PHD Guiding Basic Use and Troubleshooting

[IanL]

What a brilliant beginner's guide - thank you so much Ian.

A couple of questions re the pixel scale calculations.  I have a Canon EOS 700d, similar to the one you used as an example here.  Your EOS 500d has 15.5 megapixels (15.1 effective megapixels) and a chip size of 22.3 x 14.9mm.  That's a chip size of 22,300µm x 14,900µm = 332,270,000µm2, divided by 15,500,000 pixels, comes to 21.44µm2 per pixel (or 22µm2 per pixel if you're only counting the effective ones).  Yet you say the chip's pixels are 4.7µm2. 

1/ Your number looks like the square root of mine.  Is that right and, if so, why?

2/ Also if so, how come you are using the effective number of pixels rather than the total number?

Thanks in anticipation.

Ben

Ben,

1. I didn't say the pixels are 4.7µm², I actually said the sensor has 4.7µm square pixels, meaning the pixels are square (not rectangular as some older cameras have) and 4.7µm on a side. We are using linear measurements to determine the pixel scale, not areas, so if you've calculated the area of one pixel you need to take the square root to determine the length of one side. Sorry if the bad choice of words caused any confusion, but too late to go back and edit the post now.

2. I've used the effective pixel count since that is readily available to a beginner from images.  It is definitely more accurate to use the total pixel count and sensor area if you have that data available, but at the time I wrote the post I didn't.  Where I can get a data sheet for a sensor I will use it as the source for information in my imaging toolbox, but often one has to rely on incomplete information from supplier marketing materials and they may quote total or effective pixels (and indeed in some sensors you can have slightly fewer or more effective pixels depending on how you configure the electronics).  It's pretty immaterial in this context, the difference between using 4.7 (actually 4.69)µm and 4.63µm in my example results in a pixel scale of 1.88 arc seconds per pixel vs 1.87 arc seconds per pixel, a difference of 0.0243 arc seconds per pixel if you want the gory details without rounding up/down.  The difference matters more at longer focal lengths.

If you want to get really down in to the weeds, you need to know the number of active rows and columns on the sensor, not just the difference in the total vs. active megapixels.  The inactive pixels will not be an equal number of rows/columns in most cases, and also there are more pixels in a row that a column on a typical rectangular sensor.

Edited September 9, 2015 by IanL

[bendavidson]

Thanks - yes, marginal difference in the end, but good to understand how you made the numbers work. With a 9.25" SCT and 60mm Altair guidescope and cam, I end up with a ratio of 8. I'll see how well it does and maybe add a Barlow lens. Adding a 6.3 focal reducer takes the ratio down to about 6...

Once again thanks - really good article. The setting up walk through was excellent and I'm sure you're right it made more sense to do it indoors rather than outside in the cold, even if there weren't hitches...

[alcol620]

Having used a ZWOASI camera to guide using PHD I am soon to take delivery of a Starlight Xpress filter wheel with a SX off axis guider attached. Will PHD run this guider?

Thanks

[Freddie]

See reply to your same question in software. Hope it helps.

[alcol620]

Thanks Freddie

[mhard26339]

I have just found this thread and thank you for a comprehensive tutorial on PHD. I now know what I was doing wrong but more importantly UNDERSTAND what I was doing wrong; saved in my favourites as a reference work!. I now cannot wait for a clear night to try it out. Thanks again.

Regards

Mike

Edited January 6, 2016 by mhard26339

[gonzostar]

Thanks for this great thread. Wish i read it when i first started, However it you have given dome valuble poimters

 

Cheers

 

Dean

[wornish]

Just found this thread and it is exactly what I needed after spending 3 frustrating nights trying to get PHD2 guiding to work properly.

Thanks to IanL for providing all the info, read once and will read again to try and get it to stick.

 

I am using the FLO guiding package  (ST80 + Toupcam)  on my new mount and have only managed once to get guiding to work in 4 attempts.

I seem to get a lot of   "the camera has not responded over 15 seconds"  messages and the whole thing just hangs up.  Tried extending the wait time for response but still get this issue.

I think it most probably was high clouds causing the problem last night, but it seems a strange message when the camera is still actually connected .

I'm running the latest version of PHD2.6.2 running on Windows.  I get a sinking feeling the drivers are buggy under Windows 10.

Anyone using the FLO kit successfully?

 

[Xplode]

The message can not be caused by clouds, it's a problem with the communication to the camera so it sounds like a USB problem, could you tell more about how the guidecam is connected? Cable length? Hub?

Edited September 8, 2016 by Xplode

[wornish]

On 8 September 2016 at 13:54, Xplode said:

The message can not be caused by clouds, it's a problem with the communication to the camera so it sounds like a USB problem, could you tell more about how the guidecam is connected? Cable length? Hub?

Thanks for getting back if you were you referring to my post.  

I seem to have solved the camera message problem,  but last night (first clear night) when I set up again  I couldn't get PHD2 to start tracking because it couldn't connect to the mount this time.

My setup is :

HP Laptop running latest (Aug 2016) Anniversary version of Windows 10.  Also got current ASCOM drivers Version 6.2 

Guide camera is Toupcam GCMOS01200KMB connected by (2 Mtr) USB cable direct from Toupcam into USB port on laptop

plus an ST4 cable (1 Mtr) from Toupcam ST4 port direct to Auto Guide port on AZ-EQ6 GT mount.

Guide software PHD2.6.2  (latest version) 

PHD2 settings were :- 

     Camera :- Ascom Toupcam Driver  (Could see stars although they were very feint because of the bright moon)

     Mount :- Toupcam ST4(ASCOM)

PHD2 said camera connected but kept getting new message saying can't connect to mount because of ASCOM problem, gave up last night and went to bed !!

