavior?

I’m using an STM32G473 and I have TIM20 CHANNEL1 configured in output compare mode. When the compare value is reached, I want to move CNT values from TIM1, TIM5, TIM6, TIM7 and TIM15 to 5 static locations in memory.I’m not sure if I can use DMA to streamline this. I’m using CubeMX and HAL and I’m looking for examples of something like this, but so far I only see DMA examples for transferring to and from the triggering peripheral. Can DMA do this?

I need to load the TIM1 CCR registers into the DAC1 outputs on an STML432KC. I have this working with the line DAC1->DHR12R1 = TIM1->CCRx executing in a loop. I'd like to substitute with DMA. Others have suggested that I can do this with the available DMA features and supplying the the necessary register addresses. I gave it a shot and couldn't make it work. However, this is my first STM32 project, and I'm new to the toolchain. Is this approach really viable?

I began using an STM32G473 and the DAC outputs are acting funny. I have VREFBUF configured as an internal reference. Since I'm not using VREF+ as an output, I left the pin unconnected. I posted a thread about it and it was recommended that I connect VREF+ through a cap to ground.  I'm working with a custom board where this isn't possible so I can't test.I haven't found anything in the documentation that would suggest leaving the VREF+ pin unconnected would have ill effects on DAC outputs. Is this documented anywhere? Original thread:https://community.st.com/t5/stm32-mcus-products/triangle-wave-dac-output-on-stm32g473-seems-to-be-stuck-enabled/m-p/659118#M240313

I’m using CubeIDE, which lets me configure a timer to trigger a DMA request triggered by an output compare that transfers that timer's value to memory. I want to modify the generated code such that a separate timer’s value is sent to memory instead.Within the HAL_TIM_OC_Start_DMA function, HAL_DMA_Start_IT is called, (??) which seems to be where the to\from addresses are established. I’m hoping I can modify this call to reference the memory location of another timer’s CNT value…but I’m not certain if its as simple as modifying a single call. I’m up against other anomalous behavior, so I’m not sure if this approach works. Can it?

I am using CubeIDE, and I want an output compare event on TIM8 CH1 to trigger a DMA transfer to a single memory location. I’ve selected ‘Peripheral to Memory’ as the direction, but I’m observing the contents of the memory location being copied to the CCR1 register instead of the other way around. This also happens in ‘Memory to Peripheral’ mode, which would be expected to have opposite behavior.Using ST-link, I can seed the DMA1 registers configured in peripheral to memory mode:DIR = 0 (Per documentation this is correct)CMAR register maps to 0x40013434, which is the TIM8->CCR1 address. (Should be the memory address though)CPAR = 0x20000000 This maps to the base address of the SRAM section. This should be the peripheral address.If I switch to memory to peripheral mode, see the contents of the CMAR and CPAR registers switch, but I also see DIR switch so the two changes together seem like they may be cancelling to enable only memory to peripheral behavior.If this is indeed a bug, what

I seem to remember a way to make a reset event set a timer’s value to a non-zero value. However, I didn’t see anything in the docs for the part I’m using so I’m not sure if this is a feature in STM32 parts. Is it?An alternative is a DMA trigger on the reset that loads the CNT value. Is there a better way?I have an in input capture value that represents a signed value. If a 16 bit timer resets to a value above 65535/2, input capture values within a known range can be sent away via DMA to the FMAC without processing steps to sort out the sign.  

I have a variety behavior that didn't occur on the last version of my board. I'm trying to figure out if the board has an intrinsic electrical fault, or if incompatible programming is making the pins fight with themselves to generate what I'm seeing.  The board gets up and goes and I can read registers, so I'm hoping there are diagnostic steps I can take to figure out how to correct what I'm seeing below 1. When I program the board with ST-Link, VREFBUF HIZ=1, which is incorrect and makes VREF+ pin not work.  I’ve been pushing ahead by setting HIZ=0 via ST-Link after programming to get the right behavior.  I didn't have to do this on previous boards...so I'm wondering if I introduced an electrical failure that makes the pin go Hi-Z in a fault mode or some such...? 2. After programming, I can pull power to the board (ST_Link still engaged) and wait 5 secs for the caps to power out, and reconnect, VREFBUF shows HIZ=0 and VREF+ pin works.  Why the change in b

I’m using CubeIDE and I need to load a single memory location of the FMAC input buffer via DMA with readings from a single ADC channel that is triggered by a timer output compare.  DR register on the ADC shows a stuck value indicating conversions aren’t taking place. I’m using the same trigger for other things successfully, but I don’t see any conversions happening. I'm using HAL_ADC_START_DMA + Cube...are there more config steps for trigger mode? As a possible substitute, I disabled the trigger option and kept continuous mode enabled. I can see a single conversion and DMA transfer take place at startup, but not continuous.  I read this can be due to overloading DMA with constant requests, so I set the sample time to 640 clock cycles, but it didn’t change anything. In this state, I can see ADSTART is 0. If I set it to 1, I get another single conversion and a DMA transfer and ADSTART goes back to zero. It looks like this is nearly working...but I'm likely just missin

Hello,I'm porting code that worked on STM32U585CI using SPI1 as full duplex master with GPDMA1 CH4(Rx)/CH5(Tx) to an STM32U585QII using SPI3, again as master with GPDMA1 CH4(Rx)/CH5(Tx). Only 1 SPI is used. I'm using CubeIDE 1.13.2 to autogenerate HAL initialization code, which I confirmed calls MX_GPDMA1_Init() before MX_SPI3_Init(). I compared the autogenerated code between the two chips to see that all IOC-controlled settings came out the same. I've looked some through the reference manual to identify differences between SPI3 v SPI1/2, and also saw this ticket, warning that there is a SPI3 size limitation of 1023 bytes.While I'm also in the process of determining whether there might be a board issue, I'm wondering if there are any subtle gotchas with what I'm trying to do, using SPI3 versus one of the other SPI peripherals. There is this blocking function (partially shown below) used during application initialization that gets called before DMA is even engaged. The U585QII gets stuc

