The present disclosure relates to digital-to-analog converters (DACs) that store configuration and address information, and input data in non-volatile memory; and more particularly to, multi-channel DACs having non-volatile memory and using serial communication protocols over conventional serial interfaces, e.g., I2C, SPI, USB, SCIO, UNI/O, etc.
Present technology DAC devices store configuration information and input data in volatile memory. The configuration information and input data stored in the volatile memory are lost when operating power is removed from the DAC device and the associated volatile memory. For example, a DAC device may be used to output a programmable analog voltage. The programming bits, e.g., digital representation of the analog voltage, are stored in a DAC register which is volatile, thereby loosing its contents when powered down. Upon an initial power-up of the DAC device, the DAC register is either cleared or its contents are not predictable until the DAC register is programmed again. Thus the DAC register must be reprogrammed each time the DAC device is powered up. This necessitates additional program cycles of a master controller program so as to reprogram the DAC register. In many applications, DAC devices support operation of other devices in a system. For example, the DAC device may provide a reference voltage to other devices for proper operation thereof. Since the DAC register has to be reprogrammed, all other devices dependent upon the DAC device must wait (prevented from operating) until the DAC register contains the correct data.
DAC devices are becoming more prevalent in integrated circuits having both analog and digital functions, e.g., mixed signal devices. Typically, mixed signal devices (slaves) will communicate with a master device such as master control unit (MCU), e.g., microcontroller, microprocessor, digital signal processor, etc., over a communications bus. There may be more then one mixed signal slave device connected to the communications bus, thus each one of the mixed signal slave devices will need a device address. Generally the mixed signal device has either multiple address programming pins on the integrated circuit package, or the mixed signal device has a fixed address that is mask programmed during fabrication at the factory. For a three bit address, up to eight different integrated circuit fabrication masks are required. Having to mask program up to eight different addresses into mixed signal devices that are otherwise identical increases manufacturing time and costs, and results in having to stock and ship up to eight different parts. In addition, having mixed signal devices with non-field programmable addresses may become very inconvenient in certain applications.
Therefore there is a need to prevent loss of the DAC device configuration and address information, and input data during a power down or power loss condition. If the DAC device outputs a preprogrammed output by itself immediately when it turns-on, then the overall system application reduces several initialization and calibration steps, and can thereby initialize the system with the same conditions all the time, even when there are power interruptions thereto. This will increase system operating efficiency and the useful range of applications for such a device.
A DAC device may have both volatile and non-volatile internal memory blocks. The non-volatile memory may be used to store configuration information, digital voltage values, e.g., data, and an address for the DAC device. The non-volatile memory may be for example, but is not limited to, electrically erasable and programmable read only memory (EEPROM), FLASH memory and the like. This data may be written into the internal non-volatile memory block at any time and the stored configuration information, digital data, e.g., voltage values; and DAC device address may thereby be protected from being lost during a power outage.
According to teachings of this disclosure, a non-volatile memory, e.g., EEPROM, FLASH, etc., may be part of the DAC device. The DAC/non-volatile memory device may thereby provide a preprogrammed output voltage whenever it is powered-up. However, unlike the conventional DAC devices, DAC devices with the non-volatile memory may need special interface communication protocols for effective operation of the DAC device. For example, the system master controller unit (MCU) requires a way to access the volatile memory (DAC register) and/or the non-volatile memory (e.g., EEPROM). Therefore, the non-volatile memory in the DAC device requires effective interface communication protocols with the MCU so that the MCU may access the DAC device's memories (both non-volatile and volatile memories) effectively. Since most of the mixed signal devices such as DAC, analog-to-digital (ADC), and digital potentiometer are operated by using a standard serial interface, e.g., I2C, SPI, USB, SCIO, UNI/O, etc., the interface communication protocols for communicating with the volatile and non-volatile memories of the DAC device may operate without violating the specifications of the existing serial communications protocols.
According to teachings of this disclosure, a serial data interface communication protocol may be used to operate the DAC device and the internal non-volatile memory over a serial data bus, e.g., I2C, SPI, USB, SCIO, UNI/O, etc. For example, but not limited to, one, two, three or four channel 12 bit DAC devices with non-volatile memory, wherein these DAC devices may incorporate the same non-volatile interface communication protocol. Using an interface communication protocol solves the following problems: (a) A user may read and/or write the configuration, address and data information into non-volatile or volatile memories with a simple command(s). This also reduces the interface communication time. (b) A simple and yet effective command structure reduces the complexity of device interface circuits. (c) The same command structure may be used for reading from and writing to device test registers using the same integrated circuit package pin-out connections, thus eliminating the need for extra test interfaces.
