22. Digital-to-analog converter (DAC)

22.1 Introduction

The DAC module is a 12-bit, voltage output digital-to-analog converter. The DAC can be configured in 8- or 12-bit mode and may be used in conjunction with the DMA controller. In 12-bit mode, the data could be left- or right-aligned. The DAC features two output channels, each with its own converter. In dual DAC channel mode, conversions could be done independently or simultaneously when both channels are grouped together for synchronous update operations. An input reference pin, $V_{REF+}$ (shared with others analog peripherals) is available for better resolution. An internal reference can also be set on the same input. Refer to voltage reference buffer (VREFBUF) section.

The DACx_OUTy pin can be used as general purpose input/output (GPIO) when the DAC output is disconnected from output pad and connected to on chip peripheral. The DAC output buffer can be optionally enabled to allow a high drive output current. An individual calibration can be applied on each DAC output channel. The DAC output channels support a low power mode, the Sample and hold mode.

22.2 DAC main features

The DAC main features are the following (see Figure 157: Dual-channel DAC block diagram )

• One DAC interface, maximum two output channels
• Left or right data alignment in 12-bit mode
• Synchronized update capability
• Noise-wave and Triangular-wave generation
• Dual DAC channel for independent or simultaneous conversions
• DMA capability for each channel including DMA underrun error detection
• External triggers for conversion
• DAC output channel buffered/unbuffered modes
• Buffer offset calibration
• Each DAC output can be disconnected from the DACx_OUTy output pin
• DAC output connection to on chip peripherals
• Sample and hold mode for low power operation in Stop mode
• Input voltage reference, $V_{REF+}$

Figure 157 shows the block diagram of a DAC channel and Table 146 gives the pin description.

22.3 DAC implementation

Table 145. DAC features

DAC features	DAC1
Dual channel	X
Output buffer	X
I/O connection	DAC1_OUT1 on PA4, DAC1_OUT2 on PA5
Maximum sampling time	1MSPS
Autonomous mode	-

22.4 DAC functional description

22.4.1 DAC block diagram

Figure 157. Dual-channel DAC block diagram

Dual-channel DAC block diagram showing two channels (Channel 1 and Channel 2) with control registers, sample and hold registers, DAC converters, buffers, and output pins (DACx_OUT1 and DACx_OUT2).

The diagram illustrates the internal architecture of a dual-channel DAC. It is divided into two main sections: DAC channel 1 and DAC channel 2 .

DAC channel 1: Includes a Control registers & logic Channel1 block with TSEL1[3:0] bits, TRIG , and DMA_Request inputs. It connects to Sample & Hold Registers containing TSAMPLE1 , THOLD1 , and TREFRESH1 . These registers feed into a DAC_DOR1 (12-bit) register, which then connects to the DAC converter 1 . The converter also receives MODE1 bits and Offset calibration OTRIM1[5:0] bits . The output of the converter passes through Buffer 1 to produce dac_out1 at the DACx_OUT1 pin. The LSI clock is also connected to this channel.
DAC channel 2: Follows a similar structure with Control registers & logic Channel2 (with TSEL2[3:0] bits), Sample & Hold Registers ( TSAMPLE2 , THOLD2 , TREFRESH2 ), DAC_DOR2 (12-bit), DAC converter 2 (with MODE2 bits and Offset calibration OTRIM2[5:0] bits ), and Buffer 2 leading to dac_out2 at the DACx_OUT2 pin. This channel also connects to On-chip Peripherals .

Common components include VREF+ , VDDA , and VSSA pins. Both channels are connected to the APB1 Bus .

Dual-channel DAC block diagram showing two channels (Channel 1 and Channel 2) with control registers, sample and hold registers, DAC converters, buffers, and output pins (DACx_OUT1 and DACx_OUT2).

1. MODEx bits in the DAC_MCR control the output mode and allow switching between the Normal mode in buffer/unbuffered configuration and the Sample and hold mode.
2. Refer to Section 22.3: DAC implementation for channel2 availability.

The DAC includes:

• Up to two output channels
• The DACx_OUTy can be disconnected from the output pin and used as an ordinary GPIO
• The DAC_OUTx can use an internal pin connection to on-chip peripherals such as comparator, operational amplifier and ADC (if available).
• DAC output channel buffered or non buffered
• Sample and hold block and registers operational in Stop mode, using the LSI clock source for static conversion.

The DAC includes up to two separate output channels. Each output channel can be connected to on-chip peripherals such as comparator, operational amplifier and ADC (if available). In this case, the DAC output channel can be disconnected from the DACx_OUTy output pin and the corresponding GPIO can be used for another purpose.

The DAC output can be buffered or not. The Sample and hold block and its associated registers can run in Stop mode using the LSI clock source.

Table 146. DAC input/output pins

Pin name	Signal type	Remarks
VREF+	Input, analog reference positive	The higher/positive reference voltage for the DAC, $V_{REF+} \leq V_{DDAmax}$ (refer to datasheet)
VDDA	Input, analog supply	Analog power supply
VSSA	Input, analog supply ground	Ground for analog power supply
DACx_OUTy	Analog output signal	DACx channely analog output

22.4.2 DAC channel enable

Each DAC channel can be powered on by setting its corresponding ENx bit in the DAC_CR register. The DAC channel is then enabled after a $t_{WAKEUP}$ startup time.

