Maxim DL script

[scagman]

Hello,

I am after a script that will help me with my CCD drift alignment.

Currently I start taking a 35sec image, wait 5 sec's then press and hold the east arrow for 15sec's then press and hold the west arrow for the remaining 15secs. Inspect the image, which will show the stars making "V". I then adjust the mount, either the alt or az. depending on were I'm pointing. Then start another image wait 5sec then move the mount east for 15sec then back west again for 15secs. I do this till I get the stars forming a single line instead of a "V". I then double the duration of the exposure, and go thru the above procedure again untill I get a single line, double the exp. lenght and go again. I keep doing this using upto 10-15min exp and it gets very boring sitting there holding the buttons on the controller. Not to mention the sore fingers.

Does anyone know of a script that will let me enter a duration, wait 5secs then move the mount east for 1/2 the remaining time then move the mount back west for the rest of the duration.

Eg. put in a time duration of say 605sec. start script, script would wait 5 secs then move the mount east for 300sec at GUIDE rate then move the mount west at the GUIDE rate for the remaining 300sec.

It doesnt have to be anything fancy just a simple script that will save my fingers:) and probable the buttons on the controller.

The mount is a G11 with G2.

I have tried to do this my self but have no idea how to.

Thanks

[alistairsam]

Hi John,

This is the script I use for the same task with EQASCOM.
http://eq-mod.sourceforge.net/samples/CCDDriftAlign.vbs
All it does is it slews east for 30secs and then west for 30s.
So I have my image capture program running windowed, and then run this script and start the capture at the same time.

so if you are able to substitute the CreateObject("EQMOD.Telescope") with your gemini equivalent, this might work.

Maybe try "DriverHelper.Chooser"

I'm guessing you'd have to use the ASCOM driver for your G11.


Also look at the Maxim scripts. There might be something there that uses scope control.
for eg. I found this snippet from the recenter script at http://www.cyanogen.com/maxim_extras.php

see if you could use this with the eqascom script I mentioned.

' #2 -- Telescope driver

' Need to set this to the appropriate telescope model
' This will pop up a chooser box; you pick the telescope
' and set up the COM port etc. You can experiment with
' the Simulator
Set ch = CreateObject("DriverHelper.Chooser")
ScopeProgID = ch.Choose("")
Set t = CreateObject(ScopeProgID)
t.Connected = True
Set ch = Nothing


Else you may be able to script the whole thing in Maxim which can include the capture as well.


Finally, I think this might work partially.
try out this script. It opens a dialog to choose the scope. then you can set it to track. but you need to find the ascom command to change the tracking=true to a slew at 1x sidereal.

http://eq-mod.sourceforge.net/samples/ASCOM_DEMO.zip


I'm no software expert so I'm only guessing here.

Cheers
Alistair

[alistairsam]

http://ascom-standards.org/Developer/Chooser.htm

[alistairsam]

found this interesting as well. replace the "ScopeSim.Telescope" with yours.

look at the script here http://ascom-standards.org/Developer/AppStart.htm

[Tandum]

Have you guys tried using PHD to align? Just connect PHD to your scope and camera and do a normal PHD calibration. Once calibrated, stop everything, go into the brain button and turn off guiding. Turn the scope east or west about 25-35degrees up, start capturing, pick a star and start guiding. Get the graph up and watch the dec chart, ignore RA. Adjust alt for a flatish line. Stop guiding, turn the scope up, get a star overhead and guide again. Adjust Az for a flat line. Rinse and repeat.

Doesn't work well at long focal lengths.

[alistairsam]

Hi Robin,

yep, we've been through quite a few methods including phd guiding and alignmaster.
we've found this method to be the most accurate, correct me if I'm wrong John.
its actually quite easy and removes the ambiguity you may have with the phd graph and whether its going up or down cause sometimes, it looks like its stable, and then it changes for a bit and then stabilizes.
with this ccd method, its very straightforward and either a straight line or a V.
as explained here http://www.observatory.digital-sf.com/Polar_Alignment_CCDv1-1.pdf

It should be possible to move the G11 east and then west with a script.