It is advantageous to use a common serial communications protocol for mixed signal devices, e.g., both analog and digital circuit functions, even if different companies use their own interface protocols, a common protocol may emerge that is driven by customer demand.
An example problem: DAC device outputs a programmable analog voltage. The programming of bits are stored in a volatile DAC register which means its memory contents are cleared or not predictable at an initial power-up stage until it is reprogrammed. The user must reprogram the DAC register each time when it is powered-up. This requires that the system's master controller unit (MCU) use additional cycles to reprogram the DAC register. In many applications, DAC devices are used as a supporting device for other devices in systems. For example, the DAC device may provide a reference voltage to other devices to operate. If the DAC device outputs a preprogrammed output by itself immediately when it turns-on, then the overall application systems reduce several initialization and calibration steps, and can initialize the systems with the same conditions all the time, even when there are power interruptions thereto. This will increase systems efficiency and useful operability greatly.
According to the teachings of this disclosure, a solution to the aforementioned problem may be by having a non-volatile memory, e.g., EEPROM, FLASH, etc., as part of the DAC device. The DAC/non-volatile memory device may thereby provide a preprogrammed output voltage whenever it is powered-up. However, there may be a communications issue for controlling the non-volatile memory. For example, the system MCU needs a way to access the volatile memory (DAC Register) or non-volatile memory (EEPROM). Therefore, the non-volatile memory in the device requires effective interface communication protocols between the DAC device and the MCU. This allows the MCU to access the device's memories (both non-volatile and volatile memories) effectively. Since most of the mixed signal devices such as a DAC, ADC, and/or a digital potentiometer are operated by using a standard serial interface, e.g., I2C, SPI and the like, the necessary interface communication protocol may be operated within these standard serial interface specifications. The interface communication protocols disclosed herein do not violate existing serial communication specifications.
A more complete understanding of the present disclosure may be acquired by referring to the following description taken in conjunction with the accompanying drawings wherein:
a-11d illustrate schematic byte diagrams of an address, command and data protocol structures for reading in normal mode the DAC input register and non-volatile memory of one or more DAC devices;
While the present disclosure is susceptible to various modifications and alternative forms, specific example embodiments thereof have been shown in the drawings and are herein described in detail. It should be understood, however, that the description herein of specific example embodiments is not intended to limit the disclosure to the particular forms disclosed herein, but on the contrary, this disclosure is to cover all modifications and equivalents as defined by the appended claims.
Referring now to the drawings, the details of specific example embodiments are schematically illustrated. Like elements in the drawings will be represented by like numbers, and similar elements will be represented by like numbers with a different lower case letter suffix.
Referring to
The non-volatile memory 114 may be, but is not limited to, an electrically erasable and programmable read only memory (EEPROM), FLASH memory, etc. For example, the non-volatile memory 114 may be used to store configuration register, DAC input data (e.g., 12 bits of a digital representation of the analog voltage that the DAC 108 is supposed to produce), address bits (e.g., 3 bits for I2C address) and test mode trim bits. The charge pump 116 may be used for writing to the non-volatile memory 114. Power may be supplied to the device 100 at voltage terminals Vdd and Vss. The serial interface and address logic 102 is coupled to a serial data bus 120 of n-bit width, e.g., n=1, 2, 3, etc. Configuration and data values may be written to or read from the non-volatile memory 114 and/or the input register 104. The serial interface and address logic 102 determines whether the device 100 is being addressed by a bus master (not shown) over the serial bus 120. A specific programmable device address allows specific operation and selection from a number of devices 100. The device address is written to and stored in the non-volatile memory 114 so that a specific device address is maintained even when the device 100 has power removed therefrom. The DAC output register 106 may be loaded from the input register 104. Also, the contents of the non-volatile memory 114 may be transferred to the input register 104.
A load DAC output register input 122 may be used to (a) transfer the contents of the input register(s) 104 to the respective DAC output register(s) 106, (b) select a device 100 of interest in Read/Write Address Bit Commands, and (c) enter test mode. When there is a logic transition on the input 122, the contents of the input register(s) 104 may be loaded into the DAC output register(s) 108, thus generating a new analog voltage at the output of the analog amplifier 112 (Vout).
