Page **3** of **3**

### Re: Astrometry: Looking for Pascal/Delphi functions

Posted: **17 Jun 2019, 16:09**

by **Asterix**

han.k wrote: ↑16 Jun 2019, 10:06

My program reports an altitude of 5.0. I do not include observatory height nor the pear shape of the Earth. The altitude measurements are for me of less interest.

Thank you, Han, for giving it a try. I guess the conversion to Horizon coordinates should not be a problem and that the reason for your larger difference is somewhere before. Here I try to improve the so far verified accuracy of about 1 arcsec. As mentioned before, once I am happy with my star calculations I will come back regarding implementing JPL DE.

Until then ... best wishes

Marcel

### Re: Astrometry: Looking for Pascal/Delphi functions

Posted: **21 Jun 2019, 09:07**

by **Asterix**

Han, Sorry, I realised only afterwards that your comment: "The altitude measurements are for me of less interest." related probably to the reduced number of decimals for showing the result ALTtrue in the horizon system. With the one decimal shown the result agrees with others but it does unfortunately not allow a comparison to higher accuracy which I would require for verifying my calculations.

Now something else ...

I intend to implement your Pascal unit "hns_Uprs" with the long term precession model which you converted from the original Fortran script. My questions:

1) Have you already verified whether the Pascal results compare well with the Fortran ones?

2) My undersunding is that the prcession model 2006 (on which it is based) combines both, precession and nutation. Is this also true for this long term model, or did it require to multiply the resulting precession matrix additionally also with a nutation matrix?

Thanks and best wishes

Marcel

### Re: Astrometry: Looking for Pascal/Delphi functions

Posted: **21 Jun 2019, 19:13**

by **Asterix**

Asterix wrote: ↑21 Jun 2019, 09:07

2) My undersunding is that the prcession model 2006 (on which it is based) combines both, precession and nutation. Is this also true for this long term model, or did it require to multiply the resulting precession matrix additionally also with a nutation matrix?

It seems my understanding was wrong, i.e. it looks now to me like the nutation is additionally required.

### Re: Astrometry: Looking for Pascal/Delphi functions

Posted: **22 Jun 2019, 13:57**

by **han.k**

The precession model was converted by Patrick Patrick Chevalley. The precession unit is 95 % copy of his unit.

I did a few spot checks and the value matched with the old routine, so I assumed it worked correctly. Ideally you could compare the Fortran output with Pascal version.

The altitude is for the regular user not interesting. He only want to know the rough altitude value . The only relevant relation for him is how higher the object the better the seeing is. I could quickly create a HNSKY version with more altitude decimals if you would like to compare.

It is already some time ago I implemented the routines. After a few years I have to study the topic again. But the HNSKY program does the following:

Code line 3655 in hns_main.pas:

If equinox is not J2000

Standard year 1750 -2250:

- Precession (ESO 2011)

- nutation_aberration_correction_equatorial_classic

Using DE430:

- Precession (ESO 2011)

- nutation_correction_equatorial

- aberration_correction_equatorial

Regards, Han

### Re: Astrometry: Looking for Pascal/Delphi functions

Posted: **24 Jun 2019, 14:37**

by **Asterix**

Thank you, Han, I found it, and yes before adding aberratioon. What Nutation model are you using, may be the JPL DE one? Do you have by any chance also the Nutation IAU 2000A or 2000B models as Pascal source code or know someone else who has it and is prepared sharing it?

### Re: Astrometry: Looking for Pascal/Delphi functions

Posted: **26 Jun 2019, 14:29**

by **Asterix**

Han, it looks like I did not implement "my" nutation correctly. May I ask you for your Pascal source code for

Code: Select all

```
nutation_correction_equatorial(ra_eq, dec_eq);
aberration_correction_equatorial(ra_eq, dec_eq);
```

Or did I miss them when searching through your various source files?

Thanks and regards,

Marcel

Sorry, just found it in hns_Uast.