[Tandum]

I've never used that method alistair. I'll try it out when the eq8 arrives, ummm, next week? :rofl:

Also, if you switch to a dx/dy chart as mentioned on several sites it won't work right as those charts don't compensate for camera angle.

[scagman]

Hi Robin,

I have tried using PHD but found this to work better for me. I always had problems. I do think it was me more than then method. It may be due to me using dx/dy aslo. I will have another go. I do think it is better in some ways as you do get an instant readout on any adjustments made.

I suppose too, if I increase the rate at which I move the mount and decrease the duration I should get the same result in a shorter time.

Cheers

[Tandum]

John, the phd method bums out if your guiders F/L is over maybe 800mm. Changes you make zoom off the chart and you can't tell which way it went. In the dome I normally start with under 600mm then push F/L up to the highest I use to get it accurate. It can sometimes take 3 nights though to get it set, depends on seeing at the longer F/L. It's worth the time in the long run and I only do it if I've pulled the mount out.

If you use dx/dy charts that's straight from the camera with no adjustment for angle. We spoke to stark about it and you really need to use the normal charts and turn all guiding off and ignore RA.

[alistairsam]

Hi Robin,

The problem for me with the phd graph was that it could vary a fair bit due to seeing and you'll have to make an approximation of whether it is drifting or not.
I've seen the graph stay steady for a minute or so and then start drifting.
I've tried this a few times and it didn't work too well for me. and I wasn't using dx/dy. I was just watching dec drift.
with the ccd method, I use 2x2 bin as that makes the trails brighter and keep at 30 to 1min exposures. that takes 2 or 3 tries. when that seems ok, then I go for 2 min at 1x1bin, either 1 or 2 iterations, and then I do one 4 min just to make sure. but by then, the drift in Dec is not noticeable.
basically that's what you're seeing with this method, how far the star has drifted in dec, just that its easier to make out visually as you can see it as a point and the line. you're slewing east/west and coming back to the same point. so since you've made no adjustments to dec, the start and end points should be very close.

As I mentioned, the EQASCOM script works brilliantly for me. I just make multiple copies and edit the values for each. so I have one script on the desktop for each.

John, I'm pretty sure you can script this with maxim. I'll have a go when I get the chance.
Which driver do you use and does the ASCOM one work for you?

[Tandum]

Al, new is good. I'm looking at the ole interfaces on my machine now to see how to control it all via maxim.

[edit]
you can't. Maxim only exposes a camera and the document (the image), not the scope. So you'd need to ascom to the scope and ole/com to maxim. I'll have a look at it.

[scagman]

Yes I connect via Ascom and use gemini.net



Cool, Thanks.

[Meru]

Interesting discussion, thanks for the info guys. I just got my EQMOD cable so looking forward to trying all this out, rather than having to press E/W on the handcontroller for minutes!

[Tandum]

I manged to knock up a rough framework for a tool today. But I can't see how to make the scope move or at least it doesn't appear to make the simulator move. I'm trying to use this command in C# : scope.MoveAxis(DeviceInterface.Tele scopeAxes.axisSecondary, 100);

I have set tracking to sidereal and turned tracking on but the numbers in the RA/DEC boxes don't move. Anyone know of another command?

I'll try it on eqmod in the dome later.

[alistairsam]

Hi Robin,

Have you tried a value lower than 100 for the rate?
this is from the eqmod script

' Move mount west at sidereal x 2
' moveaxis(a,b) where a=0 for RA axis, a=1, for DEC axis, b is rate in degrees/sec
scope.moveaxis 0,0.00832

[Tandum]

Yes the rate was too small to see movement in the numbers. I upped the rate and it's away,just need to tune the rate at bit.

No it's not a script it's a C sharp app, you need timers for it to work and I don;t know how to do timers in a script.

[alistairsam]

hi

that app looks cool
does it work with the higher rate?
ideally sidereal x2 would be perfect as otherwise the trail would go out of the fov especially for 4 or 5 mins
timer in a script could be a simple loop as in here
http://eq-mod.sourceforge.net/samples/CCDDriftAlign.vbs
but if you've got it working as an app then its great
are there commands to initiate the image capture as well, would be neat to see in your app as the seconds variable can be used for the slew time and exposure time
cheers
alistair

[Tandum]

Here, see if it works for you : http://starshed.net/temp/ccd-align.zip

Unzip and run setup.exe. It will show up on the menus and in add/remove programs so you can uninstall it there.