For selection of a device 100 of interest, a logic transition on the input 122 at a certain time during a command may be used to either read a device address or write a new device address into the selected device 100. Using the input 122 in this way allows determining the address programmed into a specific device 100 and also being able to change a specific device address when the device 100 is in an end use system without requiring removal of the device 100 and/or special test fixtures.
To enter a device test mode, a higher then normal voltage, e.g., 10 volts, may be applied to the input 122.
A ready/busy output 124 may be used to indicate when a write operation to the non-volatile memory 114 is completed.
Referring to
The non-volatile memory 114 may be, but is not limited to, electrically erasable and programmable read only memory (EEPROM), FLASH memory, etc. For example, the non-volatile memory 114 may be used to store configuration register, DAC input data (e.g., 12 bits of a digital representation of the analog voltage that the DAC 108 is supposed to produce), address bits (e.g., 3 bits for I2C address) and test mode trim bits. The charge pump 116 may be used for writing to the non-volatile memory 114. Power may be supplied to the device 100 at voltage terminals Vdd and Vss. The serial interface and logic 102 may have an input (/LDAC) for transferring DAC settings from serial input latches to output latches, e.g., DAC registers 106. The serial interface and logic 102 is coupled to a serial data bus of n-bit width, e.g., n=1, 2, 3, etc. Configuration and data values may be written to or read from the non-volatile memory 114 and/or the input registers 104. The DAC registers 106 may be loaded from the respective input registers 104. Also the contents of the non-volatile memory 114 may be transferred to the respective input registers 104. The load DAC output register input 122 and the ready/busy output 124 function as described hereinabove.
A normal mode of the device allows user commands to write to and read from the device DAC registers and non-volatile memory during normal operational thereof.
Normal Mode Write Commands
Referring to
The fast mode write command shown in
The fast mode write command is used to sequentially update the input register(s) 104. The power down selection bits (PD1, PD0) 314 and the 12 bits of DAC input data bits (D11-D0) 316 and 318 are sequentially updated for each DAC channel (bytes for three DAC channels are shown in
The fast mode write command writes only the power-down selection bits 314 (PD1 and PD0) of the configuration register, and 12 bits (D11:D0) of DAC input data 316 and 318 of each DAC channel. The write data is loaded sequentially from the first channel to the last channel of the device. Each of the DAC input registers 104 are updated (written) at the acknowledge pulse of that channel's last input data byte. Once the DAC input registers 104 are loaded, the DAC registers 106 and Vout from each of the amplifiers 112 are updated at any time by changing the logic level at the load DAC output register input 122 (/LDAC). Non-volatile memory 114 is not affected.
Referring to
The multiple write command shown in
The multiple write command writes to one DAC input register 104 at a time. The DAC channel may be selected by using the DAC register selection bits 426 (DAC1 and DAC0), and only that channel is affected. More than one DAC register 106 may be written to by sending repeat bytes, e.g., bytes 400e, 400f and 400g, with the respective DAC register selection bits 426 for each DAC channel to be updated. Data in the non-volatile memory 114 is not changed by the fast write command shown in
The configuration register bits: vref bit 432a, power down selection bits 414a (PD1 and PD0) and DAC gain selection bit 440a (GX); and DAC input data bits (D11-D0) 422a and 424a may be updated after the fourth byte 400d acknowledge 410d if the logic level at the load DAC output register input 122 (/LDAC) is low or /UDAC bit 430 is cleared. Vout from the respective amplifier 112 may be updated using the /UDAC bit 430, a logic change at the DAC output register input 122 (/LDAC), or through a general call software update. When /UDAC bit 430 is at a first logic level, the Vout of the selected DAC channel is updated as soon as the fourth byte 400d is acknowledged 410d (the last byte of the selected DAC register) regardless of the logic state of the DAC output register input 122 (/LDAC). The DAC input data bits (D11-D0) 422 and 424 are the DAC input data of the selected DAC channel (bits 426). Bytes 400b, 400c and 400d are repeated for each of the next DAC channels to be updated but without the need for specifying the write command type bits 412a (C2, C1 and C0) and write function bits 428a (W1 and W0), X=don't care. For example, bytes 400e, 400f and 400g represent the update data necessary for a second DAC channel. Subsequent DAC channels may be similarly updated.
