PHD guiding - first use, got some questions

[troypiggo (Troy)]

I'm very new to this. The 3 opportunities I've had so far to set up with clear skies have been spent just setting up the mount, learning a few "important" stars, and practising aligning using the SynScan 1, 2, or 3 star alignments. Haven't taken a single photo yet.

Last night I got to a point where I was pretty happy with the alignment, so thought I've give PHD Guiding a go and try to take my first exposures. This is where some questions come in.

1. Selecting the camera - I have the Philips SPC900NC and GPUSB, and chose the Windows WDM(?) camera (I assume this was correct?), then the camera (choice of 2 here since laptop has a built-in webcam also), then got a list full of resolutions with letters after them (eg 640x480 QYUX or something like that). At this point I had no idea which one to select, so picked one with max resolution. There were several choices all with same resolution. What do the letters mean and which one should I be selecting?

2. I couldn't get the thing to focus. Was just getting a screen full of white noise. Eventually pointed at the moon and focused on that, then pointed back at some stars. Realised the gamma correction slider was half, so slid it all the way up and could finally make out some of the brighter stars only. Was surprised at the amount of noise. Any way to improve this other than what I did? What if I'm shooting something in a region where these isn't a star bright enough to show over all the noise?

3. After selecting a star and starting guiding, all appeared to be going well for a bit. It was tracking ok (I think). But then the star started drifting out of the "box" and PHD Guiding window started beeping and flashing, and I noticed the star was out of the guide crosshairs. How did this happen? Sort of defeats the purpose of autoguiding?

Any answers, tips, help appreciated.

PS - still didn't get to take any exposures

[Dennis]

I think that the Philips webcam has a 640x480 chip, so I’d use the settings for 640x480, which means that the imaging region of the chip is an area 640 pixels wide by 480 pixels high.

An all-white display normally would indicate a gross over exposure; all the pixels are saturated and are registering a value of 255 which is pure white, with 0 being black on an 8-bit scale.

On warm evenings most un-modified webcams will show noise when exposing for longer than say, 1 or 2 seconds. The way around this is to cool the chip. For example, my SBIG ST7 CCD camera can be cooled to around 30 degrees below ambient temperature and for each 5 or 6 degrees temperature drop, you roughly halve the noise.

Unfortunately, with a standard webcam you will see noise, especially on those warm Brisbane nights!

Cheers

Dennis

[peter_4059 (Peter)]

Troy,

I use PHD a bit but am no expert and others will probably do things differently to me but here goes...
1. WDM is correct and a Dennis recommends I also use 640x480. I've tried the different options (RGB24, IYUV and I420) but have not found any difference in the image - someone else might be able to explain the difference and if one is better than the others but I've found them all to work much the same.
2. I run with the gamma all the way to the right. I have a neximage which is similar to the toucam - both short exposure and not particularly sensitive in LUX terms so can also only find relatively bright guide stars with it. To find a guide star in some regions you might need a set of guide rings so you can re-position the guide scope independent of the imaging scope.
3. Not sure why your guide star drifted out of the box - normally it starts carrying on (beeping) when it looses the guide star (cloud etc). Was the GPUSB light changing colour indicating it was receiving guide commands from PHD? Also can you hear the motors in the mount making guide corrections?

I'm also in Brisbane so happy to get together one night and have a look if you want.

Good luck.

Peter

[troypiggo (Troy)]

Thanks Dennis and Peter. I'm giving it another go tonight to see if I can improve things. Might try some drift aligning too.

Peter - would really appreciate a get together some time when the skies are good. I'm at Auchenflower, but will head up to the in-law's at Upper Brookfield for darker skies tonight. Where do you go? I was thinking about heading to one of those Leyburn nights - not January, but maybe the Feb one.

[troypiggo (Troy)]

Been doing some more googling. Discovered that the letter suffixes after the camera resolutions in the camera mode window refer to camera codecs YUY2, IYUV, and I420. The YUY2 codec is newer and so I'll use 640x480 (YUY2).

choose the YUY2 codec, which delivers a higher quality image (and a larger file for the same amount of data).
A 100 frame AVI at 640x480 is about 65Mb with the YUY2 codec, compared to 45Mb with I420 or IYUV.

[Rigel003 (Graeme)]

It sounds like the guide stars you've chosen aren't sufficiently bright for it to lock on to. Set the SPC900NC to its max exposure and use the "Take Dark" feature. PhD prompts you to cover the lens, takes a dark frame, and then subtracts it from each of the displayed video frames, reducing noise in the image. You could also experiment with 2x2 noise reduction (in the "brain" menu). Even so, you will be limited to bright guide stars.

[Octane (Humayun)]

Troy,

This might sound rudimentary, but, I'll ask anyway, did you drift align the mount first? How long did the star stay in the reticle/camera before it started to drift?

Regards,
Humayun

[troypiggo (Troy)]

Originally Posted by Rigel003

It sounds like the guide stars you've chosen aren't sufficiently bright for it to lock on to. Set the SPC900NC to its max exposure and use the "Take Dark" feature. PhD prompts you to cover the lens, takes a dark frame, and then subtracts it from each of the displayed video frames, reducing noise in the image. You could also experiment with 2x2 noise reduction (in the "brain" menu). Even so, you will be limited to bright guide stars.

Thanks Rigel! That's some pretty specific advice that I'll definitely try next time I'm out.

Originally Posted by Octane

This might sound rudimentary, but, I'll ask anyway, did you drift align the mount first? How long did the star stay in the reticle/camera before it started to drift?

I didn't. Have yet to try drift aligning, although I have read a lot about it. It was on my list next time out.

I did a 3 star align, then guided to some other stars and did some guiding correction on those.

[Octane (Humayun)]

Troy,

Originally Posted by troypiggo

I didn't. Have yet to try drift aligning, although I have read a lot about it. It was on my list next time out.

I did a 3 star align, then guided to some other stars and did some guiding correction on those.

Mate, I'd hazard a guess that this was the culprit.

Give drift aligning a go next time and see if it makes a difference. I reckon it will.

This is a dastardly hobby. Keep at it; your patience will be rewarded.

Regards,
Humayun

[troypiggo (Troy)]

Just looked outside and the clouds are clearing out. Might give it a go tonight.

[RobF (Rob)]

All the best Troy. Brissy hasn't been a great place for an Astronomer to be last month or so.....

[kinetic (Steve)]

Originally Posted by peter_4059

Troy,


1. WDM is correct and a Dennis recommends I also use 640x480. I've tried the different options (RGB24, IYUV and I420) but have not found any difference in the image - someone else might be able to explain the difference and if one is better than the others but I've found them all to work much the same.

Peter

Troy,

sorry mate, no idea about PHD but I know a bit about SPC900 webcams.
For guiding purposes on an unmodded cam, you won't find any noticeable
difference between the codecs. But stick with WDM for capture device.

But here's what the codecs mean:
I420 or IYUV uses 12 bits per pixel with 8 bits brightness and the rest for colour.
YUY2 uses 16bits per pixel with 8 bits brightness and the rest for colour.

RGB is uncompressed and uses 24 bits per pixel with 8 bits per R,G,B

Also, an un-modified SPC900 should only be capable of 1/5sec
shutter speed at the best, no longer, definately not seconds long.

This would also, I'd imagine, limit you to only using brighter guide stars
in a field. Unless PHD does some sort of integration and stacks a few frames on the fly .... dunno.

Steve

[troypiggo (Troy)]

Ok, thanks for the clarification