The rate is the number of seconds to move 1 degree so you can change that but a rate change doesn't take effect if a run is already started. You need to stop it and start again. I set the range from 15 to 300 seconds. The other crap on the screen is mostly for debugging but I did add a honking big all stop button :lol:

I've only tried it on simulators.

[alistairsam]

wow that looks amazing, thats one serious looking stop button. I'll give it a whirl and let you know. John, look forward to your test results as well. sorry if I've hijacked your thread.

robin, I'm guessing that of you choose poth hub for the scope, you wont have to close maxim. I usually have stellarium and maxim connected to the scope via the hub.

cheers

[Tandum]

It seems to open a new maxim so you'll have 2 on the screen. If the first has a connection to the camera the second will fail to connect to the camera. That's the way com seems to work. I haven't turned any camera coolers on either. I'm not sure if moveaxis is implemented in poth or if the .net assemblies for poth are inside the program. eqmod is a hub anyway. Also I altered the rate range to be from 15-300. 1 degree a second was a bit quick :) It's overcast here so I can't see if it works.

[edit]
I got it now. It's only starting a second maxim when I run it inside the debugger. Run the installed app itself and it latches onto the currently running maxim :)

[Tandum]

I found a couple of problems, eqmod does not support the tracking rates command and when the app exits it disconnects the maxim camera for reasons unknown. I fixed the eqmod issue but I dunno why it's disconnecting the camera, I will ask someone.

Uploaded a fixed version (http://starshed.net/temp/ccd-align.zip) with all the debugging stuff removed.


[edit]
Fixed the maxim camera disconnect issue with a camera.DisableAutoShutDown = true;

[troypiggo]

Been a while since I've had an opportunity to do any astro-imaging or scripting, but I was writing a script that automatically slewed to where you'd want it on the horizon (you choose east or west) for ALT adjustment, started guiding with DEC off, you'd watch the Maxim guiding error graph and adjust ALT, start guiding again, then slewed to the meridian and repeat for AZ adjust.

I'll have to dig it out off my other computer etc, and can't recall if it was in a condition to make public. But it was close. Not quite the "V" image slew method you mention, but if you're using Maxim I found it handy.

[alistairsam]

thanks robin
'm hoping for some clear skies tonight so will test it outdoors else I'll try it indoors and let you know
got a few exception errors when I ran yesterdays version
would i need to install any dependencies like runtime files or anything?

hi Troy, would be cool to see your script as well. heaps of solutions, John your fingers would be happy once we get this working

cheers

[Tandum]

It was probably imploding when trying to set the tracking rate on eqmod. This is a real quicky,no error catching or any nice stuff like that. I've just uploaded v1.0.1 which has all the quick fixes and I added radiobuttons for exposure time. I guess it will take a bit of fiddling to get the dec rate and exposure time right for your field of view. I'll try and figure out how to save these settings between sessions.

It looks like this now.

[pmrid]

Robin, full marks for this app. It will be a real boon to those who have to polar align.

I have run the app in daylight and chosen a 60 second exposure.. Th exposure competed without a hitch. I'm not sure if the scope moved or not - I forgot to watch/listen. But I am wondering how to do the reverse direction movement and exposure. Am I right that the idea is still to do a forward then a backward movement on the one exposure?
Peter

[alistairsam]

Hi Peter,
fully agree, this would be a cool tool.
and yes, it needs to slewed E and W for the same exposure. But I think you get the reverse direction simply by turning off tracking so the star drifts.
that's what the eqmod script does.
so for the start and end points to be the same, the rate would need to be sidereal x 2, thats 0.00832 deg/sec. Else you'll need to do a slew in the reverse as well.
I don't think you'd need to change the rate and this rate works well for exposures till 5 minutes.
you can view multiple stars within the FOV and you could look at the brighter ones.
I usually use 2x2 bin on the 30sec exposure as that makes it bright and you can clearly see if it is a V or a straight line.

