Patriot 12 meter telescope
Last Update: 280may23
Links
Taking data with the
12meter
Documentation
Specs
History:
System performance
Dewar
monitoring wb receiver
hardware
All measurements
cal values
rfi
measurements
rfi movies using pulsar sband
observations
Cooled
wideband receiver
Xband:
Sband:
Monthly plots/movies
Measured power Levels
The telescope model
my notes
fieldsystem, rdbe,
mark6 page
Pulsars
observing with the12 meter
Pictures of the 12 meter system
receiver chain
debugging
the hardware
Daily
temperature plots from 12meter hub,ctrlbox, and
weatherStation (until 220426)
12meter
control program
acceptance test
Trouble shooting polA 210920
- Pol A has jumps , polB doesn't (01jul21)
- polA shows a freq variation of the total power amplitdude
(across the 7 bands). polB does not show this (01jul21)
- When bypassing the lna by driving the system
from a noise source (inserted after the lna)
- there are no jumps in polA or polB
- there is no freq variation of the amplitude for the 7
bands (was in polA).
- --> jumps and amp variation with freq polA is
coming from the lna or before the lna.]
- 220620:found bad connection in bias box.this fed polA B of
s and x rcvs.
Debugging history:
- 220506: tcl script mocktakedata has not implemented mock
startime. So cross, maps started when request sent.
- prior to this you can't assume that the middle of the
strip was the middle of the strip.
- You need to look at header locations: crval5(data
start),rate_dur, and req_pnttime to see if there is an
offset between the application of the rate and the sampled
data.
- 220105: saw first/last band mock band with low power (more info)
- 210909:
- data taken 210908 showed some bands with
instabilities while others looked ok.
- problem was the setfreq.proc tcl routine was selecting
the freq ok, but was getting the hi/lo sideband incorrect.
- occasionally ended up with lo inside the if band.
- 210707: while tracking sun saw total power dropouts
- (10ms duration, 2 sec period).
- occurs more in polB than A.
- occurs in all 7 80Mhz bands the same . (fiber optic
dithering?)
- 210701: monitoring for 17 hours.
- the hub amps are now 11.5 db. added short cables to
input to hub amps, to relieve strain on cables
- PolA has 2 jumps and a few places where it drifted as a
function of freq.
- This looks a little like what polB used to be doing.
On 21may21 we did see polA do things similar to this.
- polB is stable for the entire time.
- found a .073 usec ripple across the spectra
(fitting acf). (about 11 meters if vel=c). (more
info)
- 210629: 20db xb hub amp polB. The power connection to the
amp had corroded, making bad connection.
- replaced both amps with 11.5db again 5db nf
temporary amps.
- 210624: in lab found that the xb hub amps were ok, gain
and noise figure. Problem must be the cables connecting to
the amps in the hub (they are stressed where they connect to
the 2 amps).
- 210623: swapped pol cables at : lna output, hubamp input,
hubampout.
- At each swap do calonoff and measure the cal deflection
relative to cal off (using the square law detectors).
- Found that the large polA,cal deflection difference
(normalizing each to calOff) was being caused by polB
hub amp.
- It must be adding a large value to Tsys.
- 210622: 10am . xband filter in postamp chassis
- moved before the first amp in the postamp chassis to cut
out the 4-8Ghz power in that amp.
- polA,B cal deflection difference remained the same,
- 210324: (calon-caloff)/caloff shows 64 Mhz standing
wave. probably coming from the coupler (see 24mar21,07apr21)
- 210409 - fiber patch cables in pedestal replaced. (more info)
- 210412: fast dump spectra show periodic dropouts in total
power (see 12apr21)
- --> disconnecting the sband input to postamp chassis
got rid of this. sband rfi was saturating the fiber
optics.
- temp variation of gain:
- 30apr21,04may21, 21may21 on sky -1 to -2% gain
variation /deg F change
- 01jun21 using cal diodes -3% gain change/degF
change. Includes cal power variation.
- polB drifting:
- on sky : 30apr21,04may21,21may21
- with cal diode driving if/lo: 01jun21 no drifts
seen
- 01jun21 sitting.. no drifts seen .
- 02jun12 traking sky (but cals as input). no drifts or
jumps, so vibrations in system
aft2023-05-17_08:18:34
:start /share/phildat/wbrcv/230516/test_01_0017.dat
er the lna not a problem.
- --> so pol B drifts in lna or feed.
- PolB going into compression.
- 08jun21: tracked blank to see tsys variation with el
- Had rain clouds, doubled Tsys.
- cal deflection went to 0 polB when tsys doubled.
