I was looking through the XTag2 schematic and I noticed something strange.
The Oscillator/crystal used is indicated to be 13Mhz, this seemed an odd choice to me, not the usual frequencies you would expect to drive the L1. Then I realised it must be to do with the USB interface, but then shouldn't that be a 12Mhz crystal/oscillator rather than a 13Mhz, as is often the case with USB?
Hoping someone could explain if this is a documentation error or something I am missing?
XTag2 query
-
- XCore Legend
- Posts: 1274
- Joined: Thu Dec 10, 2009 10:20 pm
-
- XCore Expert
- Posts: 546
- Joined: Thu Dec 10, 2009 10:41 pm
- Location: St. Leonards-on-Sea, E. Sussex, UK.
It's high-speed USB - 480 Mbit/s.
Leon
Leon
-
- Experienced Member
- Posts: 66
- Joined: Fri Dec 11, 2009 3:38 pm
This is not an error.
The SMSC USB3318 phy needs a 13MHz reference clock. The L1 has a clock mode setting where the PLL can be used a 13MHz input to make 399.75MHz core clock frequency, so it makes sense to use one cheap pierce oscillator to drive both.
If the SMSC USB3317 phy is used, then it would need a 26MHz reference clock.
Regards,
Corin
The SMSC USB3318 phy needs a 13MHz reference clock. The L1 has a clock mode setting where the PLL can be used a 13MHz input to make 399.75MHz core clock frequency, so it makes sense to use one cheap pierce oscillator to drive both.
If the SMSC USB3317 phy is used, then it would need a 26MHz reference clock.
Regards,
Corin
-
- XCore Legend
- Posts: 1274
- Joined: Thu Dec 10, 2009 10:20 pm
Thanks Guys, I thought it might be to do with the USB.
Wow 399.75MHz is a really weird frequency requirement, I wonder where that came from anyone know why or the history?
regards
Al
Wow 399.75MHz is a really weird frequency requirement, I wonder where that came from anyone know why or the history?
regards
Al
-
- XCore Expert
- Posts: 546
- Joined: Thu Dec 10, 2009 10:41 pm
- Location: St. Leonards-on-Sea, E. Sussex, UK.
It's not a requirement, it's just the closest to the standard 400 MHz that can be achieved with a 13 MHz input.
Leon
Leon
-
- XCore Legend
- Posts: 1274
- Joined: Thu Dec 10, 2009 10:20 pm
Sorry Leon my bad, I thought the 399.75MHz was for the USB not the XCore, Doh!
So in this case there using 399.75MHz to clock the Xmos core. I guess as long as your not running anything else on the Xtag2 that requires specific timing it's not an issue. That also explains the clock complications on the usb audio reference board.
regards
Al
So in this case there using 399.75MHz to clock the Xmos core. I guess as long as your not running anything else on the Xtag2 that requires specific timing it's not an issue. That also explains the clock complications on the usb audio reference board.
regards
Al
-
- XCore Addict
- Posts: 165
- Joined: Wed Feb 10, 2010 2:32 pm
The L1 can achieve exactly 400MHz from a 13MHz input frequency. The input frequency can be divided by 13 to get 1MHz, then multiplied up to get 400MHz exactly. 399.75MHz is the closest that can be obtained avoiding an input divider.
-
- Member
- Posts: 8
- Joined: Thu Feb 11, 2010 7:46 pm
Such as?Folknology wrote:That also explains the clock complications on the usb audio reference board.
-
- XCore Legend
- Posts: 1274
- Joined: Thu Dec 10, 2009 10:20 pm
Just The fact that it uses 3 separate clocks
-
- Member
- Posts: 8
- Joined: Thu Feb 11, 2010 7:46 pm
audio master clocks?