Note: The ENx bit enables the analog DAC channelx only. The DAC channelx digital interface is enabled even if the ENx bit is reset.

22.4.3 DAC data format

Depending on the selected configuration mode, the data have to be written into the specified register as described below:

• Single DAC channel
There are three possibilities:
- – 8-bit right alignment: the software has to load data into the DAC_DHR8Rx[7:0] bits (stored into the DHRx[11:4] bits)
- – 12-bit left alignment: the software has to load data into the DAC_DHR12Lx [15:4] bits (stored into the DHRx[11:0] bits)
- – 12-bit right alignment: the software has to load data into the DAC_DHR12Rx [11:0] bits (stored into the DHRx[11:0] bits)

Depending on the loaded DAC_DHRyyxx register, the data written by the user is shifted and stored into the corresponding DHRx (data holding registerx, which are internal non-memory-

mapped registers). The DHRx register is then loaded into the DORx register either automatically, by software trigger or by an external event trigger.

Figure 158. Data registers in single DAC channel mode

Figure 158: Data registers in single DAC channel mode. A diagram showing three rows of a 32-bit register (bits 31 to 0) for single channel mode. The first row, '8-bit right aligned', has data in bits 7:0. The second row, '12-bit left aligned', has data in bits 31:20. The third row, '12-bit right aligned', has data in bits 11:0. The label 'ai14710b' is in the bottom right corner.

• Dual DAC channels (when available)

There are three possibilities:

– 8-bit right alignment: data for DAC channel1 to be loaded into the DAC_DHR8RD [7:0] bits (stored into the DHR1[11:4] bits) and data for DAC channel2 to be loaded into the DAC_DHR8RD [15:8] bits (stored into the DHR2[11:4] bits)
– 12-bit left alignment: data for DAC channel1 to be loaded into the DAC_DHR12LD [15:4] bits (stored into the DHR1[11:0] bits) and data for DAC channel2 to be loaded into the DAC_DHR12LD [31:20] bits (stored into the DHR2[11:0] bits)
– 12-bit right alignment: data for DAC channel1 to be loaded into the DAC_DHR12RD [11:0] bits (stored into the DHR1[11:0] bits) and data for DAC channel2 to be loaded into the DAC_DHR12RD [27:16] bits (stored into the DHR2[11:0] bits)

Depending on the loaded DAC_DHRyyxD register, the data written by the user is shifted and stored into DHR1 and DHR2 (data holding registers, which are internal non-memory-mapped registers). The DHR1 and DHR2 registers are then loaded into the DAC_DOR1 and DOR2 registers, respectively, either automatically, by software trigger or by an external event trigger.

Figure 159. Data registers in dual DAC channel mode

Figure 159: Data registers in dual DAC channel mode. A diagram showing three rows of a 32-bit register (bits 31 to 0) for dual channel mode. The first row, '8-bit right aligned', has data for channel 1 in bits 11:4 and channel 2 in bits 3:0. The second row, '12-bit left aligned', has data for channel 1 in bits 11:0 and channel 2 in bits 31:20. The third row, '12-bit right aligned', has data for channel 1 in bits 11:0 and channel 2 in bits 27:16. The label 'ai14709b' is in the bottom right corner.

22.4.4 DAC conversion

The DAC_DORx cannot be written directly and any data transfer to the DAC channels must be performed by loading the DAC_DHRx register (write operation to DAC_DHR8Rx, DAC_DHR12Lx, DAC_DHR12Rx, DAC_DHR8RD, DAC_DHR12RD or DAC_DHR12LD).

Data stored in the DAC_DHRx register are automatically transferred to the DAC_DORx register after one APB1 clock cycle, if no hardware trigger is selected (TENx bit in DAC_CR register is reset). However, when a hardware trigger is selected (TENx bit in DAC_CR register is set) and a trigger occurs, the transfer is performed three APB1 clock cycles after the trigger signal.

When DAC_DORx is loaded with the DAC_DHRx contents, the analog output voltage becomes available after a time $t_{\text{SETTLING}}$ that depends on the power supply voltage and the analog output load.

HFSEL bit of DAC_CR must be set when APB1 clock speed is faster than 80 MHz. It adds an extra delay of three APB1 clock cycles to the transfer from DAC_DHRx register to DAC_DORx register ( $t_{\text{SETTLING}}$ ).

The DAC_DORx update rate is limited to 1/3 of APB1 clock frequency. When HFSEL bit is set, this rate is limited to 1/8 of the APB1 clock frequency.

When HFSEL is set, it is not allowed to write the DHRx register during a period of eight clock cycles after the ENx bit is set. During this period, making software/hardware triggering is not allowed either.

Figure 160. Timing diagram for conversion with trigger disabled TEN = 0

Timing diagram for conversion with trigger disabled (TEN = 0). The diagram shows three signals over time: APB1_CLK (a square wave), DHR (Digital-to-Analog Register), and DOR (Digital-to-Analog Output Register). The DHR signal is shown as a constant value of 0x1AC. The DOR signal is shown as a constant value of 0x1AC. The output voltage available on the DAC_OUT pin is shown as a curve that rises from a previous value to the new value of 0x1AC. The time interval between the DHR update and the output voltage reaching the new value is labeled t_SETTLING. The diagram is labeled ai14711c.

22.4.5 DAC output voltage

Digital inputs are converted to output voltages on a linear conversion between 0 and $V_{\text{REF+}}$ .

