UHD Daughter Board EEPROM Format (calibration data)

GNU Radio
1

I am wondering if there has been any formalized structure for storing
the IQ
balance offset data in the dboard EEPROM.

The following snippet is taken from the UHD that is packaged with GNU
radio
3.5.1
(I’m not actually sure if the UHD version is different from the GNU
radio
package 3.7?)

////////////////////////////////////////////////////////////////////////
// format of daughterboard EEPROM
// 00: 0xDB code for ``I’m a daughterboard’’
// 01: … Daughterboard ID (LSB)
// 02: … Daughterboard ID (MSB)
// 03: … io bits 7-0 direction (bit set if it’s an output from
m’board)
// 04: … io bits 15-8 direction (bit set if it’s an output from
m’board)
// 05: … ADC0 DC offset correction (LSB)
// 06: … ADC0 DC offset correction (MSB)
// 07: … ADC1 DC offset correction (LSB)
// 08: … ADC1 DC offset correction (MSB)
// …
// 1f: … negative of the sum of bytes [0x00, 0x1e]

#define DB_EEPROM_MAGIC 0x00
#define DB_EEPROM_MAGIC_VALUE 0xDB
#define DB_EEPROM_ID_LSB 0x01
#define DB_EEPROM_ID_MSB 0x02
#define DB_EEPROM_REV_LSB 0x03
#define DB_EEPROM_REV_MSB 0x04
#define DB_EEPROM_OFFSET_0_LSB 0x05 // offset correction for ADC or DAC
0
#define DB_EEPROM_OFFSET_0_MSB 0x06
#define DB_EEPROM_OFFSET_1_LSB 0x07 // offset correction for ADC or DAC
1
#define DB_EEPROM_OFFSET_1_MSB 0x08
#define DB_EEPROM_SERIAL 0x09
#define DB_EEPROM_SERIAL_LEN 0x09 //9 ASCII characters
#define DB_EEPROM_CHKSUM 0x1f

#define DB_EEPROM_CLEN 0x20 // length of common portion of
eeprom

#define DB_EEPROM_CUSTOM_BASE DB_EEPROM_CLEN // first avail offset for
// daughterboard
specific
use
////////////////////////////////////////////////////////////////////////

The storage of calibration data for DC offset is stored in 0x05 - 0x08.

But what about the IQ offsets?
There is a gap between reserved addresses 0x09 and 0x1f.
Is the use of this area documented?

There is a utility to calibrate the board and it generates a text file.
Presumably, the authors also implemented a method to make use of this
data?

Continuing, the ettus boards driver code sets the gain, though I do not
see
any request for the offset data stored in EEPROM inside the constructor
or
elsewhere.

This is a similar issue with IQ balance, ie. formalized structure for
both
storing calibration data in EEPROM and then accessing it and using it
during
runtime.

–
View this message in context:
http://gnuradio.4.n7.nabble.com/UHD-Daughter-Board-EEPROM-Format-calibration-data-tp44187.html
Sent from the GnuRadio mailing list archive at Nabble.com.

2

Hello Alexander,
UHD does not come bundled with GNU Radio; they are two completely
independent libraries.
You’re sending some code from UHD, but from the defines below I’m not
quite sure if you’re a) referring to the firmware code or b) referring
to the host (PC) code.
What file exactly is this from?
Anyway, this has nothng to do with GNU Radio as such, but with the Code
for Ettus USRPs.
Since your question is quite complicated, please join the USRP users
mailing list () and write your question to [email protected],
the folks there will be glad to help you.

Greetings,
Marcus

3

Ah yes… I have become used to viewing them as one system.

Thank you for the link to the USRP mailing list, I will explore the
issue
with them.

Alex

–
View this message in context:
http://gnuradio.4.n7.nabble.com/UHD-Daughter-Board-EEPROM-Format-calibration-data-tp44187p44202.html
Sent from the GnuRadio mailing list archive at Nabble.com.