- 15jun21: calonoff on sun, off sunjun
- on sun double Tsys
- polB caldeflection went to 0 when on sun, ok when off
sun.
History: (top)
- 230727:
cal values measured 20jun23 with 4" shims installed.
- 230714:
cameras at 12meter shut off.. making lots of rfi.
- 230530: updated attn levels for cal measurements.
solAttn in for wb1,wb3,wb5.
- 230529 removed 12db pad from in front of foTx.
- 230525: put 15db pad in straight thru mode instead
of cable. added 12 db pad after coupler and before foTx.
- 230516: wideband cryo dewar cooled down on the
telescope.
- 230111: b5s0g0 (b5) had trouble. replaced mock box
22/12/28. didn't update adc offset till today. 12/28-1/11 b5
had large dc offset.
- 221219: had trouble polB if/lo power supply. fixed
20dec22 10:30. had similar trouble the week before.
- 221110: calPolA,B cables disconnected and then
reconnected to the cal couplers (when checking the
120hz saturation).
- 221108: updated loadramp in pntProg to fix 12meter
hangs.
- 221102: cal fiber box power disconnected 10am
02nov22 thru 10am 04nov22.
- 27oct22: sband
polB shows changing ripple across the bandpass. Has a time
of day dependence.
- 13oct22:
sband polB, 60 hz seen in some rf frequency channels
- 24jun22: sb,xb cals remeasured. attenuators in cal
line switched to xb:(3,2)db , sb:(7,7)db
- 24jun22: sb,xb receiver re installed after qrfh
tests
- 20jun22: bias box. found intermittent short in
current supply. this goes to polA,B s and x.may have caused
some of thedrifing,jumps.
- 07jun22: finished installing sun shades around
12meter base.
- 07jun22: measure cals sb,xb after atten added to cal
lines
- atten switched. xb: wanted (2,3) db-installed (7,7). sb
req:(7,7): installed (2,3)
- 06jun22 : xb has new cal diode., sb diode now
temperature controlled (but we need to add an attn pad)
- 23may22: loose bolts found on backup structure. A
shim a fallen to the ground.
- 25apr22: pm removed ethernet switches from
pedestal and hub (2250 rfi now gone)
- 08apr22: 10am. sband cal installed and active. Both
cals fire together.. (need to fix)
- 04apr22: 15,20db pads (polA,B) before udc now on
a manual switch.
- 11mar22: installed new sband filter. sband signals now
connected to postamp box in pedestal with no pads at input.
- 17jan22: measured xband cal values using sky
and absorber.
- 14jan22: banged the edge of the dish against the
scaffolding when measuring the cal with absorber. Dent in
edge of dish.
- 03jan22: sband cables into rfbox in
pedestal terminated with 50 ohms again. The sband rf was
active 07de21 to 03jan22 am. Sun map taken after 50 ohm pads
installed.
- 07dec21 : sband cables installed. put 15 db pads at
input to if/lo box in the pedestal for the sband cable.
- 09nov21: added pads before udc inputs: polA 15db,
polB 20db.
- Sun was saturating the udc synth chassis.
- 08nov21: cal diode is now temperature stabilized
with a peltier cooler.
- 29oct21: while tracking sun, saw compression in chassis 1
of the udc (more info)
- 210909:setfreq.proc tcl routine was selecting the
freq ok, but was getting the hi/lo sideband incorrect.
- occasionally ended up with lo inside the if band. fixed
210909
- 02sep21: the rate for dut1 should have been ..282668, but
i had entered .0282668 (ms/day).
- entered 12feb21. fixed 02sep21.. ..02sep21 off by aout
50ms or .75 asecs..
- 01jul21: added short cables to 11.5 db
xb hub amps to relieve tension.
- 29jun21: xb polB hub amp. in lab found that
power connection to amp had corroded. This was probably
causing our polB problems.
- replaced both amps with 11.5db gain, 5db nf amps (till
we get a replacement).
- 23jun21: found that hub amp polB was adding lots to Tsys
polB.
- brought xb hub amps down to lab for testing. leave
system connected with a bullet (no 20db amp)
- 22jun21: 10am . xband filter in postamp chassis
- moved before the first amp in the postamp chassis to cut
out the 4-8Ghz power in that amp
- 9apr21: Fiber patch cords (in the pedestal) from the
fiber patch panel to the fiber Tx's changed. They were
causing variation in the power levels, especially polB. more info.
- 07apr21: Found .35 degree offset in the elevation
encoder (relative to the model made in 2016).
- 27apr16:
- moved polA,polB to both be xband at 8600 Mhz.
- switched mock clock from 172.032 Mhz to 160Mhz. vlbi
using time domain.. don't want any glitches.
