mirror of
https://github.com/RobTillaart/Arduino.git
synced 2024-10-03 18:09:02 -04:00
141 lines
4.6 KiB
Markdown
141 lines
4.6 KiB
Markdown
|
|
[](https://github.com/marketplace/actions/arduino_ci)
|
|
[](https://github.com/RobTillaart/AD5680/actions/workflows/arduino-lint.yml)
|
|
[](https://github.com/RobTillaart/AD5680/actions/workflows/jsoncheck.yml)
|
|
[](https://github.com/RobTillaart/AD5680/issues)
|
|
|
|
[](https://github.com/RobTillaart/AD5680/blob/master/LICENSE)
|
|
[](https://github.com/RobTillaart/AD5680/releases)
|
|
[](https://registry.platformio.org/libraries/robtillaart/AD5680)
|
|
|
|
|
|
# AD5680
|
|
|
|
Arduino library for the AD5680 series digital analog convertor (18 bit).
|
|
|
|
|
|
## Description
|
|
|
|
**Experimental**
|
|
|
|
The AD5680 is a 18 bit DAC on paper but in fact it is a 16 bit DAC.
|
|
To get the 18 bits a 2 bit duty cycle is used to implement the last two bits.
|
|
This means that the output voltage will be fluctuating unless the two
|
|
least significant bits are 0 (zero). To dampen this fluctuation the datasheet
|
|
advises to add a LOW pass filter after the output.
|
|
|
|
Feedback, issues, improvements are welcome.
|
|
Please file an issue on GitHub.
|
|
|
|
#### Related
|
|
|
|
- https://github.com/RobTillaart/AD56x8 (multi channel)
|
|
- https://github.com/RobTillaart/AD568X (single channel lower resolution)
|
|
- https://github.com/RobTillaart/MCP_DAC (SPI interface)
|
|
|
|
|
|
## Interface
|
|
|
|
```cpp
|
|
#include "AD5680.h"
|
|
```
|
|
|
|
### Base class
|
|
|
|
- **AD5680(uint8_t slaveSelect)** constructor base class, sets HW SPI.
|
|
Sets internal values to zero.
|
|
- **AD5680(uint8_t spiData, uint8_t spiClock, uint8_t slaveSelect)** constructor,
|
|
sets SW SPI.
|
|
Sets internal values to zero.
|
|
- **begin()** initializes the SPI and sets internal state.
|
|
|
|
|
|
### Set DAC
|
|
|
|
- **bool setValue(uint16_t value)** set value to the output immediately,
|
|
effectively a prepare + update in one call.
|
|
Returns false if value out of range.
|
|
- **uint16_t getValue()** returns set value.
|
|
At power up the AD5680 will be reset to 0 (== 0 volt).
|
|
- **bool setPercentage(float percentage)** idem.
|
|
- **float getPercentage()** idem.
|
|
|
|
|
|
#### SPI
|
|
|
|
- **void setSPIspeed(uint32_t speed)** sets SPI clock in **Hz**,
|
|
please read datasheet of the ADC first to get optimal speed.
|
|
- **uint32_t getSPIspeed()** gets current speed in **Hz**.
|
|
- **bool usesHWSPI()** returns true if HW SPI is used.
|
|
|
|
|
|
#### SPI ESP32 specific
|
|
|
|
("inherited" from MPC_DAC library)
|
|
|
|
- **void selectHSPI()** in case hardware SPI, the ESP32 has two options HSPI and VSPI.
|
|
- **void selectVSPI()** see above.
|
|
- **bool usesHSPI()** returns true if HSPI is used.
|
|
- **bool usesVSPI()** returns true if VSPI is used.
|
|
|
|
The **selectVSPI()** or the **selectHSPI()** needs to be called
|
|
BEFORE the **begin()** function.
|
|
|
|
(experimental)
|
|
- **void setGPIOpins(uint8_t clk, uint8_t miso, uint8_t mosi, uint8_t select)**
|
|
overrule GPIO pins of ESP32 for hardware SPI. Needs to be called AFTER the **begin()** function.
|
|
|
|
Note: earlier experiments (other device) shows that on a ESP32
|
|
SW-SPI is equally fast as HW-SPI and in fact a bit more stable.
|
|
The SW pulses are a bit slower than the HW pulses and therefore more square.
|
|
The HW-SPI has some overhead SW-SPI hasn't.
|
|
|
|
|
|
## Performance
|
|
|
|
Measurements with AD5680_demo.ino - setValue() most important.
|
|
(numbers are rounded).
|
|
|
|
| version | board | clock | SPI | samples / second | Notes |
|
|
|:---------:|:-------:}:---------:|:-----:}:------------------:|:--------|
|
|
| 0.1.1 | UNO | 16 MHz | HW | 53500 |
|
|
| 0.1.1 | UNO | 16 MHz | SW | 2800 |
|
|
| 0.1.1 | ESP32 | 240 MHz | HW | 91000 | 1
|
|
| 0.1.1 | ESP32 | 240 MHz | SW | 111000 |
|
|
|
|
|
|
1. ESP32 HW is equal performant for HSPI and VSPI.
|
|
Unknown why HW SPI is 20% slower than SW SPI (transaction overhead?)
|
|
|
|
50000 - 100000 samples per second means that a 1 KHz wave can be
|
|
constructed with 50-100 values per period.
|
|
|
|
|
|
## Future
|
|
|
|
#### Must
|
|
|
|
- improve documentation
|
|
- get test hardware
|
|
- test the library
|
|
|
|
#### Should
|
|
|
|
- add examples
|
|
|
|
#### Could
|
|
|
|
- make a 16 bit interface?
|
|
|
|
#### Wont
|
|
|
|
|
|
## Support
|
|
|
|
If you appreciate my libraries, you can support the development and maintenance.
|
|
Improve the quality of the libraries by providing issues and Pull Requests, or
|
|
donate through PayPal or GitHub sponsors.
|
|
|
|
Thank you,
|
|
|