[time-nuts] And another Thunderbolt question
time at patoka.org
Tue Aug 8 19:38:41 EDT 2017
I solved an issue connecting TBolt directly to the PCIe Serial Card.
Previously I was using DigiKey PortServer device. Strange that it was
working OK for long period of time.
May be some recent MS update did something for the driver. Will
investigate it later.
Now I could see the sat. information on LH screen. As well there is no
"empty screen" any more ! Thanks everyone for support !
Regarding the date/time - personally I am using "chronyd" with 1PPS
interface via RS232. This setup works perfectly fine for my lab.
210 Number of sources = 5
.- Number of sample points in measurement
/ .- Number of residual runs with same
| / .- Length of measurement set
| | / .- Est. clock freq error
| | | / .- Est. error in
| | | | / .- Est.
| | | | | | On
| | | | | |
| | | | | |
Name/IP Address NP NR Span Frequency Freq Skew Offset
PPS0 6 3 81 -0.000 0.029 -2ns
ntp2.torix.ca 13 9 224m +0.097 0.285 -1578us
ntp3.torix.ca 20 9 327m +0.219 0.142 -1167us
time.allover.co.za 6 5 85m -0.368 2.634 +1947us
18.104.22.168 6 5 102m -0.149 0.735 -3899us
On 2017-08-08 18:14, Chris Albertson wrote:
> On Tue, Aug 8, 2017 at 10:11 AM, paul swed <paulswedb at gmail.com>
>> But the TBolt will not ever on its own be correct.
>> For frequency and 1 pps not an issue. For NTP unfortunate.
>> By the way my TBolt fired up and tracked satellites without any
>> trouble at
> It would be nearly trivial to write a t-bolt specification reference
> clock for NTP. The driver would correct the date the same way LH
> does. Maybe six lines of C code max.
> Chris Albertson
> Redondo Beach, California
More information about the time-nuts