Frequently Asked Questions
Questions Related to Specific Gage Products
Q.
My wireless communication test application requires a very stable (10 ppm) clock for the CompuScope 14100.
Your specification sheet lists the accuracy of the internal clock to be +/- 200 ppm. Is this the same as stability? I have also seen some of your competitors list their clock stability as 2 ppm. What is going on?
A. This is an excellent question. The explanation given below will most probably answer the question and also uncover some of the tricks some smaller manufacturers in the industry are playing to make themselves look better than they really are.
First of all, let us differentiate between clock accuracy and clock stability .
Clock accuracy refers to the absolute value of the frequency of the clock. CompuScope 14100 uses a 100 MHz SAW oscillator that is 200 ppm accurate, i.e. the frequency output from the oscillator can be in the range of 100.020 MHz to 99.980 MHz (100 MHz +/- 200 ppm).
Clock stability refers to the stability of this output frequency. For example, a CompuScope 14100 whose oscillator is outputting a frequency of 100.020 MHz +/- 0.002 MHz is said to have a frequency stability of +/- 0.002 MHz, or 20 ppm, even though its clock accuracy is rated at 200 ppm.
The physical nature of SAW oscillators makes them highly stable, but a little less accurate. Hence the 200 ppm specification.
Therefore, if your application requires very good clock stability, then the Clock Accuracy is not the right parameter to look at.
If your application requires clock stability better than what the CompuScope 14100 can deliver over the entire operating range, we can suggest two solutions:
a) We install a special oscillator on your CompuScope 14100. This is not economically feasible for small quantity purchases.
b) You use the built-in External Clocking capability of CompuScope 14100 and provide your own digitizing clock from a highly stable frequency synthesizer. This solution is more cost-effective if you are setting up one or two stations only.
Now let us study the competition's claims of 2 ppm clock stability.
When our engineers carefully read the fine print in the specification sheet of one such product made in Europe, they realized that the manufacturer uses a clocking scheme based on a 10 MHz reference clock and a programmable Phase Locked Loop (PLL).
The 2 ppm specification is really for the 10 MHz reference frequency oscillator and not for the sampling clock that the user really cares about!
Let us imagine that the output of the PLL was to be a 100 MHz clock. Simple arithmetic shows that the clock stability of this system will be at least (10 * 2) = 20 ppm - assuming a perfect PLL frequency multiplication.
In fact, the analog noise in the PLL circuit will, inevitably, induce phase noise in the sampling clock frequency, thereby increasing the clock instability.
Measurements done using this "2 ppm" clocking scheme showed clock stability as bad as 350 ppm over the entire operating temperature range.
The moral of this story is that the oscilloscope consumer should always be vigilant against manufacturers who want to "stretch the truth" when it comes to product specifications.
back
print this
email this
