Guiding woes in PHD with DSI-C

[Daniel]

Hi All, had a chance to take my new guide cam i bought second hand out yesterday, it's a DSI-C which i've been told is pretty good for guiding.
The image i'm getting is a massive improvement over what i had before with the sc1 modded camera, but although the image is rock solid and without noise the calibration kept going wrong, the star would fly out of the box during backlash clearing and not return to it, also after it had finished I was getting an error saying "dec calibration failed, turn off declination guiding"

one other thing i've noticed is that I don't seem to get any noticable extra brightness by pushing up the exposure, although even with it set to 0.2s i can still see guide stars most of the time

Any idea's?

Thanks

Daniel
:O)

[JohnP]

Hi Daniel,

I'm not an expert on PHD but below are the settings that I use:



The DEC guide MODE on this snapshot is set to North but that is only because I had established that I only needed to do corrections in that direction. For the best part you can just leave it set to AUTO.

With regards to seeing more stars when you increase the exposure PHD does some auto-scaling & only displays stars that it thinks are suitable for guiding so you don't necessarily start to see a lot more stars when you increase exposure & also those that are there do not necessarily start to get brighter..

Take a look at this FAQ that Fay discovered - it may help you with some of your issues.

http://www.stark-labs.com/wiki/doku.php?id=tutorials:phd:faq


Cheers & HTH - John.

[Mike]

Daniel are you coming to the meeting tonight?

[MarkS]

Daniel,

I may know why the star jumps out of the box during backlash clearing - I had exactly the same problem.   The handset on my mount was already programmed with a certain amount of backlash compensation (in fact too much).  This resulted in the star jumping wildly when PhD tried to move in a certain direction.

My solution was to switch off backlash compensation in the handset.

Mark

[Daniel]

Hi guy's thanks for  the advice, John, I shall try your settings when i have the chance to, aswell as turning off that backlash compensation.

unfortunately I can't make it to the meeting tonight as Im working evenings, realised the meetings are all on thursday nights which is the one of the nights I teach evening classes

Shall try to get along to the next meet that's not on a thursday or tuesday

Thanks

Daniel
:O)

[Mike]

OK - Observing is on Tuesday 6th and Imaging Friday 9th.

[Daniel]

Ahhhh, looks like friday will be my best bet, especially when i still owe you guys 15 quid

[RobertM]

I don't know whether it helps but I had a similar problem with maximdl and ended up setting the agressiveness to 1 (default 10).  Maxim also needs more than 5 pixel movement in the guide star so I had to set the calibration time longer by a factor of 4.

There may also be a problem if you use the autoguide cable rather than ASCOM (or some other S/W).  I had my mount controller set to slew mode rather than guide, and that's precisely what it did.  If your mount has guide settings then you may also have to set the guide speed to the right value x1, x2 or x4 etc

I didn't have any problems with PHD though it worked right out of the box with the ASCOM driver.

[Rocket Pooch]

Daniel,

1st thing just because you can see a star it does not mean it goog for guiding, all guiding software needs a good centroid to the star across several pixels.  If you can imaging the x below being pixels the centre point being the centre of the star.



               x
             x  x
            x      x
   xxxxxx         xxxxxx


Quite often people get a star on screen just i.e. short exposure and the software cannot work out where the centroid is, for example


             x
xxxxxxxxx xxxxxxxxx

Whats then happen is because the x is just about the background noise, other x's then the centre point could be lost causing erratic guiding.

So try to find a bright star to generate a good image and allow the software to do its work, typically I use bright stars or longer exposure or binning to create a good centroid.

In Maxim, AA etc it assesses the centroid, K3CCD and PHD do not so its just experiance from there, good tip would be to start with .5 second or 1 second exposure and work backwards, when I was using a colour guide can I used 1 second even on bright start it worked a treat.


Chris


P.S. go on someon discuss FWHM on the 1st and second images?

[Daniel]

Hi Chris, thanks for that, The stars I was using were quite bright (though not overly bright) and also had very little background noise around them (this DSI is incredible for guide images)

that said, Im still perplexed as to why my stars seem no brighter using a 10s exposure over a 1s exposure (i've been guiding with between 1 and 2)

Another possibility i've been pointed to is to manually take up the slack in the dec before guiding which apparently gets it not to jump like i've been having it do

[Rocket Pooch]

What mount do you have?

[Daniel]

It's a CG5 (currently looking for a cheap EQ6)

[Rocket Pooch]

This may help http://www.astronomyboy.com/cg5/

[Mike]

Ahhhh, looks like friday will be my best bet, especially when i still owe you guys 15 quid

Yes Dan - which we are still waiting for