User Tools
Sidebar
This wiki is not maintained! Do not use this when setting up AuScope experiments!
DBBC work at Yg, 2013 Jul 25
DBBC has had its CPU board replaced for the cooler ATOM processor board from Advantech.
The earthing of the VSI cable connector has been implemented.
Jamie configured and tested the DBBC. He started a test schedule and there have been no clock jumps in 24 hours.
Looking at creating clock jumps by producing electrical disturbances from things such as airconditioners coming on, antenna drives switching on etc.
There have been no clock jumps yet. Correlater evidence suggests that the problem was happening in afternoons till sun down, around 0800 UT. That raises the posibilities of temperature was the problem or the associated electrical disturbances from A/C units cutting in and out. Unfortunately, yesterday was wet at Yg, so we are not certain the problem is fixed.
timemon.log is the file being generated by FS containing the 2 counter readings, maserdelay and clkoff, which you can check or plot to detect jumps.
July 27. Today will be warmer. Randall will come and assist with creating Electrical disturbances. I will be here till Monday, performing Yg maintenance this weekend.
August 7th. Completed first experiment since maintenance today. R1597. There were no clock jumps. It is probably fixed.
Brett Reid
Discussion
Clock Plothttp://users.on.net/~cdadsl/YgClock/postDBBCupg.PDF for the last 2 days during Jamie's test schedule, post DBBC upgrade.
Clock plot http://users.on.net/~cdadsl/YgClock/r1595ygclk.PDF for the last experiment done at the start of this week, prior to the upgrade of DBBC. Clock has jumps.
Clock plot http://users.on.net/~cdadsl/YgClock/r1594ygclk.PDF for the previous experiment which shows the clock as OK.
Clock plot http://users.on.net/~cdadsl/YgClock/r4593ygclk.PDF for the next experiment prior which show that there were 8 or 9 clock jumps.
What has been going on? Why was it OK during r1594? I do not want to say it is fixed yet. Looking for ideas on how to confirm it is fixed. Brett. 27 July 0300 UT.
Spoken to Randall. There was rain during r1594. He suggests that the wet ground improves our immunity to power transients getting in to the DBBC clock. It is wet weather for the last few days and during r1594. If clock jumps return then we need to apply water to our earth stakes as a test.
In the mean time we shall implement the Rich Strand suggestion of earth straps to DBBC and Mark5 in the rack. The rack earth strap connects to the supply authority earth and to the antenna lightning earth mat.
The switching on of antenna drives now never produces clock jumps. This is since the implementation of MEN wiring (multiple earth neutral) where Neutral is tied to earth at each switch board according to AS3000 electrical wiring rules.