[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Remote VCO test/PID servo revistited
Nick Beser wrote:
> The remote vco test is going good. I started the linux software with all
> output being piped to an output file. That way I can use grep and the
> unix tail function to confirm that the commands being issued with cclient
> are working.
> Your new version is printing out the measured time interval. I should be
> able to generate a table of VCO and measured intervals. This will only
> be accurate for the APL telescope. Arne's technical note suggests the
> interval should be .9142 seconds for the Dec where the telescope is
> pointing. It occured to me that if you can measure the time interval,
> you should be able to create a command that adjusts the vco setting for
> an exact time interval.
Your table will be good for the APL telescope as long as the
temperature in the computer case stays the same. So for a given
VCO setting the scan rate is a function of the office air temp.
This is if you keep your computer powered up 24x7 as I do so
you don't have a warm up period to wait through.
What I want to do is have the user enter his exact image scale and
aim point and from that compute a line scan interval. This information
is already required to be in the config file. Arne's
TN assumes a 135mm lens. I think the lenses are all off by a small
amount, no two the same. After we do the astrometric reduction we
will know the exact image scale and aim point. I think the formula
for computing the scan interval is in the TN. So this part should be
The next part is harder. You have to continouly look at the scan
line rate and adjust the VCO to keep the scan rate on target. This
is not importent in an office environment but my computer is kept
in an unheated outbuilding where is can get down to ~50F at night.
The hard question is when there is an error in the scan rate how
much do you adjust the VCO? Do it to much and you get ocilation,
to little and you are unresponsive. Your chart of VCO v. scan rate
will be a big help. I'll assume all Mk III systems have a similar
slope on that chart. The slope should tell me how many VCO steps
are required to correct the error in scan rate. I'd apply say,
75% of the required corection.
Tom had some BASIC code that did this. Tom, is it still around?
email@example.com Voice: 626-351-0089 X127
Logicon RDA, Pasadena California Fax: 626-351-0699