The analog output voltages on each DAC channel pin are determined by the following equation:

\text{DACoutput} = V_{\text{REF}} \times \frac{\text{DOR}}{4096}

22.4.6 DAC trigger selection

If the TENx control bit is set, the conversion can then be triggered by an external event (timer counter, external interrupt line). The TSELx[3:0] control bits determine which out of 16 possible events triggers the conversion as shown in TSELx[3:0] bits of the DAC_CR register. These events can be either the software trigger or hardware triggers. Refer to Table 147: DAC trigger selection .

Each time a DAC interface detects a rising edge on the selected trigger source (refer to the table below), the last data stored into the DAC_DHRx register are transferred into the DAC_DORx register. The DAC_DORx register is updated three APB1 cycles after the trigger occurs.

If the software trigger is selected, the conversion starts once the SWTRIG bit is set. SWTRIG is reset by hardware once the DAC_DORx register has been loaded with the DAC_DHRx register contents.

Note: TSELx[3:0] bit cannot be changed when the ENx bit is set.

When software trigger is selected, the transfer from the DAC_DHRx register to the DAC_DORx register takes only one APB1 clock cycle.

Table 147. DAC trigger selection

Source	Type	TSELx[3:0]
SWTRIG	Software control bit	0000
TIM1_TRGO	Internal signal from on-chip timers	0001
TIM2_TRGO	Internal signal from on-chip timers	0010
TIM4_TRGO	Internal signal from on-chip timers	0011
TIM5_TRGO	Internal signal from on-chip timers	0100
TIM6_TRGO	Internal signal from on-chip timers	0101
TIM7_TRGO	Internal signal from on-chip timers	0110
TIM8_TRGO	Internal signal from on-chip timers	0111
TIM15_TRGO	Internal signal from on-chip timers	1000
Reserved	-	1001
Reserved	-	1010
LPTIM1_OUT	Internal signal from on-chip timers	1011
LPTIM2_OUT	Internal signal from on-chip timers	1100
EXTI9	External pin	1101
Reserved	-	1110
Reserved	-	1111

22.4.7 DMA requests

Each DAC channel has a DMA capability. Two DMA channels are used to service DAC channel DMA requests.

When an external trigger (but not a software trigger) occurs while the DMAENx bit is set, the value of the DAC_DHRx register is transferred into the DAC_DORx register when the transfer is complete, and a DMA request is generated.

In dual mode, if both DMAENx bits are set, two DMA requests are generated. If only one DMA request is needed, only the corresponding DMAENx bit must be set. In this way, the application can manage both DAC channels in dual mode by using one DMA request and a unique DMA channel.

As DAC_DHRx to DAC_DORx data transfer occurred before the DMA request, the very first data has to be written to the DAC_DHRx before the first trigger event occurs.

DMA underrun

The DAC DMA request is not queued so that if a second external trigger arrives before the acknowledgment for the first external trigger is received (first request), then no new request is issued and the DMA channelx underrun flag DMAUDRx in the DAC_SR register is set, reporting the error condition. The DAC channelx continues to convert old data.

The software must clear the DMAUDRx flag by writing 1, clear the DMAEN bit of the used DMA stream and re-initialize both DMA and DAC channelx to restart the transfer correctly. The software must modify the DAC trigger conversion frequency or lighten the DMA workload to avoid a new DMA underrun. Finally, the DAC conversion could be resumed by enabling both DMA data transfer and conversion trigger.

For each DAC channelx, an interrupt is also generated if its corresponding DMAUDRIEx bit in the DAC_CR register is enabled.

22.4.8 Noise generation

In order to generate a variable-amplitude pseudonoise, an LFSR (linear feedback shift register) is available. DAC noise generation is selected by setting WAVEx[1:0] to 01. The preloaded value in LFSR is 0xAAA. This register is updated three APB1 clock cycles after each trigger event, following a specific calculation algorithm.

Figure 161. DAC LFSR register calculation algorithm

Diagram of the DAC LFSR register calculation algorithm. It shows a 12-bit shift register with cells numbered 11 down to 0. Feedback is taken from cells 11, 6, 4, and 1, which are inputs to a 4-input XOR gate. The output of the XOR gate is also input to a 12-input NOR gate. The output of the NOR gate is fed back into cell 11. Labels X^12, X^6, X^4, X, and X^0 are placed above cells 11, 6, 4, 1, and 0 respectively. A 12-bit bus symbol is shown below the register. The identifier ai14713c is in the bottom right corner.

The LFSR value, that may be masked partially or totally by means of the MAMPx[3:0] bits in the DAC_CR register, is added up to the DAC_DHRx contents without overflow and this value is then transferred into the DAC_DORx register.

If LFSR is 0x0000, a '1' is injected into it (antilock-up mechanism).

It is possible to reset LFSR wave generation by resetting the WAVEx[1:0] bits.

Figure 162. DAC conversion (SW trigger enabled) with LFSR wave generation

Timing diagram showing APB1_CLK, DHR, DOR, and SWTRIG signals. APB1_CLK is a periodic square wave. DHR is initially 0x00. DOR starts at 0xAAA and changes to 0xD55. SWTRIG is a pulse that triggers the change in DOR. Vertical dashed lines indicate timing relationships between the signals. The identifier ai14714b is in the bottom right corner.

Note: The DAC trigger must be enabled for noise generation by setting the TENx bit in the DAC_CR register.

