Clocking SX48: Resonator or oscillator

Old_Lady

Hi folks -

I have a quick question:· which is preferable, clocking the SX48 with a 20MHz resonator, or an oscillator?· It seems that Parallax shows resonators used more frequently.

Thoughts?· I like the stability of an oscillator, but I designed the board for a resonator, and now I'm not sure that the 20MHz resonator with the 1M Rf is the correct choice.

Thanks,
··· Laura·

Bean

Laura,
I guess it depends on the accuracy you want.
The resonator is about 0.1%, where a crystal oscillator may be 0.001%.
Other than that, I don't think it matters.

Bean.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
·The next time you need a hero don't look up in the sky...Look in the mirror.


·

Old_Lady

Thanks, Bean.
~ Laura

PJMonty

Laura,

I always use TTL oscillator chips such as this:

www.mouser.com/Search/ProductDetail.aspx?qs=sGAEpiMZZMvCNRO4dXrouUsUg7nty0ZU4VlWklbLDso%3d

It guarantees me a known, quality clock that works without worrying about the the drive level in the SX or any of the other issues that often create problems. However, the tradeoff is that the clock chip itself draws a fair amount of current (about 30 ma for this particular chip), which would be a problem in a circuit that is battery operated or otherwise power constrained.

One heads-up if you go this route: Just connect the clock enable on the chip so it is always enabled, and instead use a jumper to disconnect it from the SX for debugging. These chips typically have diodes across the outputs that prevent the SX-Key from operating correctly for debugging.

Thanks,
PeterM

dkemppai

PJMonty said...
Laura,

I always use TTL oscillator chips such as this:

www.mouser.com/Search/ProductDetail.aspx?qs=sGAEpiMZZMvCNRO4dXrouUsUg7nty0ZU4VlWklbLDso%3d

It guarantees me a known, quality clock that works without worrying about the the drive level in the SX or any of the other issues that often create problems. However, the tradeoff is that the clock chip itself draws a fair amount of current (about 30 ma for this particular chip), which would be a problem in a circuit that is battery operated or otherwise power constrained.

One heads-up if you go this route: Just connect the clock enable on the chip so it is always enabled, and instead use a jumper to disconnect it from the SX for debugging. These chips typically have diodes across the outputs that prevent the SX-Key from operating correctly for debugging.

Thanks,
PeterM

FYI,

I just had a problem on a board, where the SX wouldn't clock with a 3.3V TTL Xtal (SX running at 3.3V). I tried everything.
However, I've always been able to get the SX to clock at 3.3 to 3.0 volts with a resonator (often times needing a few pf cap to ground from osc2).

Checked the osc with a high·impedance fet input probe, and it was working as it should have been. The SX Just wouldn't start up. Had to switch to a 5V Xtal on the 3.3V SX. Even the guys at parallax·didn't know what to say...

-Dan


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔

"A saint-like quantity of patience is a help, if this is unavailable, a salty vocabulary works nearly as well." - A. S. Weaver

Capdiamont

PJMonty: Do these fit directly in the 4 pin sockets of the protoboards? Is the accuracy good enough for serial communication? Thanks!

PJMonty

Capdiamont,

I checked the protoboard schematics, and yes, the "half size" TTL clocks fit directly in the protoboard's sockets. The "Half size" packages are 8 pin DIP size packages that only have 4 leads.

Also, the accuracy is very good and more than sufficient for serial communications.

Thanks,
PeterM

Post Edited (PJMonty) : 2/16/2009 6:45:10 AM GMT

Capdiamont

Thanks! I wanted different clock options, so I can get this protoboard working, with serial comms. I just can't make it work with resonators.