Robin, I'm still getting dot net errors. maybe i'm missing something.
System.Runtime.InteropServices.COME xception (0x80040154): Retrieving the COM class factory for component with CLSID {26819E05-F8BA-11D2-AA9C-444553540000} failed due to the following error: 80040154.

I'm guessing this would work for the gemini.net driver as well as that doesn't use eqmod.

[Tandum]

Peter,
If you pick say 30sec, it actually exposes for 35sec. It is stationary for 5 sec to get a bright spot from the star, then it moves dec one way for 15sec and then back for 15sec so in the image you should see a line or a V with a bright dot at the start point. There should be a balance between dec rate and exposure time which you'd need to fiddle with to get it right for your FOV. It's probably easier to look at the dec numbers on the ascom control pad for your scope to see if it's moving dec.
Of course I have a house due south of me so ..... I can't test it :)

Al,
That CLSID number belongs to maxim.CCDCamera. Try starting maxim as administrator. So right click on maxim_dl.exe and run as administrator. If you've never done this before, maxim hasn't registered itself correctly. For win7 anyway. So long as the ascom telescope responds to the moveaxis command it should work on any ascom scope. I think they will all implement moveaxis command, that's like the command onscreen direction buttons talk to.

[alistairsam]

Hi Robin,

Did you mean it moves in RA for 15sec and back?
Dec should be stationary and the V comes up due to the drift in Dec caused by the alignment being off.
so you slew East then west and if PA is off, the star would have drifted in Dec perpendicular to the East/West axis/line and that gives the gap between the start and stop.

[Tandum]

Ahhh ****, yeah, backwards. I thought I read the doc properly.

I've uploaded another version, v1.0.2

[scagman]

Thanks Robin for your hard work. This looks really good. Much more than I expected. I tried V1 last night and that worked well using the simulator. I'll try and test V1.0.2 later tonight when I get home, wont be till late thou.

Cheers

[Tandum]

Grab the newest copy here john and let me know what's wrong with it :)

Newest is labeled v1.0.3

I pulled the cameras from the dome to clean them but haven't had time to do it yet.

[Tandum]

Grab the newest copy here (http://starshed.net/temp/ccd-align.zip) john and let me know what's wrong with it :)

Newest is labeled v1.0.3

I pulled the cameras from the dome to clean them but haven't had time to do it yet.

[alistairsam]

Hi Robin,

I get this error after the 5 sec wait when I set the rate to 300s.
ASCOM.DriverAccessCOMException (0x80040401): MoveAxis set - '0.00333333333333333' is an invalid value. The valid range is: -50 to -10, 0 and 10 to 50. ---> System.Reflection.TargetInvocationE xception: Exception has been thrown by the target of an invocation. ---> System.Runtime.InteropServices.COME xception (0x80040401): MoveAxis set - '0.00333333333333333' is an invalid value. The valid range is: -50 to -10, 0 and 10 to 50.

Same with 200, 100, 60s.

[Tandum]

What scope driver is this?

The ascom docs say :
Both the Minimum and Maximum properties are always expressed in units of degrees per second. This is only using for Telescope InterfaceVersions 2 and 3

10 to 50 seems big.

[troypiggo]

Here's a link to the VBS I was talking about above. I haven't played with it for a while. Will fire up the simulator to refresh my memory some time shortly.

https://www.dropbox.com/s/dm9z6naujejccfw/DriftMax.vbs

[scagman]

Hi RObin,

Just been testing your app. I also get an error.


Unhandled exception has occurred.

MoveAxis set - 'axisPrimary' is an invalid value.


