PolA Tsys high

sep15

Intro
History
09sep15: calibration run shows that problem is not the cal values.
28sep15: TsysLoad, antenna show that large Tsys Difference only occurs on antenna
summary

Intro

The 430 ch receiver signal path is:

turnstile to
jennings relay (load,sky coax switch)
polA -> monoplexor
polA,B -> receiver.

History:

The monoplexor should add about 20 K to the system temperature.

Historically we've seen the temperature difference between polA and polB range from 20 to 50 Kelvins .
The plots show:

Top Frame: tsys polA (black), polB (red) for the year
bottom frame: TsysPolA - tsyspolB for the year
2014 tsys 430ch (.ps) (.pdf)

for most of 2014 TsysPolA-PolB was 10 to 20 Kelvins
may14-jun14 it had jumped up to a difference of 40 K
on 28Dec14 Tsys jumped to a difference of about 45 K.

2015 tsys 430ch (.ps) (.pdf)

For most of 2015 tsysPolA-polB was about 45K.

Particular dates are:

28dec14: last 430tsys with polA-B diff <=20K
28dec14:from operators log:

Ast Time:         22:20  
Project :         INFO
Obs Type:         
Comments:         There was a loud sound behind the rack 6. as the sound of a
                  short circuit. The experiment  could not continue because of
                  this. I checked the rack and the chassis 6-19 30 MHz - 260
                  , Converter box had no power. Checked the rack 6 breakers
                  and had tripped. I reset it and resolved the problem.

After this delt Tsys was about 40K
15jan15: coax switch polA (jennings relay) replaced

09Sep15 - calibration run with ch

On 9sep15 x102 calibration runs were done with the ch 430 system.
The plots show the sep15 calibration run results (.ps) (.pdf)

Top: gain (K/Jy) vs za. Black is polA, red is polB
2nd: Tsys vs za
3rd: SystemEquivalentFluxDensity (SEFD) vs za

sefd is Tsys/gain. It is independent of errors in the cal value
Tsys and gain both use the cal value. Their ratio cancels the cal value.

bottom: Ratios polB/polA

Black SEFDA/SEFDB, green: TsysB/TsysA
You can see that the tsys ratio is the same as the sefd ratio (which is independent of the cal value)

These plots show that the tsys difference is a true temperature difference and not a problem with the cal values.
processing: x101/150909/ch430.pro

28sep15: Tsys on load and sky.

Tsys was done on sky and then load around 11:15 on 28sep15... the measured values were:

	TsysPolA	tsysPolB	tsysA-B
onSky	121.1	69.1	52
onLoad	387.0	363.2	23.8
onLoad	386.6	362.7	23.9
onSky	119.1	67.5	51.6
onSky	121.5	71.5	50.0

When we were on load, the polA,B difference was only 20K. When we were on sky, the difference was 50K.
The monoplexor is after the load/sky switch. If it was the problem, we should have had a large temperature difference whether we were on load or antenna. These results point at the problem being the jennings relay load/antenna switch.

Summary

TsysPolA - TsysPolb for the ch 430 has been about 45 Kelvins since 28dec14.
Comparing the ratio of sefdA,B to the ratio of TsysA,B shows that the Tsys difference is independent of the cal value (so the problem is not a problem with the cal values)
Tsys on load,antenna from 28sep15 showed that the large pola,B difference only occurs on antenna. On load, the difference is only 20K.

This shows that the problem is not the monoplexor (since it is present on load and antenna).
The problem is probably the jennings relay.

home_~phil