The information contained in the second, third and fourth bytes described above may be repeated for each DAC channel. Write command protocols for two instances of DAC channels are shown, however, write protocols for any number of DAC channels are contemplated herein (e.g.,
Referring to
The sequential write command shown in
The sequential write command writes the configuration register bits: vref bit 532, power down selection bits 514 (PD1 and PD0), DAC gain selection bit 540 (GX), and DAC input data bits (D11-D0) 522 and 524 to the DAC input registers 104 sequentially from the starting DAC channel to the last DAC channel, and this command also writes the same data sequentially to the non-volatile memory 114. The starting DAC channel is determined by the DAC register selection bits 526 (DAC1 and DAC0) in byte 500b. Subsequent DAC channels are written after the completion of each respective byte pair, e.g., bytes 500e and 500f for a second DAC channel. Additional DAC channels are just repeats of byte pairs up to the maximum number of DAC channels of the device 200.
When writing to the non-volatile memory 114, the ready/busy output 124 (
Referring to
The single write command shown in
The single write command writes the configuration register bits: vref bit 632, power down selection bits 614 (PD1 and PD0) and DAC gain selection bit 640 (GX), and the DAC input data bits (D11-D0) 622 and 624 to the DAC input register 104 for the DAC channel specified by the DAC channel selection bits 626 (DAC1 and DAC0), and also writes the same information into associated locations in the non-volatile memory 114.
Referring to
The writing of a new address command shown in
The writing of a new address command is only valid if there is a logic level transition at the load DAC output register input 122 (/LDAC) during the slave acknowledge 710b of the second byte 700b, and the logic level at the input 122 remains in its new state for at least the end of the third byte 700c. The load DAC output register input 122 (/LDAC) may be used to select a device when programming a new address therein.
Referring to
The change voltage reference select bits command shown in
Referring to
The write power-down select bits command is used to select either a normal or power down mode for each of the DAC channels. Two power-down bits 980 (PD1, PD0) are used for each of the DAC channels, e.g., DAC channels A, B, C and D. When a normal mode is selected for a DAC channel, there will an analog voltage output. When a power down mode is selected, there will be no analog voltage output, instead a fixed resistance value to ground or common will be substituted depending upon the logic values of the two power-down bits 980 (PD1, PD0).
The write power-down select bits command shown in
Referring to
The write gain selection bits to the DAC input registers command begins with a start bit 1002, followed by bytes 1000a and 1000b (8 bits each), each byte 1000 is followed by a device (slave) acknowledge 1010, and terminates with a stop bit 1020. A first byte 1000a comprises a device code 1004, a device address 1006 (A2, A1, A0) and a read/write bit 1008 set to zero. A second byte 1000b comprises write command type bits 1012 (C2=1, C1=1 and C0=0), and gain selection bits 1040. Each of the gain selection bits 1040 may be used to select the gain for its respective DAC channel, e.g., gain of one when the gain selection bit 1040 is at a first logic level or gain of two when at a second logic level. Non-volatile memory 114 is not affected by this command.
Read Command and Output Data Format
If the read/write bit (X08) in the first byte of each command is set to a logic “high” (1), then the device enters a read mode. There are two types of read modes: (a) normal read mode for reading register data, and (b) test mode read for accessing lock and trim bits. The test mode read may be entered into by asserting a high voltage on the load DAC output register input 122 (/LDAC). The read command executes only when the ready/busy output 124 indicates that the non-volatile memory 114 is not busy.
Referring to
Next a first slave byte 1150a is sent by the slave device. The first slave byte 1150a comprises the present status of the following data contained in the DAC register 106 of DAC channel A (
After the slave device receives the master acknowledge 1160a, a second slave byte 1150b is sent by the slave device. The second slave byte 1150b comprises the following data contained in the DAC register 106 of DAC channel A (
After the slave device receives the master acknowledge 1160b, a third slave byte 1150c is sent by the slave device. The third slave byte 1150c comprises the eight (8) least significant data bits 1124a (D7:D0) contained in the DAC register 106 associated with the DAC channel indicated in the selection bits 1126a. Bytes 1150a-1150c comprise the present contents of the DAC register of the indicated DAC channel 1126a (DAC1, DAC0). After the third slave byte 1150c has been read by the bus master, the bus master sends a master acknowledge 1160c.
Then the addressed slave device sends a fourth slave byte 1150d indicating the present status of the data contents contained in the non-volatile memory 114. The fourth slave byte 1150d comprises a ready/busy bit 1196a that indicates the completion status of a write to the non-volatile memory 114 (e.g., logic 1 indicates write complete, logic 0 indicates otherwise), the present status of the associated power-on-reset bit 1198a, the DAC channel indicated in the selection bits 1176a (DAC1, DAC0), a zero (0), and the three address bits 1156b (A2, A1, A0). After the fourth slave byte 1150d has been read by the bus master, the bus master sends a master acknowledge 1160d.