- 12apr21:
disconnected sband cables to chassis in pedestal. Was
causing glitches in xband signal downstairs
- 31mar16: inject tones into cal coupler to see how large a
cal we need (more
info)
- 28mar16: tx,rx for receivers have paired serial numbers.
- polA 5009, polB 5010.
- We had them reversed so tx 5009 went into rx 5010.
- on 28mar16 we switched the fiber rcvr cables so 5009
tx,rx are together, as well as 5010 tx.rx
- polA now had 5-6 db less power than polB (coming out of
the fiber rcvr.. see specs below)
- 15mar16:
model3A taken feb16-mar16
- 15mar16: updated "
making the model page"
- 10mar16: replaced
attenuator after xb rf filter (to reduce ripples,jumps)
- 16apr15: aorouter no longer part of path to 12m. now
galfas2 -> .224 router -> p12m
- 28jan15: crosses
on casA to verify model2A
- 27jan15: 17:00 installed model2A
- 26jan15: found that grp read addresses were lost on
19jan15 14:00 after power dip. The reboot hadn't stored
them?
- 19jan15: switched to use group read for status info.
- 19jan15: online for model correction was using:
- elRefr=el + refrCor
- azModCor,elModCor= computeMod(az,elRefr)
- azUse = az + azModCor
- elUse = el _ elModCor
- Should have been:
- elUse=el + refrCor + elModCor
- (hasleft off the refractoin correction..
- fixed 19jan15
All Measurements:
- 230726
a4002 3ghz rfi 1-9 jul23
- 230714
testing the cameras at the 12meter with the cooled receiver.
- 230627:
comparing noise and radar pshifts when taking 3GHz pulsar data
- 230627 a4005
sees instability in upper portion of 9103 Mhz band (probably
8730-9000 Mhz rfi)
- 230615
B0329+54 pulsar showed lots of extra rfi in polb. probably 16
bit saturation.
- 230613
shimming the horn to find the wb focus
- 230606
3ghz az swings. el=6,10,20..80 deg
- 230530
25,29 may23.. step cal 2.7 to 12. Ghz. plotting TcalOn/Tsys.
- 230518
updated rcvr chain tsys contributions. added straight thru with
a 15 db pad.
- 230517
azswings with field fox at filter box output on telescope.
- 230518
measured power levels at coupled output before the fiberOptic
transmitter. Straight thru mode is saturating.
- 230516
dewar cooldown on the telescope.
- 230515
sparameters for fiber optic chassis in pedestal, cable loss
dewar to filter box
- 230502
added xband s2p measurements to filter, and filter box.
- 230426
rcvr chain levels. updated 0-14ghz lopass filter in filterbox,
added set using xband cables.
- 230414
s-parameters for filter box as a unit.
- 230413
remeasured cable loss in sband cables lna to pedestal.
- 230412
cooled wide band receiver chain levels.
- 230405
S-parameters for filters in filterbank.
- 230331 Trcv
remeasured after dewar mods.
- 230329
pset error time outs caused by disc io slowdown.
- 230328
re-measured s-parameters wbrcvr cal input to lna output after
cal cables replaced.
- 230328
removing linear baselines before 2d gaussfits on crosses.
- 230328
dewar monitoring box rfi measured in screen room after osvaldo
worked on it.
- 230306
rfi measurements of filter bank box in the screen room.
- 27feb23
summary of 8730 to 9100 MHz rfi.
- 16feb23
solar attenuator increases xband Tsys.
- 10feb23
measured s-parameters wbrcvr cal input to lna output
- 31jan23
remeasured Trcv 4-14 GHz with preamp on.
- 30jan23
xband sidelobes using 3C405 from the jun22 crosses.
- 25jan23
Treceiver for 12m wideband dewar measured at antenna test
building
- 11jan23
first rfi test of dewarmonitor box in the screen room.
- 28nov22
nov22 processing of 60hz data showed jump in freq when cal box
removed, reinstalled.
- 18nov22
testing voltage sampling at sband looking at siriusXm signal.
- 10nov22
more sband calbox tests. reverse cal cables, reverse bias
cables.
- 09nov22
sband polB. check calbox for 120hz saturation.
- 04nov22
sband polB has 120 hz saturation when cal fiber box in pedestal
disconnected.
- 03nov22
sband rms/freqChan for first rec of pulsar sources. added
apr22-jul22
- 02nov22 sband
dynamic spectra of pulsar B1641-45 showed dispersed signal in
polA (probably instrumental)
- 02nov22 tested new
camera on top of 12meter control building.
- 27oct22
Sband polB shows variation in the ripple across the
bandpass.
