| Electronic Components Datasheet Search |
|
LTC6820 Datasheet(PDF) 49 Page - Analog Devices |
|
|
|||||||||||||||||||||||||||||
LTC6820 Datasheet(HTML) 49 Page - Analog Devices |
|
49 / 80 page ![]() LTC2949 49 Rev A For more information www.analog.com REGISTER DESCRIPTION Table 29. FIFO Register Read Format. DATA BYTE NAME DESCRIPTION 0 SN[15:8] MSB of measured sample N 1 SN[7:0] LSB of measured sample N 2 TAGN Tag of sample N 3 SN+1[15:8] MSB of measured sample N+1 4 SN+1[7:0] LSB of measured sample N+1 5 TAGN+1 Tag of sample N+1 … … … 3*M SN+M[15:8] MSB of measured sample N+M 3*M+1 SN+M[7:0] LSB of measured sample N+M 3*M+2 TAGN+M Tag of sample N+M The column Data Byte does not count the PEC bytes, which depend on the setting of data bytes per PEC (ID-byte for DCMD or fixed to six for RDCV). For maximum data throughput it is recommended to read the FIFOs with DCMDs with 16 bytes per PEC and multiples of 16 samples (= 48 data bytes + 3 • 2 PEC bytes). Table 30. FIFO TAG Definitions. TAG NAME DESCRIPTION 0x00 OK Valid, new sample 0x55 RDOVR Read overrun, sample was already read 0xAA WROVR Write overrun, FIFO was filled completely and at least one sample was already overwritten It is recommended to connect LTC2949 always in parallel to a daisy chain. Also, in the scenario where LTC2949 is con- nected at one end of a reversible isoSPI chain, the default communication should be done with direct read commands. Only in case the direct isoSPI link to LTC2949 fails, the communication would be routed through the daisy chain. In this configuration where LTC2949 is on top of a daisy chain, after the broadcast RDCV command, the stacked LTC68xx Cell Monitors turn into a cascaded shift register, in which data is shifted through each device to the next device in the stack. In this scenario, at the end of the read transaction, there is always a fixed number of samples stuck in the shift register, that is never transmitted to the master. Several approaches are possible to avoid or minimize the loss of samples. Either fast continuous mode is stopped (FACONV=0), the FIFO is read until empty and then fast continuous measurement is repeated (FACONV=1). This results in a time window without measurements, equal to the time it takes to read the samples. For example, at 1Mbit/s serial clock rate, it takes 0.8 ms (approximately one fast conversion time) to read 24 samples (four byte command and PEC, 72 data bytes, 24 data PEC bytes equals in total 100 bytes or 800 bits). For lower clock speeds or longer cycle times, where more samples are read per cycle, the following approach is more efficient. Alternatively, the FIFO read burst must be long enough to always empty the FIFO and read at least one sample with TAG RDOVR. Still, also here, one sample with TAG OK can arrive after the RDOVR and would then be stuck in the daisy chain’s shift register. If the latency of the shift register (number of devices in the daisy chain times eight bytes times eight bit per byte divided by SPI clock speed) is longer than one fast conversion time, even more than one sample can get stuck in the daisy chain. |
|
Link URL |
| Does ALLDATASHEET help your business so far? [ DONATE ] |
About Alldatasheet | Advertisement | Contact us | Privacy Policy | Link to Datasheet | Link Exchange | Manufacturer List All Rights Reserved©Alldatasheet.com |
| Russian : Alldatasheetru.com | Korean : Alldatasheet.co.kr | Spanish : Alldatasheet.es | French : Alldatasheet.fr | Italian : Alldatasheetit.com Portuguese : Alldatasheetpt.com | Polish : Alldatasheet.pl | Vietnamese : Alldatasheet.vn Indian : Alldatasheet.in | Mexican : Alldatasheet.com.mx | British : Alldatasheet.co.uk | New Zealand : Alldatasheet.co.nz |
|
Family Site : ic2ic.com |
icmetro.com |