Tips on selecting a crystal for an ATTiny841

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

I've built a number of AVR projects, but I've always used either the internal or an external oscillator.  I'm starting a project now that I'd like to use an external crystal with and Digikey lets me filter on load capacitance and ESR.  I've found the Atmel recommendations for the capacitors to use with external crystals, but nothing on load capacitance or ESR.  Can anyone point me to some info on this?  Thanks.

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

Try Atmel A.N AVR042.

1) Studio 4.18 build 716 (SP3)
2) WinAvr 20100110
3) PN, all on Doze XP... For Now
A) Avr Dragon ver. 1
B) Avr MKII ISP, 2009 model
C) MKII JTAGICE ver. 1

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

Thanks!  I'll check it out.

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

Actually, that note doesn't answer my question (though it's poorly edited and seems to have been written by someone for whom English is a second language).  Any other suggestions? 

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

lautman wrote:
I've found the Atmel recommendations for the capacitors to use with external crystals, but nothing on load capacitance or ESR.  Can anyone point me to some info on this?  Thanks.

 

Usually the caps to use, are designed to sum to the vendors load capacitance.

ESR is not a critical spec, but lower ESR usually means better Q  - some Oscs will spec a maximum ESR they can tolerate.

 

So you should be fine filtering by package and price first, then you add C's to suit the Vendors spec & it is always a good idea to actually verify the Xtal frequency via a Frequency counter on a Pin. Most common, is for designs to underestimate the equivalent total C.

I've seen cases where 15pF / 18pF (ie similar but unequal values)  were used to give a better average point.

 

Last Edited: Tue. Jun 23, 2015 - 05:14 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks.  The ATTiny841 datasheet suggests each crystal cap be between 12 pf and 22 pf.  Would I choose a crystal with a load capacitance of 24 pf - 44 pf?

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

lautman wrote:

Thanks.  The ATTiny841 datasheet suggests each crystal cap be between 12 pf and 22 pf.  Would I choose a crystal with a load capacitance of 24 pf - 44 pf?

 

Nope, not quite - see

https://en.wikipedia.org/wiki/Cr...

and especially

https://en.wikipedia.org/?title=...

 

where it says "Load capacitance CL can be calculated from the series combination of C1 and C2, taking into account Ci and Co, the input and output capacitance of the inverter, and Cs, the stray capacitances from the oscillator, PCB layout, and crystal case (typically 3-9 pF)"  - note they fail to mention any miller capacitance effects, so assume Ci and Co include that, but not all vendors measure the pin C, in an active bias state.

 So you can see there is plenty of room for error creepage effects, hence the suggestion to verify by measurement if precision matters to you.

 

 If you do not need super-small, I like these ones, as they are cheap, and look cute too :)

http://www.digikey.com/product-d...

 

& if you want to pay a little more for size/precision, then the EPSON TSX-3225 series look a good choice.

Last Edited: Tue. Jun 23, 2015 - 05:50 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

lautman wrote:

Actually, that note doesn't answer my question (though it's poorly edited and seems to have been written by someone for whom English is a second language).  Any other suggestions? 

It does talk about load capacitance...AND how to calculate it.

1) Studio 4.18 build 716 (SP3)
2) WinAvr 20100110
3) PN, all on Doze XP... For Now
A) Avr Dragon ver. 1
B) Avr MKII ISP, 2009 model
C) MKII JTAGICE ver. 1

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

I've seen cases where 15pF / 18pF (ie similar but unequal values)  were used to give a better average point.

I thought the asymmetry was also sometimes used to help insure the oscillator actually oscillates on power up, and doesn't lock up.

That said, it adds to the parts list, and I think the likelihood of the osc not oscillating on power up, even with "matched" caps, is incredibly small.

 

JC