22.4.9 Triangle-wave generation

It is possible to add a small-amplitude triangular waveform on a DC or slowly varying signal. DAC triangle-wave generation is selected by setting WAVEx[1:0] to 10 ¹¹. The amplitude is configured through the MAMPx[3:0] bits in the DAC_CR register. An internal triangle counter is incremented three APB1 clock cycles after each trigger event. The value of this counter is then added to the DAC_DHRx register without overflow and the sum is transferred into the DAC_DORx register. The triangle counter is incremented as long as it is less than the maximum amplitude defined by the MAMPx[3:0] bits. Once the configured amplitude is reached, the counter is decremented down to 0, then incremented again and so on.

It is possible to reset triangle wave generation by resetting the WAVEx[1:0] bits.

Figure 163. DAC triangle wave generation

Figure 163: DAC triangle wave generation graph showing a triangular waveform oscillating between a base value and a maximum amplitude. The rising slope is labeled 'Incrementation' and the falling slope is labeled 'Decrementation'. The y-axis is labeled 'MAMPx[3:0] max amplitude + DAC_DHRx base value' at the top and 'DAC_DHRx base value' at the bottom, with '0' at the origin. The x-axis has an arrow pointing right. The identifier 'ai14715c' is in the bottom right corner.

Figure 164. DAC conversion (SW trigger enabled) with triangle wave generation

Figure 164: Timing diagram for DAC conversion with triangle wave generation. It shows four signals over time: APB1_CLK (a square wave), DHR (Digital-to-Analog Register, holding 0xABE), DOR (Output Register, showing values 0xABE, 0xABF, 0xAC0), and SWTRIG (Software Trigger, a pulse). Vertical dashed lines indicate the sequence of events: a SWTRIG pulse followed by a DOR update, then another SWTRIG pulse followed by another DOR update. The identifier 'ai14716b' is in the bottom right corner.

Note:

The DAC trigger must be enabled for triangle wave generation by setting the TENx bit in the DAC_CR register.

The MAMPx[3:0] bits must be configured before enabling the DAC, otherwise they cannot be changed.

22.4.10 DAC channel modes

Each DAC channel can be configured in Normal mode or Sample and hold mode. The output buffer can be enabled to allow a high drive capability. Before enabling output buffer, the voltage offset needs to be calibrated. This calibration is performed at the factory (loaded after reset) and can be adjusted by software during application operation.

Normal mode

In Normal mode, there are four combinations, by changing the buffer state and by changing the DACx_OUTy pin interconnections.

To enable the output buffer, the MODEx[2:0] bits in DAC_MCR register must be:

• 000: DAC is connected to the external pin
• 001: DAC is connected to external pin and to on-chip peripherals

To disable the output buffer, the MODEx[2:0] bits in DAC_MCR register must be:

• 010: DAC is connected to the external pin
• 011: DAC is connected to on-chip peripherals

Sample and hold mode

In Sample and hold mode, the DAC core converts data on a triggered conversion, and then holds the converted voltage on a capacitor. When not converting, the DAC cores and buffer are completely turned off between samples and the DAC output is tri-stated, therefore reducing the overall power consumption. A stabilization period, which value depends on the buffer state, is required before each new conversion.

In this mode, the DAC core and all corresponding logic and registers are driven by the LSI low-speed clock in addition to the APB1 clock, allowing to use the DAC channels in deep low power modes such as Stop mode.

The LSI low-speed clock must not be stopped when the Sample and hold mode is enabled.

The sample/hold mode operations can be divided into 3 phases:

1. Sample phase: the sample/hold element is charged to the desired voltage. The charging time depends on capacitor value (internal or external, selected by the user). The sampling time is configured with the TSAMPLEx[9:0] bits in DAC_SHSRx register. During the write of the TSAMPLEx[9:0] bits, the BWSTx bit in DAC_SR register is set to 1 to synchronize between both clocks domains (APB and low speed clock) and allowing the software to change the value of sample phase during the DAC channel operation
2. Hold phase: the DAC output channel is tri-stated, the DAC core and the buffer are turned off, to reduce the current consumption. The hold time is configured with the THOLDx[9:0] bits in DAC_SHHR register
3. Refresh phase: the refresh time is configured with the TREFRESHx[7:0] bits in DAC_SHRR register

The timings for the three phases above are in units of LSI clock periods. As an example, to configure a sample time of 350 µs, a hold time of 2 ms and a refresh time of 100 µs assuming LSI ~32 KHz is selected:

12 cycles are required for sample phase: TSAMPLEx[9:0] = 11,
62 cycles are required for hold phase: THOLDx[9:0] = 62,
and 4 cycles are required for refresh period: TREFRESHx[7:0] = 4.

In this example, the power consumption is reduced by almost a factor of 15 versus Normal modes.

The formulas to compute the right sample and refresh timings are described in the table below, the Hold time depends on the leakage current.