- 27oct22
added sband tsys movies using pulsar cal files.
- 26oct22
azimuth swing shows no az dependence of 60Hz seen in sband polb.
- 21oct22
sband spectra at input to the postamp chassis (before rf
filter)
- 13oct22
problems with pulsar observing with sb corresponds to 60hz in
polB.
- 06oct22
Fit for Tsys contribution from atmosphere using 20sep22 tipping
scans.
- 02oct22
dispersive signal seen in polA of B1641-45 pulsar observation.
- 28sep22 2d
crosses 25-27jun22 done after sb/xb receiver reinstalled. Fits
to Tsys, gain,sefd.
- 08sep22
using jun22 crosses compare 2dgaussfits to casA, cygA to get the
extended src Correction for casA.
- 31aug22 rfi
test stationary surveillance camera on 12meter control building
roof.
- 30aug22
look at rfi in 30aug22 low elevation azswnings.
- 30aug22
redo azswings at low elevation after brush cleared to see
improvement in Tsys.
- 26aug22
testing from from surveillance camera located at the 12meter
pad.
- 24aug22
testing rfi from surveillance cameras on tower by visitor center
hill.
- 19aug22
test rcvr stability using on,off position switching with winking
cal.
- 19aug22
added azimuth dependence of rfi from the qrfh 2-12 ghz
measurements.
- 12aug22
xband az swings done at low elevation.
- 12aug22
testing the qrfh horn on 12meter. on,off position switch and
cross on the sun
- 06aug22:
look at ratio of cal deflections for each sun map to check
electronic gain stability
- 28jul22
added qrfh spectra measured with no hipass filter for 21jun22
data.
- 27jul22
monitor tsys with hardware winking cal data.
- 21jul22
xband cygnus mapping sees saturation from rfi at 8732 to
8968 MHz.
- 14jul22
testing the QRFH on the 12m meter (jun22).
- 26jun22
xb 4% totPwr dip that lasted for 400 seconds.
- 24jun22
measured sband cal diodes after cal attenuators changed.
- 24jun22
measured xband cal diodes after cal attenuators changed.
- 21jun22
measured power levels with wideband QRFH on the telescope.
- 07jun22
measured xband cal after new diode installed in peltier cooler.
- 07jun22
measured sband cal after diode moved to peltier cooler.
- 27may22
measure sband cal value (before temperature compensation)
- 26may22
links to ben's plots of the processed timing runs.
- 24may22
pulsar dynamic spectra and rms/mean by freq channel movies
- 12may22
xband az swing using azstep routine. 8319-8349 MHz rfi
- 05may22 xband
crosses done on 3C144, 3C461 with cal
- 04apr22
sband system performance from crosses (but no
cal)
- 04apr22
az strip through 3C405 for sband and xband.
- 23mar22
12meter postamp chassis power levels on,off the sun
(sband section, and foTx)
- 22mar22
sband azimuth swings looking at rfi.
- 18mar22
cable loss sband cables lna to postamp chassis.
- 16mar22 15
minutes of winking cal to remove gain variations.
- 11mar22
felix measured s-parameters for sband portion of postamp
chassis.
- 11mar22
measure sband spectra,power levels in postamp chassis after new
sb filter installed.
- 09mar22
S-parameters for new sband filter.
- 02mar22
Correct for rcvr gain variations with winking cal.
- 28feb22
diagnostics from 9,14feb22 recombination line measurements.
- 09feb22
recombination line measurements (includes 14feb22)
- 21jan22
azswing -180 to 360 and back with el=60 showing az motor
torques.
- 17jan22
measured xband cal values using sky and absorber.
- 11jan22
08jan22: B0833-45 pulsar obs. all bands have normal power.
- 11jan22
05jan22: B0329+54, B0833-45 pulsar observation. low power 2
bands for B0833-45
- 11jan22
03jan22: observation of B0329+54
- 11jan22
31dec21: observation B0329+54. rain 1/2 way through obs. 25%
jump, slow return to normal.
- 02jan21
xb psr data (vela,B0329+54) taken on 17dec21.
- 23dec21
cal spectrum and fit using winking cal on blank sky (data taken
07de21)
- 07dec21 spectra,
total power measurements with with sband signal installed in
system.
- 24nov21
check if if/lo, mock spectrometer compress while looking at the
sun.
- 18nov21
make map of sun. Look at strips through the center to if there
is compression.
- 22nov21
measured power levels on,off sun in pedestal and fiber.
- 19nov21
linearity test of the udc.
- 10nov21
calOn,off for 26 hours after temperature controlled cal
installed.