************** Exception Text **************
ASCOM.DriverAccessCOMException (0x80040401): MoveAxis set - 'axisPrimary' is an invalid value. The valid range is: guiding, centering, or slewing speeds. ---> System.Reflection.TargetInvocationE xception: Exception has been thrown by the target of an invocation. ---> System.Runtime.InteropServices.COME xception (0x80040401): MoveAxis set - 'axisPrimary' is an invalid value. The valid range is: guiding, centering, or slewing speeds.
--- End of inner exception stack trace ---
at System.RuntimeType.InvokeDispMethod (String name, BindingFlags invokeAttr, Object target, Object[] args, Boolean[] byrefModifiers, Int32 culture, String[] namedParameters)
at System.RuntimeType.InvokeMember(Str ing name, BindingFlags bindingFlags, Binder binder, Object target, Object[] providedArgs, ParameterModifier[] modifiers, CultureInfo culture, String[] namedParams)
at System.Type.InvokeMember(String name, BindingFlags invokeAttr, Binder binder, Object target, Object[] args, CultureInfo culture)
at ASCOM.DriverAccess.MemberFactory.Ca llMember(Int32 memberCode, String memberName, Type[] parameterTypes, Object[] parms) in c:\BuildAgent\work\c9b0aabbc621eced \ASCOM.DriverAccess\MemberFactory.c s:line 429
at ASCOM.DriverAccess.MemberFactory.Ca llMember(Int32 memberCode, String memberName, Type[] parameterTypes, Object[] parms) in c:\BuildAgent\work\c9b0aabbc621eced \ASCOM.DriverAccess\MemberFactory.c s:line 469
at ASCOM.DriverAccess.Telescope.MoveAx is(TelescopeAxes Axis, Double Rate) in c:\BuildAgent\work\c9b0aabbc621eced \ASCOM.DriverAccess\Telescope.cs:li ne 686
at ASCOM.tele.Form1.moveScopeWest(Obje ct sender, EventArgs e)
at System.Windows.Forms.Timer.OnTick(E ventArgs e)
at System.Windows.Forms.Timer.TimerNat iveWindow.WndProc(Message& m)
at System.Windows.Forms.NativeWindow.C allback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)

Looks really good.

[Tandum]

What is your mount and driver ? It seems it don't work on old ascom mount drivers.

Are you using drive ASCOM Gemini Telescope .NET version 1.0.18 with ascom 6.1?

I've uploaded ver 1.0.4 which will not solve your issue but crash in a more graceful way :)

[Tandum]

Al was using simulator for .net. yahoo ascom group says that simulator has bad code, don;t use it.

[scagman]

I'm running ASCOM Gemini Telescope .NET version 1.0.29 with ascom 6.1.

[Tandum]

John, it looks like the gemini driver does not conform to V2 or V3 of the ascom driver standard, it must still be V1. There are no docs I can find online that describes the V1 standard so I'm a bit lost.

Just maybe, if you connect to poth and then connect poth to the gemini driver, poth might convert it to V3.

I'll see if I can knock up something that will dump it's guts to a txt doc so we can see what is going on inside the thing. Without a Gxx mount here I can't do anything.

[alistairsam]

Hi John,

try this script with your gemini driver. it doesn't do anything with the capture, so you'll have to start that manually after running this.
it won't bring up any window, but you should see the RA change.

its the same eqmod script I mentioned earlier, but I've added the geminitelescope object.
in theory it should work if I got the object name correct.

just save the attached file, rename to .vbs and run it. watch the RA values, it should increase and then decrease or vice versa.

if it works, just create copies of the file and edit the 60 to whatever you want. just use half the value. so if you want 30secs, use the value 15.

this works fine for the eqmod driver. But I'll be trying out Robins app. that's way better. didn't get a chance last night.

Cheers

[alistairsam]

Hi Troy,

That is a cool script, I'm still trying to understand how it works, will try it and let you know.

Thanks

[troypiggo]

Yeah, sorry about that. I'll have a play on the weekend and try to document better. I think most instructions were in the script text comments.

[Tandum]

Al, what is timer in your script :
endtime = timer + 60

It's not initialised or dimensioned, is it built into windows script?

Also doing a loop like that drives the cpu to 100%. You're better off with a sleep command. WScript.Sleep(60) That pushes it into the background.

Did that C# app work ok on your eq6?

[alistairsam]

Hi Robin,
i have no idea.
I found it at the bottom of this page.
http://eq-mod.sourceforge.net/tutindex.html
the ccd align script and I just adjusted the time. it works for me with cpu at 5%. so not sure why its 100%

nope, haven't had the chance to try the c#app with the eq6 yet. hoping for some clear skies tomorrow or will test it indoors and let you know.