Table 148. Sample and refresh timings

Buffer State	$t_{\text{SAMP}}^{(1)(2)}$	$t_{\text{REFRESH}}^{(2)(3)}$
Enable	$7 \mu\text{s} + (10 * R_{\text{BON}} * C_{\text{SH}})$	$7 \mu\text{s} + (R_{\text{BON}} * C_{\text{SH}}) * \ln(2 * N_{\text{LSB}})$
Disable	$3 \mu\text{s} + (10 * R_{\text{BOFF}} * C_{\text{SH}})$	$3 \mu\text{s} + (R_{\text{BOFF}} * C_{\text{SH}}) * \ln(2 * N_{\text{LSB}})$

1. In the above formula the settling to the desired code value with $\frac{1}{2}$ LSB or accuracy requires 10 constant time for 12 bits resolution. For 8 bits resolution, the settling time is 7 constant time.
2. $C_{\text{SH}}$ is the capacitor in Sample and hold mode.
3. The tolerated voltage drop during the hold phase "Vd" is represented by the number of LSBs after the capacitor discharging with the output leakage current. The settling back to the desired value with $\frac{1}{2}$ LSB error accuracy requires $\ln(2 * N_{\text{lsb}})$ constant time of the DAC.

Example of the sample and refresh time calculation with output buffer on

The values used in the example below are provided as indication only. Please refer to the product datasheet for product data.

C_{\text{SH}} = 100 \text{ nF}

V_{\text{DDA}} = 3.0 \text{ V}

Sampling phase:

t_{\text{SAMP}} = 7 \mu\text{s} + (10 * 2000 * 100 * 10^{-9}) = 2.007 \text{ ms}

(where $R_{\text{BON}} = 2 \text{ k}\Omega$ )

Refresh phase:

t_{\text{REFRESH}} = 7 \mu\text{s} + (2000 * 100 * 10^{-9}) * \ln(2 * 10) = 606.1 \mu\text{s}

(where $N_{\text{LSB}} = 10$ (10 LSB drop during the hold phase))

Hold phase:

D_V = i_{\text{leak}} * t_{\text{hold}} / C_{\text{SH}} = 0.0073 \text{ V} \text{ (10 LSB of 12bit at 3 V)}

i_{\text{leak}} = 150 \text{ nA} \text{ (worst case on the IO leakage on all the temperature range)}

t_{\text{hold}} = 0.0073 * 100 * 10^{-9} / (150 * 10^{-9}) = 4.867 \text{ ms}

Figure 165. DAC Sample and hold mode phase diagram

Figure 165. DAC Sample and hold mode phase diagram. The diagram illustrates the timing relationship between the DAC output voltage, the LSI clock, and the DAC internal state. The top graph shows the voltage level over time, transitioning through 'Sampling phase', 'Hold phase', 'Refresh phase', and another 'Sampling phase'. Voltage levels V1, Vd, and V2 are marked. The middle graph shows a continuous high-frequency LSI clock signal. The bottom graph shows the DAC state as 'ON' during Sampling and Refresh phases, and 'OFF' during the Hold phase.

Like in Normal mode, the Sample and hold mode has different configurations.

To enable the output buffer, MODEx[2:0] bits in DAC_MCR register must be set to:

• 100: DAC is connected to the external pin
• 101: DAC is connected to external pin and to on chip peripherals

To disable the output buffer, MODEx[2:0] bits in DAC_MCR register must be set to:

• 110: DAC is connected to external pin and to on chip peripherals
• 111: DAC is connected to on chip peripherals

When MODEx[2:0] bits are equal to 111, an internal capacitor, $C_{Lint}$ , holds the voltage output of the DAC core and then drive it to on-chip peripherals.

All Sample and hold phases are interruptible, and any change in DAC_DHRx immediately triggers a new sample phase.

Table 149. Channel output modes summary

		Mode	Buffer	Output connections
0	0	Normal mode	Enabled	Connected to external pin
0	1		Enabled	Connected to external pin and to on chip-peripherals (such as comparators)
1	0		Disabled	Connected to external pin
1	1		Disabled	Connected to on chip peripherals (such as comparators)

Table 149. Channel output modes summary (continued)

MODEx[2:0]			Mode	Buffer	Output connections
1	0	0	Sample and hold mode	Enabled	Connected to external pin
1	0	1		Enabled	Connected to external pin and to on chip peripherals (such as comparators)
1	1	0		Disabled	Connected to external pin and to on chip peripherals (such as comparators)
1	1	1		Disabled	Connected to on chip peripherals (such as comparators)

22.4.11 DAC channel buffer calibration

The transfer function for an N-bit digital-to-analog converter (DAC) is:

V_{out} = ((D/2^N) \times G \times V_{ref}) + V_{OS}

Where $V_{OUT}$ is the analog output, D is the digital input, G is the gain, $V_{ref}$ is the nominal full-scale voltage, and $V_{os}$ is the offset voltage. For an ideal DAC channel, $$ G = 1 $$ and $V_{os} = 0$ .

Due to output buffer characteristics, the voltage offset may differ from part-to-part and introduce an absolute offset error on the analog output. To compensate the $V_{os}$ , a calibration is required by a trimming technique.

The calibration is only valid when the DAC channelx is operating with buffer enabled (MODEx[2:0] = 000b or 001b or 100b or 101b). if applied in other modes when the buffer is off, it has no effect. During the calibration:

• The buffer output is disconnected from the pin internal/external connections and put in tristate mode (HiZ).
• The buffer acts as a comparator to sense the middle-code value 0x800 and compare it to VREF+/2 signal through an internal bridge, then toggle its output signal to 0 or 1 depending on the comparison result (CAL_FLAGx bit).

