ATmega328P - what to do without full swing crystal oscillator ?

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

Hello all !

 

I am running a project on an 'old', pre-Rev.K, ATmega328P using a full swing crystal oscillator clocked @ 11.0592 MHz.

This makes a perfect UART @ 230800 bps for a bluetooth module, which works a treat.

 

I was very concerned reading about that Atmel has recently stripped the full swing crystal oscillator feature from

recent ATmega328P chips, after Rev.K. :(((

 

What are my options now ? Should I change the crystal + 2 caps with a dedicated oscillator chip and connect its

output to XTAL1, together with new flags CKSEL=0000 & CK=10 ? I will be greatful if someone could confirm !
 

Last Edited: Sun. Nov 13, 2016 - 08:37 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

But the change is only evident if you switch from using 328/328P to the new 328PB. If you stick with the 328P (the "Arduino processor") then nothing changes? 

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

From my understanding, this change will not only affect the 328PB chips but also the newly sold 328P ones !

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

Have you tried your device with one of the newer parts?

If you don't know my whole story, keep your mouth shut.

If you know my whole story, you're an accomplice. Keep your mouth shut. 

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

What evidence do you have for it affecting existing 328P? For one thing if Atmel/Mchp made such a radical change they'd have to change silicon/fuses/model number.  Tons of existing designs still in manufacture use and rely on every aspect of the original 328P design. 

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

ATmega328-328P datasheet in Errata section page440

 

Errata ATmega328/P

The revision letter in this section refers to the revision of the ATmega328/P device.

Rev. K

1 – Full swing crystal oscillator not supported

The full swing crystal oscillator functionality is not available in revision K.

Fix/Workaround:

Use alternative clock sources available in the device.

 

P.Ashok Kumar

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

Exactly that

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

vassilispapanikolaou wrote:

I am running a project on an 'old', pre-Rev.K, ATmega328P using a full swing crystal oscillator clocked @ 11.0592 MHz.

 

I was very concerned reading about that Atmel has recently stripped the full swing crystal oscillator feature from

recent ATmega328P chips, after Rev.K. :(((

 

What are my options now ? ..

 

My reading of that, is you are fine if you are under 16MHz, as the alternative of the Low Power Crystal Oscillator works to 16MHz 

 

However, if you were using a Full Swing 20MHz Crystal Design, you are 'dropped in it' by this change.

 

Crystals are analog in nature, so it may be a Low-C, Low R, focused part selection 20MHz unit can still work.

Atmel/Microchip have not yet spec'd that. 

 

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

Thanks for all answers, however I am not sure if the low power oscillator will actually work accurately enough to

support 230800 baud in USART.

 

However, is the suggested alternative (external oscillator) a correct choice ?

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

if it's for a new design and you really need full speed, then use a external oscillator.

 

But if it's just for the 230400 (not 230800 I guess), then all speeds that are n*1,8432 MHz (n integer) will work

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

Oops, sorry, its 230400, my mistake ! Thanks !

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

vassilispapanikolaou wrote:

Thanks for all answers, however I am not sure if the low power oscillator will actually work accurately enough to

support 230800 baud in USART.

 

However, is the suggested alternative (external oscillator) a correct choice ?

 

? Whatever makes you think the  Low Power Crystal Oscillator  has accuracy problems ?

 

At your 11.0592MHz, to future proof this, you just need to change modes to  Low Power Crystal Oscillator 

No other changes should be required, and this is easy enough to try.

 

Does anyone know what a "after Rev.K" part actually does, if you still try to run the Full Swing Crystal Oscillator ?

ie how much of a nasty surprise will some users get ?

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

Who-me wrote:
Does anyone know what a "after Rev.K" part actually does, if you still try to run the Full Swing Crystal Oscillator ?
See other thread - the is no such thing as "rev K" - it was cancelled.