|
Al's Astronomical Timer is a windows application for
recording and timing astronomical events - all you need is a PC or laptop
computer with you when you make your observations.
System Requirements
The application is written in Visual Basic 2005 Express
Edition, so it uses the .NET framework. If you do not have the .NET framework
installed on your PC, the application will try to download the required support
files from Microsoft via the internet when you install the program. If you are
not able to, or do not want to, install the required files or the .NET
framework, it is recommended you don't install the program. Other than that the
system requirements are very light so just about any PC running windows with the
.NET framework (and that has a working mouse!) should do.
Features
- Record event times to a precision of 0.01 second.
Times are displayed to 0.001 second but the accuracy of the milliseconds
depends on processor speed and other processes running at the time because
the program reads the time and milliseconds separately. This error was felt
to be insignificant relative to the observer's PE.
- Elapsed times between events are measured relative to
the first event with potential precision well in excess of a millisecond by
using the system clock tick counter (1/10000000 second per tick).
- Statistical measurement of the observer's Personal
Equation (PE) for valid determination of most likely event times and
measurement errors.
- Two separate PE measurement routines for Predictable
and Unpredictable events.
- The event measurement log and the results of the PE
measurements can be saved to a text file for later editing, calculations and
presentation with a word processor of your choice.
- The Windows border can be turned off (and back on)
when the application is positioned as desired to eliminate it as a source of
light adversely affecting your night vision. This done by double clicking
just inside the windows border.
Determining your Personal Equation (PE)
Both PE measurement routines assess your PE statistically.
A sample of 30 measurements of your PE is collected. The results of the
statistical analysis are:
- Your Mean (Average) Reaction Time. This
is what is classically referred to in astro circle as your PE. This figure
is subtracted from the time measurement to indicate the most likely time
that the observed event actually occurred.
- The Standard Deviation of your reaction time.
This a stepping stone to calculate the Range of your mean reaction time, but
may be useful for calculating the probability of a particular time being the
time of the event based on you measurements. You will need to know a bit
about statistical analysis to do this.
- The Range of your reaction time. The Range
defines the size of the error applicable to your measurement. It is a
measure of precision. The measured time of an event would normally be
written as: (the most likely time of the event) +/- (Range of reaction
time).
- 2 x Range of your reaction time. This figure
is calculated for convenience only. It represents the error applicable to
the elapsed times measured between two events of the same type i.e. 2
predictable events or 2 unpredictable events.
Predictable Events
Predictable Events are events that can be seen approaching
for example, a visible moon or star approaching the visible limb of a planet. It
is an event that can be anticipated. Because of this anticipation, your PE for
this type of event will be different to that for an unpredictable event.
The PEp (Personal Equation predictable) routine starts
when you press the PEp button. The text on the button will change from "PEp" to
start to count down: 4...3...2...1..."PEp". Press the button as soon as the text
on the button changes back to "PEp". The program will measure your reaction time
and start the cycle for the next test. After you reaction time has been measured
30 times the results are statistically analyzed and presented on the display.
Because this type of event can be anticipated, if you respond before the event
actually occurs you can actually have a negative PE, or mean reaction time.
Unpredictable Events
Unpredictable Events on the other hand are events that you
must observe before you can react. You cannot validly anticipate an
unpredictable event. Such events may be predictable by calculation and computer
simulation, but we are not concerned with that here. These events are things
like the reappearance of an occulted star. You can't see the star behind the
moon or planet in order to anticipate the event. Typically, the mean reaction
time for Unpredictable Events is longer than for Predictable Events.
The PEu (Personal Equation unpredictable) routine starts
when you press the PEu button. At a random time between 1 and 3 seconds after
pressing the PEu button, the text on the button will change to "Press Now". This
you cue to press the button again, allowing the program to measure your reaction
time and start the cycle again. Because the event is not predictable, a negative
PE or reaction time is not valid - it will always be positive.
Correcting your Timer Measurements
(The most likely time that an event occurred) = (Time
recorded by timer) - (Mean Reaction Time)
(The error applicable to an event time) = +/- (Range of
Reaction Time)
(The most likely time between events) = ((Elapsed time of
Event 2) - (Mean Reaction Time for Event 2) - ((Elapsed Time of Event 1) - (Mean
Reaction Time for Event 1)). If Event 2 and Event 1 are the same type of event,
the Mean Reaction Times are the same and cancel out.
(The error applicable to the time between events) = +/-
((Range of Reaction Tim for Event 1) + (Range of Reaction Time for Event 2)). If
Events 1 and 2 are the same type, the error simplifies to +/- 2 x (Range of
Reaction Time)
Accuracy of the System Clock
The accuracy of timings collected with Al's Astronomical
Timer depends, of course, on the accuracy of the system clock. The system clock
should be synchronized to an atomic clock server via the web prior to taking
measurements. Again after the observations, the time should be synchronized, and
any discrepancy noted. This time variation needs to be added to the error band
for your measurements.
For example, say we timed an event at 21:39:22.567 +/-
0.234 sec. If the PC clock was synchronized before the measurement, but was say
2 seconds slow afterwards, then we don't know when in between the clock lost
time. So the best way to represent this is to say the time for the event was
21:39:22.567 +2.234 -0.234 secs (remember a slow clock will indicate the time
later than it really is). So if your clock is slow, the discrepancy adds on
to the + error band, and if the clock is fast, it adds on to the minus error
band.
You may be able to test and prove that your system clock
loses or gains time at a specific rate. If you know this to be the case, this
error can be added to or subtracted from the measured or most likely time of the
event. For example, say we know our clock loses 1 second every minute. We note
the time when the clock was synchronized, and again when we take the
measurement. If we took our measurement 7 minutes after the clock was
synchronized, then we need to add 7 seconds to the measured time or the event.
Remember to add for losing time, and subtract for gaining time.
Al's Astronomical Timer records and displays times to
greater precision than most amateur astronomers can hope to achieve in real
time. Don't be fooled by this apparent accuracy. The true precision of your
measurements is indicated by the error bands calculated from your Range of
Reaction times and clock variation errors.
Download Al's Astronomical
Event Timer here - it's free.
Build History:
V1.0.0.0
First Release.
V1.0.0.1
Warning Beep when cursor leaves Event
Timer button. PE data not saved to
file unless the PE procedure has been completed.
Go back to Out of this
World. |