Tool/software:
Hi,
I'm dealing with a communication issue with the TMP1826 chip. I believe the timing is correct, as the chip reset, device detection, as well as data writing and reading are all working. This is supported by the fact that when I read SCRATCHPAD-1, the CRC always matches. The problem is that the chip returns data which appears to be invalid. Below is the result of the READ SCRATCHPAD-1 operation (RESET, WRITE 0xcc, WRITE 0xbe, READ 9 bytes, check CRC):
[0] 0x43
[1] 0x80
[2] 0x0c
[3] 0xff
[4] 0x00
[5] 0x00
[6] 0x00
[7] 0xff
[8] 0x8e
Calculated CRC: 0x8e
Status reg (0x0c = 0b00001100):
ALERT_HIGH R/RC 0
ALERT_LOW R/RC 0
RESERVED R 00
DATA_READY RC 1
POWER_MODE R 1
ARB_DONE R 0
LOCK_STATUS R 0
Configuration-1 reg (0x00 = 0b00000000):
TEMP_FMT RW 0
RESERVED RW 0
CONV_TIME_SEL RW 0
ALERT_MODE RW 0
AVG_SEL RW 0
CONV_MODE_SEL RW 000
Access Type Codes:
R Read
RC Read to Clear
R-0 Read returns 0s
W Write
W0CP Write 0 to clear
TEMP: 0x8043 = 32835 = 0b1000000001000011
There are several things I don't understand:
- Why do bits 5:4 (RESERVED) in the Status Register not have the value 1?
- Why does bit 2 (POWER_MODE) have the value 1 when the chip is powered via VDD?
- But most importantly: how do I convert the Temperature Register LSB and MSB (0x43 and 0x80) into a temperature in degrees Celsius?
The chip is on a PCB in a room at approximately 25 degrees Celsius. The Temperature Register sometimes reads values such as 0x8040, 0x8043, 0x8077, and so on. The values respond to warming and cooling of the chip.
I have two identical boards populated with the TMP1826 chip, and on both, the chip behaves similarly. The data read has a correct CRC, but it doesn't make sense.
I have read the datasheet, as well as application notes SBOA542 and SBAA588A, but I still can't work out what I'm doing wrong.
The chip is connected as follows:
There was no UART available when we decided to add the TMP1826 chip to the PCB, so I had to resort to bit-banging.
Thanks in advance for any advice!
- Ross
