Thinking of using XMEGA A3 - Errata Beware

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Looking at the latest datasheet - there are many NEW errata published for the same revision device.
Some of these are quite serious and will undoubtedly cause you problems and delays in your projects - they did mine! The green ones are from the earlier edition.
I've no idea when or if Atmel intends to resolve these.

Bandgap voltage input for the ACs can not be changed when used for both ACs simultaneously
"¢ VCC voltage scaler for AC is non-linear
"¢ ADC has increased INL error for some operating conditions
ADC gain stage output range is limited to 2.4 V"¢ ADC Event on compare match non-functional
Bandgap measurement with the ADC is non-functional when VCC is below 2.7V
"¢ Accuracy lost on first three samples after switching input to ADC gain stage
"¢ Configuration of PGM and CWCM not as described in XMEGA A Manual
"¢ PWM is not restarted properly after a fault in cycle-by-cycle mode
BOD will be enabled at any reset
"¢ DAC is nonlinear and inaccurate when reference is above 2.4V or VCC - 0.6V
DAC has increased INL or noise for some operating conditions
"¢ DAC refresh may be blocked in S/H mode
"¢ Conversion lost on DAC channel B in event triggered mode
"¢ EEPROM page buffer always written when NVM DATA0 is written
"¢ Pending full asynchronous pin change interrupts will not wake the device
Pin configuration does not affect Analog Comparator Output
"¢ NMI Flag for Crystal Oscillator Failure automatically cleared
Writing EEPROM or Flash while reading any of them will not work
"¢ Crystal start-up time required after power-save even if crystal is source for RTC
"¢ RTC Counter value not correctly read after sleep
"¢ Pending asynchronous RTC-interrupts will not wake up device
"¢ TWI Transmit collision flag not cleared on repeated start
"¢ Clearing TWI Stop Interrupt Flag may lock the bus
"¢ TWI START condition at bus timeout will cause transaction to be dropped
"¢ TWI Data Interrupt Flag (DIF) erroneously read as set
"¢ WDR instruction inside closed window will not issue reset

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Someone ought to compile an errata league table for all the manufacturers - number of errata and how long they have been outstanding. It would make interesting reading. I'm not volunteering. :)

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The A1 also got a datasheet Errata update recently. They now list 42 issues for the H revision.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

doc8068-Revision Q-page 99:

Quote:
Some ADC configurations or operating condition will result in increased INL error.
In differential mode INL is increased to:
– 6 LSB for sample rates above 1 Msps, and up to 8 LSB for 2 Msps sample rate.
– 6 LSB for reference voltage below 1.1V when VCC is above 3.0V.
– 20 LSB for ambient temperature below 0 degree C and reference voltage below 1.3V.
In single ended mode, the INL is increased up to a factor of 3 for the conditions above.

doc8068-Revision S-page 104:

Quote:
Some ADC configurations or operating condition will result in increased INL error.
In signed mode INL is increased to:
– 6 LSB for sample rates above 1 Msps, and up to 8 LSB for 2 Msps sample rate.
– 6 LSB for reference voltage below 1.1V when VCC is above 3.0V.
– 20 LSB for ambient temperature below 0 degree C and reference voltage below 1.3V.
In unsigned mode, the INL is increased up to a factor of 3 for the conditions above.

Which are correct?

Ozhan KD
Knowledge is POWER

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'd assume that the later one was correct.

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

At least it's a signal that ATMEL has not abandoned the XMEGAs...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

22. Pending asynchronous RTC-interrupts will not wake up device
Asynchronous Interrupts from the Real-Time-Counter that is pending when the sleep instruction is executed, will be ignored until the device is woken from another source or the source triggers again.

OMW Atmel! So much for ultra low power apps that need to keep track of time! Pleeeeeeeeze fix these!!! Soon!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There was a nasty eratta with writing to the eeprom on the 256a3 (doesn't seem to efect the 128a3), it wouldn't write while the cpu was running. Fix is to enable interrupt on write complete, start the write, sleep the cpu, cpu wakes up on write complete. I verified this fix works.