- 01nov21
measure dynamic range in udc while tracking the sun. polB
compresses
- 29oct21 12m
xband. plot ratio sefdPolB/polA for 9sep21 model data.
- 28oct21
xband sefd from model4A crosses.
- 31sep21
xb miteq lna gain measured in lab.
- 24sep21
measure tsys vs elevation at different azimuths
- 19sep21
monitor xb rcvr for 12 hours (lots of rain). specta
results and standing wave.
- 19sep21
monitor xb rcvr for 12 hours (lots of rain). total power
results.
- 23jul21
01jul21 16 hrs of data. Look at spectra and spectral ripples.
- 07jul21:
track sun for 1 hour with 2ms sampling. 2 sec total power
dropouts.
- 06jul21:
m42a recombination lines: 14 5 min on,offs
- 01jul21:
monitor xband for 17 hours. PolA jumps and drifts with freq.
polB stable.
- 24jun21: remove xb
hub amps, then calon,off on,off sun as well as scan across the
sun.
- 17jun21: do
calon,offs while tracking the sun, and then off the sun.
- 11jun21:
measure lna output while looking at blank sky
- 08jun21:
track blank sky for 7 hour firing cal. PolB going into
compression.
- 02jun21:
mock bandpasses used for 12m total power measurement (may/jun).
- 02jun21:
monitor xb rcvr for 4 hours while connected to the cal diode.
Track a position to check vibrations.
- 01jun21:
monitor xb rcvr for 23 hours while connected to the cal noise
diode (no lna or sky).
- 21may21:
monitor xb rcvr for 19 hours.
- 04may21:
monitored xb rcvr for 19 hours.
- 30apr21:
monitor xb rcvr for 14 hours
- 26apr21:
10 Hz sampling for 48 minutes. Tp vs time and tp spectra
(sband terminated)
- 26apr21:
fast sample (.238 millisecs) spectra. look at fiber optic
birdie.
- 26apr21:
measure xb spectra in hub to see where the 3 to 8 Ghz
extra power is coming from.
- 22apr21:
measured rfi : 0 to 1 Ghz using discone antenna at 12meter
site.
- 20apr21:
measure levels input to the 12meter postamp chassis (only
polB)
- 12apr21: dump 1ms
spectra to look at total power dropouts.
- 09apr21:
9 crosses on the Crab taken as xband.
- 07apr21: xb on/off position
switch on casA shows spectra ripple in cal spectra came from
cal.
- 24mar21:
monitoring xband for4 hours. standing wave, polB noisier
- 07may16: fringes 305m, 12m
taken 04may16
- 08apr16: 6 crosses on the
moon to measure Tsys
- 06apr16: tsys vs
elevation for sb, xband.
- 02apr16: sefd from
1d gauss fits to model 3A az,el cross strips.
- 31mar16:Inject tones
into cal couplers. Size of cal we need.. dana's memo.
- 29mar16: fitting
2d gaussians to sband crosses on 3C405
- 13mar16: monitor xb power
for 16 hrs after xb rf atten replaced. pwrjumps and ripple.
- 07mar16: sband
spectra shape variation with time.
- 07mar16: xband total power
ripple, power variation coming from inside the 12meter
pedestal
- 28jan15: crosses
on casA to verify model2A
- 04dec14: comparing
cross positions using Track Array and ramp modes.
- 19jun12:
first light
- 19aug11:dana's
measurements of fiber phase stability. (.pdf) (.eml)
- 28feb11:
rfi measurments from within the pedestal.
- 04feb11: plots of servo
performance tests from acceptance test.
- 04feb11: description of dana's rf
fiber test.
- 26oct10: measure
communications time for controller.
Specs:
|
az
|
el
|
|
|
position limits
|
-180 to +360 deg
|
5.8 to 88 deg
|
|
|
|
sband
|
xband
|
date
|
notes
|
freqRange:
|
2.185 - 2.364 GHz
|
8.05 - 9.2 GHz
|
|
rf filters @12m feb16
|
|
2.208 - 2.343 GHz
|
|
11mar22
|
new sband rf filter installed
|
|
|
|
|
|
bmWdith
fwhm
|
40 Amin
|
10 Amin
|
mar22
|
|
secondary apex to
feedhorn mouth
|
18.125" (46cm)
|
|
|
measured 26may22
|
|
SPD @ fiber
rcvr output
|
-81.5 dbm polA
-76.0 dbm polB
|
-76.4 dbm polA
-70.5 dbm polB
|
160328
|
- 3Mhz rbw, average 100.
- 1db extra pad in xband polB (10mar16)
- switch tx,rx so each pair is same serial #
- 5ft cable rcvr to specAna.
- .5db loss/ft at 8.6ghz
- signal 2.5 db higher at xb at rcver.
