A different program controls the camera for sampling the fringes. It is
also in the "c:\nic_code directory" and it is presently
The pixels to be sampled are specified in the frnges5.cfg file; and they
are the only parameters you need to change there.
As before, you get out of fringes5 by typing "Shift-S", and then choosing
to continue with different parameters or quitting.
This program reads the two selected pixels and sends the data to the data
acquisition Quadra, which displays it and saves it to disk if so desired.
You must first start the fringes5 program and THEN start the Quadra program
(MacInscope1.1.0); the actual sampling of the nicmos and sending of the
data starts when you turn on the piezo scanner electronics (each
scan is triggered by the piezo sending a TTL signal to the Quadra when
it stars on its 60 micron trip). The display on the Quadra shows the two
raw data outputs (complementary fringes), on a scale 1-65535, when the
gain is set to 1; and the difference between the two fringes normalized
by a dc average, and filtered if you so choose. The sky background is not
subtracted. If the background were a small contribution, a 100% fringe
would be full scale on the plot showing the difference fringe.
For historical reasons (the program running in the Quadra was inherited
from the visible experiment) the nicmos sampling must send exactly
256 data points to the Quadra; which imposes some limitations on
the fringe sampling modes. Most notably, it implies that :
at K the fringes must be oversampled
at approx. 9 samples/fringe
at H the fringes must be oversampled
at approx. 7 samples/fringe
at J the fringes must be oversampled
at approx. 5 samples/fringe
The readout mode is specified by selecting a number of reads per sample
and a number of samples per frame; I call a frame the final average of
(reads per sample x samples per frame) for both pixels.
While the Pentium samples the two pixels in the selected mode, the short
delay line line is following the sidereal fringe position, and the
piezo mirror is passively scanning its 60 microns of max. range at a frequency
that must be selected, and which determines the fringe frequency together
with the wavelength and sampling time.
These are some calibrated modes :
||Set Scanning Freq. to
Fringe Freq. = 1/(Samp. Rate x Samp. Time)
For a given choice, you must actually go and set the piezo scanning freq.
to the appropriate value on the piezo electronics box. You can monitor
its value by looking at the driving waveform on the cope.
The fringe frequency can be determined in each case from the above sample
rates (in samples/fringe) for each spectral band and the above sampling
times as :
You must establish communication between the Short Delay Control Quadra
and the Data Acqusition Quadra so that you get the correct header information
in your files.
To quit the program, the less violent way is to stop the piezo scanner
electronics (stop the TTL signal), then quit the fringes5
program on the Pentium and then quit the MacInscope program on the
Quadra. Then turn off the camera power supply.
Back to Instrument Subsystems Page