Two calibration techniques are provided:

• Factory trimming (default setting)

The DAC buffer offset is factory trimmed. The default value of OTRIMx[4:0] bits in DAC_CCR register is the factory trimming value and it is loaded once DAC digital interface is reset.

• User trimming

The user trimming can be done when the operating conditions differs from nominal factory trimming conditions and in particular when $V_{DDA}$ voltage, temperature, VREF+ values change and can be done at any point during application by software.

Note: Refer to the datasheet for more details of the Nominal factory trimming conditions

In addition, when $V_{DD}$ is removed (example the device enters in STANDBY or VBAT modes) the calibration is required.

The steps to perform a user trimming calibration are as below:

1. If the DAC channel is active, write 0 to ENx bit in DAC_CR to disable the channel.
2. Select a mode where the buffer is enabled, by writing to DAC_MCR register, MODEx[2:0] = 000b or 001b or 100b or 101b.
3. Start the DAC channelx calibration, by setting the CENx bit in DAC_CR register to 1.
4. Apply a trimming algorithm:
1. a) Write a code into OTRIMx[4:0] bits, starting by 00000b.
2. b) Wait for $t_{TRIM}$ delay.
3. c) Check if CAL_FLAGx bit in DAC_SR is set to 1.
4. d) If CAL_FLAGx is set to 1, the OTRIMx[4:0] trimming code is found and can be used during device operation to compensate the output value, else increment OTRIMx[4:0] and repeat sub-steps from (a) to (d) again.

The software algorithm may use either a successive approximation or dichotomy techniques to compute and set the content of OTRIMx[4:0] bits in a faster way.

The commutation/toggle of CAL_FLAGx bit indicates that the offset is correctly compensated and the corresponding trim code must be kept in the OTRIMx[4:0] bits in DAC_CCR register.

Note: A $t_{TRIM}$ delay must be respected between the write to the OTRIMx[4:0] bits and the read of the CAL_FLAGx bit in DAC_SR register in order to get a correct value. This parameter is specified into datasheet electrical characteristics section.

If $V_{DDA}$ , VREF+ and temperature conditions do not change during device operation while it enters more often in standby and VBAT mode, the software may store the OTRIMx[4:0] bits found in the first user calibration in the flash or in back-up registers. then to load/write them directly when the device power is back again thus avoiding to wait for a new calibration time.

When CENx bit is set, it is not allowed to set ENx bit.

22.4.12 Dual DAC channel conversion modes (if dual channels are available)

To efficiently use the bus bandwidth in applications that require the two DAC channels at the same time, three dual registers are implemented: DHR8RD, DHR12RD and DHR12LD. A unique register access is then required to drive both DAC channels at the same time. For the wave generation, no accesses to DHRxxxD registers are required. As a result, two output channels can be used either independently or simultaneously.

11 conversion modes are possible using the two DAC channels and these dual registers. All the conversion modes can nevertheless be obtained using separate DHRx registers if needed.

All modes are described in the paragraphs below.

Independent trigger without wave generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2.
2. Configure different trigger sources by setting different values in the TSEL1 and TSEL2 bitfields.
3. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a DAC channel1 trigger arrives, the DHR1 register is transferred into DAC_DOR1 (three APB1 clock cycles later).

When a DAC channel2 trigger arrives, the DHR2 register is transferred into DAC_DOR2 (three APB1 clock cycles later).

Independent trigger with single LFSR generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2.
2. Configure different trigger sources by setting different values in the TSEL1 and TSEL2 bitfields.
3. Configure the two DAC channel WAVEx[1:0] bits as 01 and the same LFSR mask value in the MAMPx[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a DAC channel1 trigger arrives, the LFSR1 counter, with the same mask, is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). Then the LFSR1 counter is updated.

When a DAC channel2 trigger arrives, the LFSR2 counter, with the same mask, is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). Then the LFSR2 counter is updated.

Independent trigger with different LFSR generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2.
2. Configure different trigger sources by setting different values in the TSEL1 and TSEL2 bitfields.
3. Configure the two DAC channel WAVEx[1:0] bits as 01 and set different LFSR masks values in the MAMP1[3:0] and MAMP2[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a DAC channel1 trigger arrives, the LFSR1 counter, with the mask configured by MAMP1[3:0], is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). Then the LFSR1 counter is updated.

When a DAC channel2 trigger arrives, the LFSR2 counter, with the mask configured by MAMP2[3:0], is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). Then the LFSR2 counter is updated.

Independent trigger with single triangle generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2.
2. Configure different trigger sources by setting different values in the TSEL1 and TSEL2 bitfields.
3. Configure the two DAC channel WAVEx[1:0] bits as 1x and the same maximum amplitude value in the MAMPx[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a DAC channel1 trigger arrives, the DAC channel1 triangle counter, with the same triangle amplitude, is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). The DAC channel1 triangle counter is then updated.

When a DAC channel2 trigger arrives, the DAC channel2 triangle counter, with the same triangle amplitude, is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). The DAC channel2 triangle counter is then updated.

Independent trigger with different triangle generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2.
2. Configure different trigger sources by setting different values in the TSEL1 and TSEL2 bits.
3. Configure the two DAC channel WAVEx[1:0] bits as 1x and set different maximum amplitude values in the MAMP1[3:0] and MAMP2[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a DAC channel1 trigger arrives, the DAC channel1 triangle counter, with a triangle amplitude configured by MAMP1[3:0], is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). The DAC channel1 triangle counter is then updated.