- polA has 5 to 6 db less power then polB
|
|
UDC output levels
500-1000 vlbi outputs: dbm/Hz
|
skyCfr (Ghz)
|
polA
|
polB
|
|
|
2.19
2.27
2.364
|
-94.1
-94.7
-97.2
|
-90.4
-88.7
-93.5
|
160331
|
udc atten was set to
15.5 db atten
|
8.1
8.6
9.1
|
-89.5
-89.9
-91.4
|
-85.1
-84.5
-85.3
|
Schematic of system.
Fiber optic output (sband not
terminated)
Measurements with power meter.
-
date
|
PolA
dbm
|
polB
dbm
|
Notes
|
210419
|
-27
|
-31
|
most of the power in 2350 birdie
these levels will saturate the udc (1db compression at
35dbm)
|
211207
|
-46
|
-43
|
15 db pads before input to postamp
chassis in pedestal
15,20 db pads added between FOrcv output and udc input.
measured using pwrmeter at input to udc then subrtraced
pads (assumed cable loss of 0).
|
220315
|
-52.5
|
-47.9
|
new sband filter added 11mar22. pwrmeter.
used short cable from forcvr to udc (but pads 15,20db
pads were not included).
|
spectra fiber optic output
Fiber tx,rx performance
- 220317: Felix measuredthe fiberOptic tx,Rx gain,NF for
2-4 GHz
- do measurements with a short fiber cable (spare tx,rx), and
then measure using installed fiber to 12meter
-
|
shortfiberCable
|
fiber 12mToCtrlRoom
|
gain (db)
|
-16db
|
-21 db
|
Noisefigure (db)
|
24db
|
28 db
|
Equivalent Temp
DegK
|
72555 K
|
182688 K
|
Equivalent noise Temperature at FO
input/output for sband
- The spectral density from the 11mar22
spectra were used to compute temperatures at the input and
output of the FO tx,rx pair (for polB)
- The output temperature of the FO rcv was also computed using
the measured FO TX temp and the Gain and NF that felix
measured for the device.
-
location
Using PolB
|
pwr/rbw
dbm
|
rbw
|
bwDbHz
|
dbm/hz
|
equivalent Temp
DegK
|
notes
|
output diplexor
|
-55dbm
|
300Khz
|
54.77
|
-109.8
|
1.21e9
|
11mar22 spectra
|
input to FO tx
(after 7db pad)
|
-62dbm
|
300Khz |
54.77 |
-116.8
|
152e6
|
just add 7db pad
|
output FO rcvr
|
|
|
|
|
1.2e6
|
computed from fo tx,rx gain,NF
measuremetns
gain -21db
,NF 28db
(152e6 + 182.687e3)*10^(-2.1)
|
-82.3dbm
|
300Khz |
54.77 |
-137.1
|
1.4e6
|
measured with spectrum analyzer
|
|
|
|
|
|
|
|
-
- The measured spectral values agree with the value computed
using the input temperature and the gain,NF of the FO
tx,rx pair
- To get this same value at input to FO tx, we need an input
temp to lna of about 100K.
udc levels
Xband power levels
210426 - power meter in pedestal and cone
- The amp numbers match the names on dana's schematic:
amp2 = A2
- We used the power meter to measure the levels.
- xband : hub,pedestal
- lna ->20ft cable->(X1) amp2 in cone (X2) ->cable to
pedestal->(X3)
PostAmpChassisInp->amp3(X4)->3dbPad->Filter->5dbPad->
-
amp4(X5) ->2dbPad -> diplexor input
- see also (210426 spectrum
analyzer measurements).
- --> 210629: 20db hub amps
replaced with 11.5 db amps (5db noise figure)
-
Measured Xband power levels
location
|
polA
dbm
|
polB
dbm
|
Notes
|
X1 before amp2 in hub
|
|
-37
|
After 20' cable from lna
|
X2 after amp2 in hub
|
|
-15
|
|
X3 input to postamp chassis
|
-26.5
|
-22.57
|
7.5 db loss polB cable
|
X4 after amp 3.
|
|
+2dbm
|
after 1st amp in postamp chassis(before
filter)
|
X5 after amp4
|
|
-8.3dbm
|
amp after filter, before 2db pad
and diplexor input
|
diplexor input
|
|
-10.dbm
|
input to diplexor
|
|
|
|
|
|
|
|
|
- --> using totPwr from spectrum analzyer, the lna to amp2
cable loss is about 10db.
