diff --git a/prod/doc/analysis.hmac b/prod/doc/analysis.hmac index 354aa4e..4210fef 100644 --- a/prod/doc/analysis.hmac +++ b/prod/doc/analysis.hmac @@ -289,7 +289,7 @@ depending on how much code was successfully flashed. - + The clock frequency of atben can be measured with an accuracy of about +/− 100 ppm using the program atrf-xtal. atrf-xtal @@ -297,7 +297,7 @@ runs directly on the Ben and measures the duration of packet transmissions. The transmission time depends on the bit clock which is in turn derived from the oscillator.
-atrf-xtal 100
+ben# atrf-xtal 100
 
The number reported is the number of poll loops the CPU counted. This value should be compared to a reference count obtained with a known to @@ -305,13 +305,35 @@ be good atben board on the same Ben at a comparable temperature.

DifferenceAction -
> +/− 50 ppmCorrect operation +
< +/− 50 ppmCorrect operation
< −80 ppmCheck soldering of capacitors; check for contamination around crystal
> +120 ppmidem
OtherDeviation can be compensated by adjusting trim value
+ + + + + + +The clock frequency of atusb can be measured with an accuracy +of about +/− 1 ppm relative to the PC's NTP-disciplined clock. +

+pc# atrf-xtal 10000
+
+This measurement runs for 10−60 seconds, depending on system load. +Note that, in order to reach such a high precision, it is imperative +that the PC clock be disciplined by NTP. +

+ +
DifferenceAction +
< +/− 30 ppmCorrect operation +
OtherSee the atben section above +
+ +