PGA305: read EEPROM register

Hello Andrea,

The EEPROM space is actually part of a different slave address. If I2C_ADDR = 0, then you access the EEPROM for reads by using the slave address 0x45:

For a read directly from the EEPROM, you will just use the EEPROM address shown in the register map 0x46. Note that even though Table 20 shows the PADC_GAIN_MSB as a "Control and Status Register" since it has an EEPROM address, you will access it by using the EEPROM register slave address.

Regards,

ok, as I just read all the register listed in table 20, i can use for all of them 0x46 as slave address and the proper register map as register address? Even those register who have DI Page and DI offset?

My final purpose is to copy on my MCU almost all of the register listed in table 20.

So I can read the register directly from the EEPROM using:

• slave address 0x45
• register address: the last two bytes of the EEPROM addres listed in table 20

Do I need to write the page i would like to read in the EEPROM_PAGE_ADDRESS before doing that?
How can i read the register without the EEPROM addres of table 20? Should I use the DI Offset? DI offset shoud be interpreted like the offset from 0x40 00 00 00?

Thank you

Only the EEPROM registers can be read using slave address 0x45.

To write to the EEPROM, you have to use the EEPROM cache. Please refer to the "EEPROM Programming Procedure" section of the datasheet. There are 16 pages of 8 bytes each, and you can write a single page with the contents of the EEPROM cache. The page of a specific register is determined by using the last byte of the EEPROM address and counting up from 0x0. So 0x0 through 0x07 will be on page 0, 0x08 - 0x0F will be on page 1, etc.

If the register is also available outside of the EEPROM, you can use the DI offset while using slave address 0x40.

Regards,

Ok, the writing process was ok, now the reading is almost fine. I can't read the EEPROM_CRC_VALUE, i use 0x45 as slave address and as EEPROM ADDRESS 0x7F but it always return Zero, even after trying to writedifferent values into EEPROM.

Thank you for your support, is very precious.

Hi Scott,

I'm here again, I'm having some problem with I2C that delayed my last question.
The CRC placed in the register 0x8D is calculated from PGA if I set the CALCULATE_CRC_bit and is calculated every time I write any page of the EEPROM (page 36) even if I don't set the CALCULATE_CRC_bit?
In the second case I need to wirte the new CRC every time I wrote a page, correct?

Last thing, for reading the CRC calculated from PGA I just need to set 0x42 as slave address and 0x8D ad register address correct? I can't test it due to my I2C issue..

I'm confused because I haven't understood the difference between offset address and eeprom address. EEPROM_CRC_VALUE and a few have both.

Thank you very much, after that i think all my issue can be considered solved.

I've read some device had issues.

The marking on my PGA is:

PGA305A

RHH

TI 96I

A5GV G4

Hello Andrea,

There should not be any problem with the I2C on this part. What issue are you seeing? 

Yes you will need to write a new CRC value in the last byte of the EEPROM any time you change any register in the EEPROM. The calculated EEPROM CRC value is actually still on the same interface page as the EEPROM, so you will write to slave address 0x45 and read register 0x8D.

Some EEPROM registers have the same function as some of the control and status registers. When you set it in EEPROM it will set that control and status register to the saved EEPROM value every time you power on the PGA305. That is why there is an offset (used to access the control and status register) and an EEPROM address (used to read and write to the EEPROM).

We are in the process of updating the PGA305 datasheet to make this more clear. The PGA300 is a very similar device, and has an updated datasheet that explains the register map in a more coherent way. I recommend checking the PGA300 datasheet to see how the EEPROM registers are different from the control and status registers.

Regards,

Scott Cummins said:

There should not be any problem with the I2C on this part. What issue are you seeing? 

I'am having problems reading the register, unfortunatly we use a uC with any I2C port so we emulated it with PWM. I try to debug with an oscilloscope and it seems fine. Sometimes I successfully read the register some other no.

Anyway really thanks for your support so far.

Regards

Yes, i put the PGA in the Reset mode. It reads casually some registers and usually different, not always the same.

Here the register 0x35, it should read 132 (decimal). The second picture read 0x54 and sould read 0x12

In the upper post you will see that the receive byte has 9 bit instead of 8. In this way it works wometimes.
I tried to modify for receiving 8 byte as the I2C protocol say but in that version it never read. i post a screen for this too.