210621: lna output looking at blank
sky
On 11jun21 felix hooked the field fox spectrum
analzyer to the output of the lna's while looking at blank sky:
- 13:07 ast, az=320, el=13.03 deg
- 3 to 11 Ghz, rbw=3Mhz, npnts=1001, vbw=3Mhz, atten=0db, preamp
on , detector=pwr avg, avg 50 sweeps.
- The noise floor of the field fox with this setup is less than
-85dbm across the entire band.
The plots show the measured
spectrum output from the lnas (.ps) (.pdf)
- Black is polA, red is polB
- The purple lines show the region used for the xband reciever.
- Most of the power through the front end is below this band.
The question is where is this power coming from?
- The rectangular waveguide before the lna module has a
long dimension of 1.122 inches (wr112).
- Using the waveguide cut off calculators on the web, this
would have a cutoff frequency of 5.269 GHz.
- I think there are many wavelengths through this piece of
waveguide (the cal coupler), so it should to a good job of
filtering out anything about 1.4% below this (i'm
extrapolating from what our lband waveguide filters did). At
5.19 Ghz it should be down by 40 to 50 db.
- There is a much longer run of circular waveguide (horn to
transition). It's cutoff will be lower than the rectangular
waveguide (it also has a dielectric down its middle).
- previous measurements of lna polB output showed
similar results (be careful,, black, red in avg, peak hold
rather than polA, polB)
processing: x101/210611/12m_lnaoutput.pro
sband power levels
210426 power meter in pedestal
- Sband: pedestal signal path
- lna -> cableToHup ->
cableHupToPedestal->(S3)inputToPostAmpChassis->filter->(S3f)2dbPad->Amp11->3dbPad->Amp12->3dbPad->
-
Amp13(S4)->7dbPad ->diplexor input.(S5)->7dbpad->
FO xmter input
- Note: measurements below do not include the 15db
pads installed 07dec21 in front of the sband rf box inputs
in the pedestal .
-
Measured Sband power levels
Location
|
PolA
dbm
|
PolB
dbm
|
Notes
|
|
|
|
|
S3 input to postamp chassis
|
-29.0
|
-27.6
|
|
S4 output of 3rd amp (before diplexor)
|
|
6dbm
|
|
diplexor input
|
|
-1dbm
|
:q
|
220311 power meter in pedestal after
new sband filter installed
- the 15 db pads at postamp chassis input in pedestal were
removed.
- see 210426 above for signal path naming.
- Telescope pointing at az,el=10,30
- also see sband
spectra 2202311
-
location
|
PolA
dbm
|
polB
dbm
|
Notes
|
output of lna in hub
|
-13.4
|
-15.72
|
this was done later in the day, the el
was down to 7deg. the spectrum at the input to the
postamp chassis saw power from 1700 to 2700 Mhz so this
is wide open.
|
cable loss amp to pedestal
|
12.5db
11.6db
|
12.5db
13.3db
|
measured 18mar22 2.0 to 2.7Ghz using S11
(more info)
using power levels at lna out and postamp chassis input.
These were taken at different times, az,el's so it
is approximate
|
S3 input to postamp chassis
|
-25.0
|
-29.01
|
this is before filter so should be
similar to 210426 measurements
|
S3f after filter
|
|
-70
|
Below minimum power level of power
meter.
|
diplexor input
|
|
-17.5
|
computed using -25.5 dbm
+7dbpad+1db diplexor
|
S5 before FO xmter input
|
|
-25.5
|
The xband input was terminated with
50ohms at the input to the box (so this is only the
sband contribution)
|
|
|
|
|
-
- see also : 220311
sband postamp chassis s-parameters.
xband receiver chain levels
The parameters of the components in the xband
receiver chain were used to compute the levels as you move through
the chain.
- The program sysTemp
(/share/megs/phil/svn/aosoft/src/gen/srcTemp.c) was used for the
computation.
- If you want to play around with the parameters:
- /share/megs/phil/svn/aosoft/p12m/rcvtemp has a script to run
the different configs (rcvtemp.sc)
- You can copy,edit the xxx.dat files to play with things.
- If you want to write the output to a file, then use a
directory you have write permissions to.
- sysTemp should be in your path by default. If not use:/pkg/aosoft/fedora4/x86_64/bin/sysTemp
- For each element the noiseFigure (or temp Kdeg) and the
gain/loss was specified.
- For each step through the chain the program output.
- step name
- noise figure
- device temperature
- device gain
- cumulative gain to that input point in the chain
- Temp (degK) at the input to the device
- Tsys contribution of this element
- The output temp, system temp, and cumulative gain are then
output.
Notes:
- The actual temperature of the system is probably double
the computed 52 K. This could come from
- sky = 10K. There is probably more scattered radiation input
- the feed may have some losses in it (dielectic?), bad
matches/resonances, etc.