I have STM32CubeIDE Version 1.15.1 installed and was unable to open an example for the STM32U575 because the installer said:Other downloads worked fine and I checked the connection in the Firmware Updater:I have other previous versions installed but the drop down box is disabled so I couldn't select any of them.  It appears that v1.6.0 is not yet available: This seems like a bug in the IDE.  Is that the case or am I missing something? FYI:I was able to manually find the example code I needed from:%USERPROFILE%\STM32Cube\Repository\STM32Cube_FW_U5_V1.5.0\Projects\NUCLEO-U575ZI-Q\Examples  

my requirement is i have to initialise a variable in M7 core and access that variable in M4 core, please let me know the process t achieve this one. board - NUCLEO-H745ZI-Q.ide - stm32cube(1.13.1 version).

I don't know if this is the previously mentioned login issues or not, but I will explain.Last week I received my new Macbook with an M2 MAX cpu. I installed the new macOS 15.0 Sequoia Beta on it, as I'm primarily an iOS developer. Then I installed STM32CubeIDE 1.15.1, because I also do firmware development sometimes, and I want to have the IDE working when I need it.But when selecting myST ->Login from the menu, all I get is a small window with a spinner and the message 'Loading.... Please wait.". It never goes away.Sometimes selecting myST->Login does nothing. But generating a project always fails with "Code generation could not be done most probably because the necessary firmware package is missing. etc. etc.".I have read most or all the posts about login issues. But nothing seems to be helpful.I have read this: https://community.st.com/t5/stm32-mcus/how-to-set-the-network-connection-parameters-for-stm32cubeide/ta-p/585694, but it does not make any sense to me. I am not ru

I just find my HAL_UART_Transmit() function cost 15ms. why does it cost so much time?The UART speed is 19200, and I use GPIO set and reset function to check the time cost.Total data length of send buffer is 29 bytes. and I use a 15ms polling loop to call this function, so it cost me over 30ms for every msg sending task, but for 19200 speed, it should only need 12.1ms for each 29 bytes.I assume this reason may be caused by the HAL_UART_Transmit() function is a sync function, it needs to wait all the data tranferring finished. is it right? If yes, is it means I can reduce my robin task period? Kind RegardsRichKnight

HelloI have CubeMx v6.11.1.Am trying to start a BLE_HeartRate project targeting P-Nucleo-WB55-Nucleo, and CubeMx is needing STM32Cube_FW_WB_V1.20.0This is not available at the ST website. Can you send me this or tell me how to create this project using V1.19?Thanks.Giri

Will you also review the newly designed stm32mp157 artwork file?

board: stm32U585 + stsafe-a110 before the stm32u585 enter the low power mode ,I want to set stsafe-a110 enter the low power consumpion mode. I can't find the methods to enter Hibernate      

I was trying to test the capability of the STSAFE-A110 chip to store and verify a password, but it didn't work.StSafeA_PutAttribute() returned STSAFEA_UNSATISFIED_ACCESS_CONDITION.StSafeA_VerifyPassword() returned STSAFEA_ENTRY_NOT_FOUND.I'm not sure what any of this means. When I call other functions, like StSafeA_GenerateSignature() or StSafeA_VerifyMessageSignature(), they work just fine. Why would StSafeA_PutAttribute() and StSafeA_VerifyPassword() not work? Here's my code: char* passwordStr = "banana101"; // Same as in the OpenSSL example int32_t statusCode; StSafeA_VerifyPasswordBuffer_t commandResult; // This call returns STSAFEA_UNSATISFIED_ACCESS_CONDITION instead of STSAFEA_OK statusCode = StSafeA_PutAttribute(gStsafeHandle, STSAFEA_TAG_PASSWORD_SLOT, (uint8_t*)passwordStr, strlen(passwordStr), STSAFEA_MAC_NONE); // This call returns STSAFEA_ENTRY_NOT_FOUND instead of STSAFEA_OK statusCode = StSafeA_VerifyPassword(gStsafeHandle, (uint

Problem Statement: Amplifying Low-Amplitude Audio Signal on STM32 Nucleo64 BoardHello STM community! I'm new here and seeking some guidance. I'm working on a project using the STM32 Nucleo64 board to create an announcement maker. My setup involves using an interrupt to handle an 8kHz frequency sampled audio, which is converted into an array. This array is then outputted from a PWM pin.However, I'm facing an issue with the audio amplitude. The output signal amplitude is very low, less than 3.3V, which is insufficient to trigger the amplifier circuit and effectively broadcast the audio.Is there a way to internally amplify the signal output coming from the STM32 board? Any suggestions or guidance would be greatly appreciated!

Dear Gfx TeamCould you please preserve/follow the cap[pitalisation in file names, folders etc in auto-generated files and folders ?E.g. the file's name is "ScrTestTextArea" but all folders, filenames etc are ignoring upper case letter. Preserving the case improves readabbility.uVision project:Include paths:Folder names:MSVS projectThanks 

Dear Gfx Team,Please be so kind and fix the code generating Visual Studio project so the capitalization of the screen names is the same as seen in GfxDesigner. It is very difficult to read in all lower case. Why all lower case ? Does not help with readibility. Thank youFerro