Yes. Pointing accuracy still matters in interferometry, even if the timing between the two receivers is perfect.
The important distinction is:
TIMING
Determines whether the signals from the two dishes can be combined coherently.
POINTING
Determines how strongly each dish receives the astronomical source.
Ideally:
SKY SOURCE
*
/ \
/ \
/ \
/ \
/ \
DISH A DISH B
\ /
\ /
\ /
RECEIVERS
|
CORRELATOR
SKY SOURCE
*
/ \
/ \
/ \
/ \
/ \
DISH A DISH B
\ /
\ /
\ /
RECEIVERS
|
CORRELATOR
Both dishes should have the source reasonably close to the centre of their beams:
DISH A DISH B
* *
/ | \ / | \
/ | \ / | \
/ | \ / | \
| |
V V
DISH A DISH B
If one dish is mispointed:
DISH A DISH B
* *
/ | \ \
/ | \ \
/ | \ \
| \ beam
V V
DISH A DISH B
Dish B still receives the source, but at lower gain.
The correlated interferometer signal is approximately:
Signal proportional to:
__________________
/
/ G1 x G2
V
where G1 and G2 are the gains of the two dishes in the direction of the source.
Therefore:
Perfect timing
|
+---- does NOT compensate for poor pointing.
However, there is an important bit of good news.
The dishes do NOT have to be pointed as accurately as the angular resolution or fringe spacing of the interferometer.
There are two different angular scales:
INDIVIDUAL DISH
<------ broad beam ------>
\ /
\ /
\ /
\ /
*
|
DISH
INTERFEROMETER
/\/\/\/\/\/\/\/\/\/\/\/\
narrow fringe pattern
The individual dish diameter determines the broad primary beam.
The separation between the dishes determines the much narrower interferometer fringe pattern.
For example, at the hydrogen line:
Frequency = 1420 MHz
Wavelength = approximately 0.211 m
For a 1.5 m dish:
Beamwidth approximately:
1.2 x wavelength
----------------
dish diameter
1.2 x 0.211
= -----------
1.5
= 0.169 radians
= approximately 9.7 degrees
So a 1.5 m dish at 1420 MHz has a surprisingly broad beam of roughly:
about 10 degrees
<---------------->
\ /
\ /
\ /
\/
*
|
DISH
Consequently, if your two dishes differ in pointing by only:
0.1 degree Excellent
0.5 degree Very good
1 degree Probably little problem
2 degrees Still likely quite usable
several deg Increasing loss of correlated signal
Now compare this with the interferometer baseline.
Suppose:
Dish separation = 10 metres
Wavelength = 0.211 metres
Then the approximate interferometric angular scale is:
wavelength
----------
baseline
0.211
= -----
10
= 0.0211 radians
= approximately 1.2 degrees
So you could have:
Individual dish beam:
|<----------- about 10 degrees ----------->|
Interferometer fringes:
| /\ /\ /\ /\ /\ /\ /\ /\ |
|/ \/ \/ \/ \/ \/ \/ \/ \|
about 1.2 degrees/fringe
This is the crucial point:
DISH POINTING ACCURACY
is NOT
INTERFEROMETER ANGULAR RESOLUTION
The dishes simply need sufficient overlap of their primary beams:
Dish A beam
\ /
\ /
\ /
\ /
\ /
\ /
\ /
X <-- useful common sky area
/ \
/ \
/ \
/ \
/ \
/ \
/ \
Dish B beam
For your 1420 MHz interferometer, I would therefore aim to get the dishes pointing in the same direction to within roughly 1 degree or better. There is no need to achieve arcminute pointing accuracy with dishes of this size.
And perfect clocks solve a different problem:
GOOD POINTING + BAD TIMING
|
V
Strong signals but poor correlation
BAD POINTING + PERFECT TIMING
|
V
Perfectly synchronised weak signals
GOOD POINTING + GOOD TIMING
|
V
BEST RESULT
So for your ezCon interferometer, getting both dishes reasonably well aligned is important, but you don’t need extraordinarily precise mechanical pointing.