- the lna's may not be the quoted 41K, the match to the feed
may not be great.
input file format
The input file (for the original
configuration) was:
# 12meter xband chain
# if possible use values 8-9ghz
#name noiseFigure/Tsys
gain/loss db
#
sky
10K
0db
lna
41K
40db from patriot measurements
cLnaToAmp2
-
-10db 26apr21 spectrum analzyer
measurements 8-9 ghz
hubAmp2
5db
20db 26apr21 measurement 8-9 ghz.
datasheet says 18
cAmp2ToPed
-
-7.5db 26apr21 total power measurements
amp3PostA
7db
24db datasheet
padr3
-
-3db dana drawing
xbfilter
-
-2db dana drawing
padr4
-
-5db dana drawing
amp4PostA
7db
24db dana drawing
padr5
-
-2db data drawing
xbdiplexor
-
-1db dana drawing
padR20
-
-7db dana drawing
Key:
- Cx is a cable
- pad is a resistor
- Amp is an amplifier
- a - in the noise figure col says to use the loss as the noise
figure (pads).
- The #'s come from dana's
drawing.
Xband receiver chain levels
date
|
link
|
state
|
description
|
210621
|
(.txt)
|
|
sys config before xb filter moved
|
210622
|
(.txt) |
|
after xb filter moved before 1st amp in
postamp chassis
|
210623
|
(.txt) |
|
remove hub amp for testing (testing). will
eventually replace them
|
210623
|
(.txt) |
not used
|
remove hub amp, replace 5db pad with 3db pad
(testing)
|
210629
|
(.txt)
|
current
|
polB 20db hub amp bad. temporarily replace
both amps with 11.5 db amps
|
Notes:
- 210622 we moved the xbfilter to before the 1st
amp in the post amp chassis so the amp would not see all the
power from 4 to 8 ghz.
sband receiver chain levels
date
|
link
|
state
|
description
|
220310
|
(.txt)
|
testing
|
test values.
|
|
|
|
|
Documentation: (top)
-
Software:
- Datataking
- Udc
- vlbi software: sched,vex2snap,snap,field
system,rdbe
- acronyms
- dbe-Digital BackEnd
- ddc-Digital Down Conversion
- dbe mode (personality). 4 to 8 filters per subband with
arbitrary freq offsets and bandwidths.
- pfb-Polyphase FilterBank mode
- dbe mode (personality). 16 bands of 32mhz each over the
512 Mhz band.
- rdbe- Roach Digital BackEnd
- digital backend using casper roach board.
- vex-Vlbi EXperiment - describe a
complete description on a vlbi experiment
- independent of any particular data acquisition or
correlator.
- vdif-Vlbi Data Interface Specification
- defines format of recorded data.
- vsi -Vlbi Standard Interface
specification.
- vsi-s software interface
- vsi-h hardware interface
- Files,programs order
- sched < xx.key > xx.vex
- vex2snap < xx.vex > xx.snp, xx.prc
- fieldsystem < xx.snp <
xx.prc > takes data
- > jive5ab (running on mark6 computer) ->
disc
- >Or Cplane/Dplane running on mark6 computer)
-> disc
- sched
- rdbe
- mark5C recorder
- mark6 recorder
- vdif
- vex: Vlbi EXperiment definition)
- vsi: Vlbi Standard Interface
- Field System
-
Telescope:
- AO if/lo
- Motion controller documentation
- The Feed from patriot
Daily temperature plots
Daily plots are made of the temperature in the
12meter hub, 12meter motor control box, and the temperatures
measured from the orion weather station (atop the lidar lab). The
plots are generated at 1am each day via a crontab
running on megs3. A postcript (.ps) and pdf (.pdf) file are
generated for each day.
Directory
listing of daily temperature plots
processing: x101/12meter/temp/plottempdaily.sc
My notes
- xband wave guide wr112 (1.12 inches long dimension)
- te10 mode cutoff freq: 5.263 GHz.:
History extended
210409 - fiber patch cords in pedestal
replaced
When opening and closing the cabinet (holding
postamps,fiber tx) in the pedestal , saw large change in the
levels in the ctrlroom (1/6 Tsys). We found the polB fiber
connector at the patch panel loose. When we tightened it,
the output power in the ctrlroom went way down. When
loosening it, it went back to the original level.
Osvaldo changed both of the fiber patch cables.
After this, opening and closing the cabinet made little
change in the power levels seen at the control room. We
still much larger dropouts in polB relative to polA (although they
both seem to happen together). May be some kind out of band
compression? More notes: x101/210409/chksys.txt
<-
page
up
home_~phil