Thanks
Alistair

[Tandum]

Loops make the cpu shift data in/out continuously.
If you got an i7 there are 8 execution threads on 4 cpus so you should see 12% usage.
One execution thread should be flat out.

But the similarity between theory and practice is purely coincidence :)

[alistairsam]

so would the WScript.Sleep(60) solve the issue as you'd suggested?

I'd very likely be using your c# app with eqmod if there are no errors, but since John's v1 driver doesn't work with the app, I thought the script might be an alternative for him.

Cheers
Alistair

[alistairsam]

would this be correct?

[troypiggo]

I've taken my script down. Had an error that I need to look at before public use.

[Tandum]

It's not that important Al but that should do the same thing.

According to the gemini group you can't just shoot any old numbers at the driver using moveAxis. You have to read the AxisRates collection to see what it supports and only shoot one of those numbers at it. http://www.ascom-standards.org/PlatformHelp/html/M_ASCOM_DeviceInterface_ITelescopeV 3_AxisRates.htm

Don't be like that Troy, everyone enjoys a sold crash every now and then :)

[Tandum]

John,
Can you get this app, install and run it. It connects to the G11 driver, asks it for it's RA axis rates, dumps the info to a text box. Can you copy paste that to notepad or something. I need to see what comes back from it.

http://starshed.net/temp/axistest.zip

[scagman]

Hi Robin,

attached is the AxisRates.

Alistair,

Your script returns the following error

MoveAxis set - 'axisPrimary' is an invalid value. The valid range is :guiding, centering, or slewing speeds.

Cheers

[Tandum]

Cheers John, That's what I thought would come back and peter ran it on his titan and got a similar response. The way gemini works is the user sets these rates in the gemini driver as guiding, centering and slewing speeds so they can change depending on what the user has set them too. Gemini won't accepts any other rates. I've altered the app and if it sees rates with a minimum = maximum it scrolls through all the rates and picks the slowest rate figure. I told it to pop a box and tells you it has found a gemini so you'll know if it picked it up.

You can get v1.0.5 from the same place. I'd uninstall the old one before installing the new one. I'm not sure how visual studio creates these install scripts so it would be best to uninstall first. Also uninstall that axis test. Hopefully this one will work on gemini.

One point, to stop the scope movement I send it a zero in the moveaxis command. Theoretically it should reject a zero as it's not in the list of accepted rates. If this will crash it, it will do it at the end of the run.

[scagman]

Thanks Robin,

I will give it a go tonight.

Cheers

[Tandum]

Does this work for anyone?

It looks like a great method which is why I spent time doing it.

It's been raining here for days so I can't try it and I'm just guessing at the numbers :)

[alistairsam]

Hi Robin,

I'm just trying to fix my windows permission issues. will then try it out and let you know.
not sure about others, but for me and I'm sure John and I'm guessing Meru, this would definitely simplify polar alignment.


Cheers
Alistair

[Tandum]

It's just always **** outside here Al. I think all of queensland will become one giant rain forest, or swamp.

I have managed to rebuild my bmw K100RS instrument cluster to tonight :)

And I got a jack for the bike :)

http://starshed.net/K100RS/Jack.jpg

[scagman]

sorry Robin,

I didn't test it last night. It was a cold, wet and windy night here last night and I didn't venture outside once I got home next to the fire.

Great looking jack. Should make life a bit easier.

Cheers

[scagman]

Hi Robin,

Tried V1.05 tonight but still getting error popups. The camera operation works fine. but as soon as the 5secs are up I get error messages about not being able to set rates.

Cheers

[Tandum]

Bugger, I can see I'll need a gemini. I've pm'ed you john.

[alistairsam]

yeah, maybe a teamviewer session, debug and code in real time!! ;)

[scagman]

Hi Robin,

I have teamview, so if youd like to use that for testing that would be fine.

[Tandum]

No good John, you'd have to also have Visual Studio 2010 installed.

Also going by your last PM the scope is moving but at slew speed. I found that bug, it wasn't scanning the full rate list and just selecting the top rate. I fixed that in v1.0.7 which I have uploaded.

I have my fingers crossed it does the trick.

Your rate should be 0.00208903731114552

[scagman]

Hi Robin,

Thankyou very much. V1.07 works great. Its does have about a 6-7sec delay in the change of direction.