 

When I tried it again today in the daylight everything connected OK as long as I start Ascom tools first to get Ascom running in the background.

I plan on trying again next clear night.   This tracking is driving me crazy.   

Am I doing something wrong in my settings or is it just bad luck ?

 

 

 

 

[alcol620]

Suggest you get Ascom EQmod running first.

[Eri]

@wornish

I'm using the exact same guiding kit from FLO, it was working fine for a few nights then I took 2 days off to get 2 full nights imaging in and the first night it refused to track or guide at all, no idea why and gave up after 2 hours. next day i uninstalled and reinstall all software and that night it guided and tracked fine.

[Haggish]

Hi guys,
My name is Geir and I'm from Norway I'm new to this forum.
My observatory equipment are: Telescope meade 14 " lx200R OTA @ f:6.2. Mount EQ8 Camera eos450d without IR filter, but with LPS filter.
Guiding: Telescope: Skywatcher 400mm refractor Camera DSI II.
I use phd2 to guiding and EQmod for telescope control.
So my problem. Here one evening I took 20 exposures a 5min of NGC7331 with guiding within +/- 1 "and the stars were round and fine throughout the image, but when I check the frames I see that all the stars move significantly towards the west in the image.
How can this be possible?

[Xplode]

33 minutes ago, Haggish said:

Hi guys,
My name is Geir and I'm from Norway I'm new to this forum.
My observatory equipment are: Telescope meade 14 " lx200R OTA @ f:6.2. Mount EQ8 Camera eos450d without IR filter, but with LPS filter.
Guiding: Telescope: Skywatcher 400mm refractor Camera DSI II.
I use phd2 to guiding and EQmod for telescope control.
So my problem. Here one evening I took 20 exposures a 5min of NGC7331 with guiding within +/- 1 "and the stars were round and fine throughout the image, but when I check the frames I see that all the stars move significantly towards the west in the image.
How can this be possible?

This sounds like a classic problem with flex.

I guess you'r guidescope is an ST80, these are known to have a lot of flex in them like they come from the facory, there's also some mirror flop in your LX200R scope and both these problems cause streaking stars.

I'll also mention that your setup is not ideal, your DSLR is not suited to the scope at all since you are way oversampled at 0.49 arcsec/pixel, this with an older noisy camera is a setup that is hard to get good results out of without a lot of trouble.

[Haggish]

1 hour ago, Xplode said:

This sounds like a classic problem with flex.

I guess you'r guidescope is an ST80, these are known to have a lot of flex in them like they come from the facory, there's also some mirror flop in your LX200R scope and both these problems cause streaking stars.

I'll also mention that your setup is not ideal, your DSLR is not suited to the scope at all since you are way oversampled at 0.49 arcsec/pixel, this with an older noisy camera is a setup that is hard to get good results out of without a lot of trouble.

Hi Ole,

Thanks for replying

If you read my post again you will see that my stars is " were round and fine throughout the image ". so I dont thing flex is the problem nor mirror flop (the LX200r has mirrorlock)

That's why this is a great mystery too me

due to the eos 450d is one of the DSRL that is the best due to noise

see link. http://dslrmodifications.com/DSLRcomparison.html

-Geir

PS Ole Bare for og presentere meg på Norsk, Jeg er den Geir som solgte dere 16" som dere har i observatoriet deres på Hågar ;-

[Xplode]

If i understood you correctly stars move from frame to frame between images? Might not be enough to show up in a single frame , but that definitely sounds like flex.

The mirror lock helps, but isn't perfect.

Those tests are old and it's only a measure of dark current. It does not reflect real image quality, 550D and newer Canon cameras all have lower noise and better image quality in the actual images that come out of the cameras.

[Haggish]

Hi,

yes you understood correctly, but how can it move in the 5 sec between frames and not in the 5 min during exposures !!!

-Geir

[Xplode]

It moves all the time, but when the flex is small enough seeing(fuzzy stars) will mask the movement in a single frame 

[Haggish]

Ok.

I'll check when I'm home but I think the movement is larger then half a star and I can't see trails even at the dim stars

-Geir

[Owmuchonomy]

I suggest you post examples of your light frames.  In that way we can ensure that we are not missing something in your description.  At the moment it is too much of a puzzle and the images will tell the story hopefully.

[Haggish]

Hi guys,

Ole got my lightframes and found that the stars were only slightly oval, but enough to make the stars moved thru the  27 frames and he made this.

Thanks.

Edited November 7, 2016 by Haggish

[iapa]

Regards calibration.

This needs to be done +/-20 off equator.

But, everyone then says to recalibrate at target.

OK - unless the target is about c60degrees, then PHD alerts that it will not be particularly accurate?

What do I do ? 

[alcol620]

Suggest you calibrate in the area you are going to image in. If you move to another area of the sky and guiding is Ok - carry on. If not recalibrate. It only takes a few minutes.

[alcol620]

Having been using PHD guiding for well over 12 months, believe it or not, I have only just noticed something odd about the setting (probably down to me, but I would like to know what is going on). I have always had the Global setting in the Brain which has "Dither in RA only" ticked. However, when I look at the graph when PHD is guiding and dithering occurs, the spike is always to the DEC trace. 

Anyone have any idea why this is happening?

[wornish]

Just re visited this thread.  In the first post it is said that long focal length scope imaging  is beyond the capabilities of most non premium grade mounts. i.e. mounts costing less than £2000.

I am successfully guiding and capturing acceptable images using my existing kit Esprit 100ED ( 550mm)  so no problems there.

Now the but:   I "was" thinking about getting a longer focal length OTA say a Celestron HD 9.25 to capture the smaller DSO's

Will I be dissappointed because my AZ-EQ6 isn't able to guide to the accuracy needed.

What mounts are others using to support a long focal length scope for long exposure  imaging.