After the slave device receives the master acknowledge 1160d, a fifth slave byte 1150e is sent by the slave device. The fifth slave byte 1150e comprises the following data contained in the non-volatile memory 114: Vref 1182a status, power down selection bits 1164a (PD1 and PD0) status, the DAC gain selection bit 1190a (GX) status, and the four most significant DAC data bits 1172a (D11:D8) associated with the DAC channel indicated in the selection bits 1176a. After the fifth slave byte 1150e has been read by the bus master, the bus master sends a master acknowledge 1160e.
After the slave device receives the master acknowledge 1160e, a sixth slave byte 1150f is sent by the slave device. The sixth slave byte 1150f comprises the eight (8) least significant data bits 1174a (D7:D0) contained in the non-volatile memory 114 associated with the DAC channel indicated in the selection bits 1176a. Bytes 1150d-1150f comprise the present contents of the non-volatile memory 114 of the indicated DAC channel 1176a (DAC1, DAC0). After the sixth slave byte 1150f has been read by the bus master, the bus master sends a master acknowledge 1160f, and a stop bit 1120a is asserted on the serial bus 120.
b shows the seventh (7th) through the twelfth (12th) slave bytes 1150g-1150l that may be used to supply all of the previously mentioned status and data for the next DAC channel B (
A test mode of the device may entered into when the device receives a “Read/Write Command for Test Mode” while a high voltage, e.g., about 10 volts, is applied to the load DAC output register input 122 (/LDAC). No specific device address is required since only the device under test will have a high voltage at its input 122.
Test Mode Write Commands
Referring to
Referring to
Referring to
Referring to
Test Mode Read Command
Referring to
Next a first slave byte 1650a is sent by the slave device. The first slave byte 1650a comprises the present status of the following data contained in the DAC register 106 of DAC channel A (
After the slave device receives the master acknowledge 1660a, a second slave byte 1650b is sent by the slave device. The second slave byte 1650b comprises the following data contained in the DAC register 106: Bandgap absolute value trim bits 1677 (Vbg3, Vbg2, Vbg1 and Vbg0), and Bandgap amplifier offset trim bits 1684 (Bba3, Bba2, Bba1 and Bba0). After the second slave byte 1650b has been read by the bus master, the bus master sends a master acknowledge 1660b.
After the slave device receives the master acknowledge 1660b, a third slave byte 1650c is sent by the slave device. The third slave byte 1650c comprises buffer amplifier A offset value trim bits 1686a and buffer amplifier B offset value trim bits 1686b. The fourth, fifth and sixth slave bytes 1650d, 1650e and 1650f, respectively, present the same types of data stored in the non-volatile memory 114 as read by the master in slave bytes 1650a, 1650b and 1650c above. After the sixth slave byte 1650f has been read by the bus master, the bus master sends a master acknowledge 1660f, and a stop bit 1620 is asserted on the serial bus 120.
Referring to
As shown in
As shown in
As shown in
As shown in
Referring to
While embodiments of this disclosure have been depicted, described, and are defined by reference to example embodiments of the disclosure, such references do not imply a limitation on the disclosure, and no such limitation is to be inferred. The subject matter disclosed is capable of considerable modification, alteration, and equivalents in form and function, as will occur to those ordinarily skilled in the pertinent art and having the benefit of this disclosure. The depicted and described embodiments of this disclosure are examples only, and are not exhaustive of the scope of the disclosure.
This application is a continuation-in-part and claims priority to commonly owned U.S. patent application Ser. No. 12/017,582, filed Jan. 22, 2008; entitled “Read and Write Interface Communications Protocol for Digital-to-Analog Signal Converter with Non-Volatile Memory,” by Thomas Youbok Lee, Jonathan Jackson, John Austin, Andrew Swaneck and Yann Johner; and claims priority to commonly owned U.S. Provisional Patent Application Ser. No. 61/021,448; filed Jan. 16, 2008; entitled “Read and Write Interface Communications Protocol for Digital-to-Analog Signal Converter with Non-Volatile Memory,” by Thomas Youbok Lee, Yann Johner, Philippe Gimmel, Tim Sherman, Jonathan Jackson and John Austin; both are hereby incorporated by reference herein for all purposes.
Number | Date | Country | |
---|---|---|---|
61021448 | Jan 2008 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 12017582 | Jan 2008 | US |
Child | 12336726 | US |