Polar aligned - now PHD only has issue in Dec?

[g__day]

So I spend some time this evening perfecting my polar alignment. I saw no drift in my reticle after an hour - so I'd say its a fair alignment - that source of error removed.

But I still found PHD giving me double stars. Turning the zoom up high of Cartes Du Ciel (CDC) I could see the OTA move slightly back and forth every 5, 10 or 15 seconds.

I suspected it was only the DEC, so I did 3 shots of M17 at 150 secs

1. Guided 3 secs - double stars
2. Unguided - stars slightly elongated (good polar alignment - PEC training on the agenda soon)
3. Guided 3 secs, but only in RA not DEC - very round stars

I ratcheted it up to 300 secs - same again.

So I've narrowed down my tracking / auto-guiding challenges as being in my DEC. Even with the pulse length turned down, jump still happen when I enable DEC corrections.

What are folks thoughts?

[Doug]

Some things to consider:
1. Balance of the OTA about the Dec axis.
2. Dec backlash compensation settings in PHD incorrect.
3. Alignment of Autoguider axis to RA/Dec axis not the best.
4. Agressiveness of Dec corrections needs to be right; too little is just as bad or worse than too much, especiallly if Dec backlash is not properly compensated for.

HTH,
Doug

[Dennis]

Great analysis and questions Matthew, with an equally great set of answers from Doug - thanks to both of you as this information may also help me track down some guiding issues I had at Duckadang.

Cheers

Dennis

[g__day]

Doug, my thoughts:

1. I'd say at present DEC balance is close to perfect and the Atlux absolutely eats the load its under at the moment
2. Maybe - need to analyse - but manual slew commands reveal no jerkiness or slop in either RA or DEC - so althought not perfect, there is a deeper hidden cause
3. Almost spot on alignment to two axes judging by calibration - by the time its moves 30 paces on Dec - its RA movement is under 1 - so its within 3 degrees of perfect - better than I ever managed with the CG5. Its either out on both RA and DEC or neither - so if RA is fine and DEC is off - its a deeper, hidden problem again.
4. You can set RA aggressiveness but not DEC- so maybe...

Now real promise of a solution has come from the stark-labs Yahoo forums on seeing vs pixel size and minimum pixel movement (0.25 pixel) before a correction is set. A member has stated with average seeing on a LPI with ED80 - you want it to be 0.7 - 1.1 pixel movement else you'll be chasing stars and yoyo-ing all evening (why only on DEC - not sure).

To me that sounds like a big step forward - so weather permitting I'll try it tonight.

alah:

Re: PHD setup - removing - It's just a jump to the left, then a jump to the right

You might want to try increasing your Min motion (pixels) value. I had a major improvement the other night by increasing this number. I guide with an LPI on an Orion ED80. This combo yields 2.75 arcsec/pixel.

Assuming seeing is on the order of 2-3 arcsec means that I should set my min motion between 0.73 and 1.1 otherwise I'll be chasing the seeing all night. I set my value to 0.75 and the yo-yoing and doubled stars disappeared. Still need to fine tune it a bit but I think I'm on the right track. Determine your guiding resolution and use (seeing/guiding resolution) to figure out what min. motion value to use. Now if I can figure out how to accurately determine what the seeing actually is it would be even better.

Rich M.

http://tech.groups.yahoo.com/group/stark-labs-astronomy-software/message/4339

[g__day]

Yah,

Downloaded V1.52, set min pixels around 0.75 - 0.95 - about 0.9 was best - now 600 second shots look great!

Thanks all!