When a DAC channel2 trigger arrives, the DAC channel2 triangle counter, with a triangle amplitude configured by MAMP2[3:0], is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). The DAC channel2 triangle counter is then updated.

Simultaneous software start

To configure the DAC in this conversion mode, the following sequence is required:

• Load the dual DAC channel data to the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

In this configuration, one APB1 clock cycle later, the DHR1 and DHR2 registers are transferred into DAC_DOR1 and DAC_DOR2, respectively.

Simultaneous trigger without wave generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2.
2. Configure the same trigger source for both DAC channels by setting the same value in the TSEL1 and TSEL2 bitfields.
3. Load the dual DAC channel data to the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a trigger arrives, the DHR1 and DHR2 registers are transferred into DAC_DOR1 and DAC_DOR2, respectively (after three APB1 clock cycles).

Simultaneous trigger with single LFSR generation

1. To configure the DAC in this conversion mode, the following sequence is required:
2. Set the two DAC channel trigger enable bits TEN1 and TEN2.
3. Configure the same trigger source for both DAC channels by setting the same value in the TSEL1 and TSEL2 bitfields.
4. Configure the two DAC channel WAVEx[1:0] bits as 01 and the same LFSR mask value in the MAMPx[3:0] bits.
5. Load the dual DAC channel data to the desired DHR register (DHR12RD, DHR12LD or DHR8RD).

When a trigger arrives, the LFSR1 counter, with the same mask, is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). The LFSR1 counter is then updated. At the same time, the LFSR2 counter, with the same mask, is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). The LFSR2 counter is then updated.

Simultaneous trigger with different LFSR generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2
2. Configure the same trigger source for both DAC channels by setting the same value in the TSEL1 and TSEL2 bitfields.
3. Configure the two DAC channel WAVEx[1:0] bits as 01 and set different LFSR mask values using the MAMP1[3:0] and MAMP2[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a trigger arrives, the LFSR1 counter, with the mask configured by MAMP1[3:0], is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). The LFSR1 counter is then updated.

At the same time, the LFSR2 counter, with the mask configured by MAMP2[3:0], is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). The LFSR2 counter is then updated.

Simultaneous trigger with single triangle generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2
2. Configure the same trigger source for both DAC channels by setting the same value in the TSEL1 and TSEL2 bitfields.
3. Configure the two DAC channel WAVEx[1:0] bits as 1x and the same maximum amplitude value using the MAMPx[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a trigger arrives, the DAC channel1 triangle counter, with the same triangle amplitude, is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB1 clock cycles later). The DAC channel1 triangle counter is then updated.

At the same time, the DAC channel2 triangle counter, with the same triangle amplitude, is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). The DAC channel2 triangle counter is then updated.

Simultaneous trigger with different triangle generation

To configure the DAC in this conversion mode, the following sequence is required:

1. Set the two DAC channel trigger enable bits TEN1 and TEN2
2. Configure the same trigger source for both DAC channels by setting the same value in the TSEL1 and TSEL2 bitfields.
3. Configure the two DAC channel WAVEx[1:0] bits as 1x and set different maximum amplitude values in the MAMP1[3:0] and MAMP2[3:0] bits.
4. Load the dual DAC channel data into the desired DHR register (DAC_DHR12RD, DAC_DHR12LD or DAC_DHR8RD).

When a trigger arrives, the DAC channel1 triangle counter, with a triangle amplitude configured by MAMP1[3:0], is added to the DHR1 register and the sum is transferred into DAC_DOR1 (three APB clock cycles later). Then the DAC channel1 triangle counter is updated.

At the same time, the DAC channel2 triangle counter, with a triangle amplitude configured by MAMP2[3:0], is added to the DHR2 register and the sum is transferred into DAC_DOR2 (three APB1 clock cycles later). Then the DAC channel2 triangle counter is updated.

22.5 DAC in low-power modes

Table 150. Effect of low-power modes on DAC

Mode	Description
Sleep	No effect, DAC used with DMA
Low-power run	No effect.
Low-power sleep	No effect. DAC used with DMA.
Stop 0 / Stop 1	The DAC remains active with a static value if the Sample and hold mode is selected using LSI clock.
Stop 2	The DAC registers content is kept. The DAC must be disabled before entering Stop 2.

Table 150. Effect of low-power modes on DAC (continued)

Mode	Description
Standby	The DAC peripheral is powered down and must be reinitialized after exiting Standby or Shutdown mode.
Shutdown

22.6 DAC interrupts

Table 151. DAC interrupts

Interrupt acronym	Interrupt event	Event flag	Enable control bit	Interrupt clear method	Exit Sleep mode	Exit Stop mode	Exit Standby mode
DAC	DMA underrun	DMAUDRx	DMAUDRI Ex	Write DMAUDRx = 1	Yes	No	No

22.7 DAC registers

Refer to Section 1 on page 86 for a list of abbreviations used in register descriptions.

The peripheral registers have to be accessed by words (32-bit).