If it was possible to extend the move time for the west(return move) by 6-7secs would be good. Its not a problem on the longer runs but on the 30sec exp. the return line will be very short.

But otherwise it works great. M fingers and buttons thankyou very much for the time and effort you put in.

Cheers

[Tandum]

Excellent, now take that gemini and ......... :)

Er, it should go west first I would have thought. I fixed that too.

There is no delay between them, unless the timer is slow or the mount has backlash.

Or is it just star movement, I've ignored that completely. Still overcast here :(




[edit]
Here's the new location (http://starshed.net/apps/ccd-align.zip) for it if you want to try it. Newest version is v1.1.0
And here's the PDF (http://www.observatory.digital-sf.com/Polar_Alignment_CCDv1-1.pdf) to tell you what it tries to do.

[scagman]

Hi Robin,

I could have gotten the e/w directions the wrong way. RA would count down then increase. I was pointing at the SCP (park postion) when I tested it. I dont think it matters which way it moves first, or does it?

I havnt tested it on a star yet, clouds here too. I dont think the delay is your app, it seems to take time from when I think the app issues the change of direction to the mount stopping and changing direction. There is some backlash but I would hope that it's not 5secs worth.

I'll try and give 1.1 ago tonight.

[alistairsam]

Hi Robin,

Great to know its working with the G11 as well. John would be a happy man.
I've also tested it with the eqascom simulator with stellarium showing the scope slewing east and then west.
Very nice work and thanks from me as well.

one thing I noticed is although the rate and duration are the same in both directions, it doesn't go back to the same position. that's because the stars are constantly moving west at sidereal and that would explain why John is seeing the return trail shorter.

the basic math
tracking rate is 15arcsec/sec
so 1deg = 3600arcsec in 240secs.

so if we enter 240 in the program and slew east, that would mean that the star would be more or less at the same place as you're travelling east the same rate as west. i've tested this with the scope in stellarium and it stays put.

so simple solution would be

slew west at 120s, thats sidereal x 2 for half the duration, and then stop tracking. the star would then drift to the start point at sidereal.
that would give the complete V.
Then resume tracking at sidereal.
so you don't really need to adjust the tracking rate and you could default it to 120, but just start west instead of east.

This is what the eqmod script does as well.

what do you think?

btw, there's a small demo video I captured here showing the reticle moving the way it is and your software

https://www.dropbox.com/s/hgv0rvl5ucargd4/ccd-align-demo.avi

Cheers
Alistair

[scagman]

Hi AListair,

I havnt actually tried it on a star yet. I just watch the RA value and how it changes. I have tested v1.1.0 tonight, again with no stars.
RA ends up back were it started from, albiet after the image has finished exposing. I have a 5-7sec turn around between changing directions.

Eg. 30 sec selected. Press "move", maxim starts to capture, at 5sec mount starts to move, at 20sec CCD-Align issues change of direction, it then takes 5-7sec before the RA changes direction, at 35sec maxim stops exp. at about 41secs RA stops, usually back at start postion.

So for the 30 I have a star trail that is at least 15secs long one way but the return trail is only about 8secs long.
60 I would have a trail 30sec long and the return trail would be about 27-28 sec long.

If this weather clears I will be able to try it with a star but I cant see any problems with it.


Hope this makes sense.

cheers

[alistairsam]

Hi John,

Do try it with the simulator and the skyx and see if it goes back to the start.
I'm not sure why you're seeing the 6-7sec delay, i can see the change in direction immediately with the eqascom simulator. i havent checked the ra value though. will check again.
Cheers

[Tandum]

If I connect the ascom simulator to sky6 I see it going west first, no delay in reversal and it comes back to the star it started on. Skippy says it will be clear here tonight so I might be able to try it.

Maybe the gemini needs a stop command before reversing?

[edit]
OkyDoky, got to run it. v1.1.0 moves east first and yes it needs more maths to take star movement into account. I guess there's no point having rates that are less than sidreal either. I'll look into it. On a side note, I get straight lines out of it :)

[Tandum]

I've found the rate needs to be much higher than I thought. I've removed the rate box and fixed the rate at 12sec per degree and selected centering speed for gemini, which is probably what they use for the hand set buttons anyway.

