Split clock signal (provided by oscillator)

Click For Summary

Discussion Overview

The discussion revolves around the feasibility of splitting a clock signal from an oscillator to synchronize multiple USB devices. Participants explore methods for amplifying the clock signal and the implications of using internal oscillators in the devices.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant inquires about the possibility of removing the internal oscillator from USB devices and using a single oscillator to provide a synchronized clock signal to multiple devices.
  • Another participant suggests using a clock buffer chip to split the clock signal if the digital clock source is external, but notes that some devices may not have accessible output pins for external clocks.
  • A participant emphasizes the need to amplify the clock signal after splitting, questioning whether an operational amplifier in an open-loop configuration would suffice for this purpose.
  • Concerns are raised regarding transmission line effects and the need for a proper digital driver for signals above 1 MHz to ensure clean rise and fall times.
  • Another participant mentions that precision instruments often use sine wave signals as external reference signals and questions the practicality of using a digital clock as an external reference.

Areas of Agreement / Disagreement

Participants express differing views on the methods for splitting and amplifying the clock signal, as well as the practicality of using digital clocks as external references. The discussion remains unresolved with multiple competing perspectives.

Contextual Notes

Participants highlight limitations related to the internal oscillators of the devices and the potential challenges of signal integrity when splitting and amplifying the clock signal.

Tom48
Messages
37
Reaction score
0
hi,

i was wondering how its possible to split a clock signal provided by an oscillator in order to use it to drive multiple devices with this one split clock signal.
Let's assume we have a kind of USB device (like an USB stick), that gets its clock signal from an oscillator.
Let's say i would like to provide the same clock signal to multiple USB devices of the same kind (to synchronize them). Would it be possible to take the already build in oscillator out, split the clock signal of this oscillator, amplify it and then provide this clock signal to all devices so they all have the same clock signal?
How would i amplify the clock signal? Could i just use an operational amplifier with no feedback to amplify the clock signal?

At the moment i have no certain device, I am just curious how this could be done?

Thanks a lot!
 
Engineering news on Phys.org
If the digital clock source (Fosc) is external on the master clocked device then a simple clock buffer chip using that signal will work. Some devices have a internal oscillator with an output pin to drive external devices with a programmed configuration option for a external clock that may not be accessible on a production USB stick.
3.3v device: http://www.ti.com/product/CDCLVC1112
 
nsaspook said:
If the digital clock source (Fosc) is external on the master clocked device then a simple clock buffer chip using that signal will work. Some devices have a internal oscillator with an output pin to drive external devices with a programmed configuration option for a external clock that may not be accessible on a production USB stick.
3.3v device: http://www.ti.com/product/CDCLVC1112
thanks for your reply!
Im indeed talking about devices that all have internal oscillators only.
So i had to remove the internal oscillator from the device first.
After i would clock multiple devices of the same kind with one shared oscillator.
But i guess i first had to amplify the given clock signal somehow in order to clock mutiple devices with it. (because its going to be to weak after the signal split)
Could i just amplify this signal using an omp amp (open loop gain configuration)?

thanks!
 
Tom48 said:
Could i just amplify this signal using an omp amp (open loop gain configuration)?

thanks!

Assuming the clock is a digital signal.
Because of transmission line effects/fanout (current drive capability) and the need for clean,fast rise/fall times on the clock a proper digital driver would be better for any signals greater than maybe 1 MHz.
 
  • Like
Likes   Reactions: Tom48
nsaspook said:
Assuming the clock is a digital signal.
Because of transmission line effects/fanout (current drive capability) and the need for clean,fast rise/fall times on the clock a proper digital driver would be better for any signals greater than maybe 1 MHz.

That depends. Measurements instruments usually use a 10 MHz sine as a external reference signal and will have a "ref in" and "ref out" connector at the back. In precision instruments (which I presume is what we are talking about here, not much point locking several instruments to the same oscillator otherwise) the oscillator is a loop and they can lock to an external reference as long as it is reasonably close to the internal reference.

I don't think I've ever come across a digital clock being used as an external reference (not even in say backplanes of rackmounted equipment); it just seems like an unnecessary complication when you can so easily do clock recovery (and generate a digital clock) from a sinewave
 
  • Like
Likes   Reactions: Tom48 and nsaspook

Similar threads

  • · Replies 7 ·
Replies
7
Views
1K
  • · Replies 21 ·
Replies
21
Views
6K
  • · Replies 40 ·
2
Replies
40
Views
4K
Replies
9
Views
5K
  • · Replies 7 ·
Replies
7
Views
2K
Replies
13
Views
3K
Replies
23
Views
4K
  • · Replies 17 ·
Replies
17
Views
2K
Replies
37
Views
7K
  • · Replies 5 ·
Replies
5
Views
2K