22.7.1 DAC control register (DAC_CR)

Address offset: 0x00

Reset value: 0x0000 0000

31	30	29	28	27	26	25	24	23	22	21	20	19	18	17	16
Res.	CEN2	DMAUDRIE2	DMAEN2	MAMP2[3:0]				WAVE2[1:0]		TSEL2[3]	TSEL2[2]	TSEL2[1]	TSEL2[0]	TEN2	EN2
	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw

15	14	13	12	11	10	9	8	7	6	5	4	3	2	1	0
HFSEL	CEN1	DMAUDRIE1	DMAEN1	MAMP1[3:0]				WAVE1[1:0]		TSEL1[3]	TSEL1[2]	TSEL1[1]	TSEL1[0]	TEN1	EN1
rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw	rw

Bit 31 Reserved, must be kept at reset value.

Bit 30 CEN2 : DAC channel2 calibration enable

This bit is set and cleared by software to enable/disable DAC channel2 calibration, it can be written only if EN2 bit is set to 0 into DAC_CR (the calibration mode can be entered/exit only when the DAC channel is disabled) Otherwise, the write operation is ignored.

0: DAC channel2 in Normal operating mode

1: DAC channel2 in calibration mode

Note: This bit is available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bit 29 DMAUDRIE2 : DAC channel2 DMA underrun interrupt enable

This bit is set and cleared by software.

0: DAC channel2 DMA underrun interrupt disabled

1: DAC channel2 DMA underrun interrupt enabled

Note: This bit is available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bit 28 DMAEN2 : DAC channel2 DMA enable

This bit is set and cleared by software.

0: DAC channel2 DMA mode disabled

1: DAC channel2 DMA mode enabled

Note: This bit is available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bits 27:24 MAMP2[3:0]: DAC channel2 mask/amplitude selector

These bits are written by software to select mask in wave generation mode or amplitude in triangle generation mode.

0000: Unmask bit0 of LFSR/ triangle amplitude equal to 1

0001: Unmask bits[1:0] of LFSR/ triangle amplitude equal to 3

0010: Unmask bits[2:0] of LFSR/ triangle amplitude equal to 7

0011: Unmask bits[3:0] of LFSR/ triangle amplitude equal to 15

0100: Unmask bits[4:0] of LFSR/ triangle amplitude equal to 31

0101: Unmask bits[5:0] of LFSR/ triangle amplitude equal to 63

0110: Unmask bits[6:0] of LFSR/ triangle amplitude equal to 127

0111: Unmask bits[7:0] of LFSR/ triangle amplitude equal to 255

1000: Unmask bits[8:0] of LFSR/ triangle amplitude equal to 511

1001: Unmask bits[9:0] of LFSR/ triangle amplitude equal to 1023

1010: Unmask bits[10:0] of LFSR/ triangle amplitude equal to 2047

≥ 1011: Unmask bits[11:0] of LFSR/ triangle amplitude equal to 4095

Note: These bits are available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bits 23:22 WAVE2[1:0]: DAC channel2 noise/triangle wave generation enable

These bits are set/reset by software.

00: wave generation disabled

01: Noise wave generation enabled

1x: Triangle wave generation enabled

Note: Only used if bit TEN2 = 1 (DAC channel2 trigger enabled)

These bits are available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bits 21:18 TSEL2[3:0]: DAC channel2 trigger selection

These bits select the external event used to trigger DAC channel2

0000: SWTRIG2

0001: dac_ch2_trg1

0010: dac_ch2_trg2

...

1111: dac_ch2_trg15

Refer to the trigger selection tables in Section 22.4.6: DAC trigger selection for details on trigger configuration and mapping.

Note: Only used if bit TEN2 = 1 (DAC channel2 trigger enabled).

These bits are available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bit 17 TEN2: DAC channel2 trigger enable

This bit is set and cleared by software to enable/disable DAC channel2 trigger

0: DAC channel2 trigger disabled and data written into the DAC_DHR2 register are transferred one APB1 clock cycle later to the DAC_DOR2 register

1: DAC channel2 trigger enabled and data from the DAC_DHR2 register are transferred three APB1 clock cycles later to the DAC_DOR2 register

Note: When software trigger is selected, the transfer from the DAC_DHR2 register to the DAC_DOR2 register takes only one APB1 clock cycle.

These bits are available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bit 16 EN2 : DAC channel2 enable

This bit is set and cleared by software to enable/disable DAC channel2.

0: DAC channel2 disabled

1: DAC channel2 enabled

Note: These bits are available only on dual-channel DACs. Refer to Section 22.3: DAC implementation .

Bit 15 HFSEL : High frequency interface mode enable

This bit is set and cleared by software to enable/disable DAC interface high speed mode

This bit need to be set when APB1 clock frequency is higher than 80 MHz.

0: High frequency interface mode disabled

1: High frequency interface mode enabled

Bit 14 CEN1 : DAC channel1 calibration enable

This bit is set and cleared by software to enable/disable DAC channel1 calibration, it can be written only if bit EN1 = 0 into DAC_CR (the calibration mode can be entered/exit only when the DAC channel is disabled) Otherwise, the write operation is ignored.

0: DAC channel1 in Normal operating mode

1: DAC channel1 in calibration mode

Bit 13 DMAUDRIE1 : DAC channel1 DMA Underrun Interrupt enable

This bit is set and cleared by software.

0: DAC channel1 DMA Underrun Interrupt disabled

1: DAC channel1 DMA Underrun Interrupt enabled

Bit 12 DMAEN1 : DAC channel1 DMA enable

This bit is set and cleared by software.

0: DAC channel1 DMA mode disabled

1: DAC channel1 DMA mode enabled

Bits 11:8 MAMP1[3:0] : DAC channel1 mask/amplitude selector