I've also made it go west longer than east so it should get back close to where it started and I stopped it before reversing.

The ASCOM simulator has east/west arse about which confused things.

I've uploaded v1.1.2

[alistairsam]

hi Robin

how did you get that rates need to be higher? was the line not long enough?
12sec per deg is about 10x sidereal and for a 5 min exposure, might ovetshoot the fov. not that it matters. I guess pixel scale is also a factor, I didn't think of that.
have you tried this, in a real test, stop your tracking for 1min while exposing and see what the length is.
can I suggest, having two modes, one for eqascom where rate is 2xsidereal or maybe a choice between 2 4 and 8x. and a gemini mode with the centering speed you mentioned.
I still think going west for 30s at 2x and turning tracking off for 30 would get it to the same start point, you could easily do this with the eqascom buttons.
will be interesting to see your result when you did this.
btw congrats on the straight line.. its the 4 or 5min exposure that confirms your PA.
cheers

[Tandum]

It worked better. I can't tell a gemini to do double sidereal rate, I can only pick from it's rate list and centering speed is about 12. John's guiding speed is pretty much sidereal rate. I don't think it matters if it goes out of field. I also tried balancing movement times by calculating an offset based on rate minus sidereal and adding that to time west then subtract it from time east. Doesn't do much though :(


[edit]
I found the problem with the time offset. It wasn't converting a double to an integer correctly resulting in a time offset of zero. I've fixed it and uploaded another version (v1.1.3) :) It will pop a box when you change exposure time and show the rate and time offset.

[scagman]

If I could get some clea skies I'd be able to test it on some real stars. Last night was looking good early on but truned to crap.

Doesn't look good for the next few days either.

[Tandum]

John,
V1.1.4 is up and it's very close. I'm trying to tune the time offset settings to get the star back where it started. This version still puts the star back a little too far east on the 600sec exposure option.

[scagman]

Finally the clouds parted and I was able to get a chance to use CCD_Align last night.

Brilliant!.

I tried V1.1 and V1.3. I didnt realise V1.4 was there. I'll see if I can try that tonight.

The stars looked like they ended up pretty close to where they started.

Thanks again Robin.

[alistairsam]

great news. I'm hoping to try it tmrw.

Cheers

[Tandum]

How'd we go guys? Close enough?

If it works for you, I'll turn off the popup boxes and finish it.

[alistairsam]

Hi Robin

the weathers been really bad out here clouded up or raining quite often
hoping to give it a try tmrw if the weather holds

have you tried it? how does this method measure up against others you've tried?l

cheers
Alistair

[scagman]

Hi Robin,

I was able to try V1.1.4 out last night. It moves way to fast. A 30sec exp moves well out of my FOV then comes back. Were V1.1 a 120sec sec exp. mostly says within my Fov. The last version I have that is at the correct speed is V1.1. I havent got 1.1.2 or 1.3. I thought I did, but when I run them they are 1.0.7. Not sure If I did ever run them missed them due to weather I think.

Cheers

[Tandum]

Yes I changed it to gemini centering speed instead of guiding speed. Guiding speed on yours is about double sidereal rate. Does it matter that it goes off the ccd? I can change it back but I'll need an algorithm to work out the exposure offset figure otherwise it will go east 60% faster than it goes west due to star movement. This offset doesn't matter as much at higher speeds.

[scagman]

I dont think it will matter if it goes of screen, It does anyway at guide speed at the longer exp. I'll try and test a 10min exp tonight to see how far it actually does move. Providing the clouds behave. I'm getting a very nice single line at 10min exp's at the moment which makes it hard to see where the star is returning to. ( I'm not complaining about that thou).

[peter_4059]

Not meaning to hyjack this however if you have a G11 then pempro has a very nice drift alignment tool - it calibrates itself then tells you how far out you are in alt and az. Because the G11 alt and az adjustments are calibrated it is very quick to get aligned.

http://www.losmandy.com/g-11-manual.html
Each complete revolution of the altitude knob raises or lowers the polar axis 3.2°
Each complete revolution of the azimuth knob is .82°.

Here's the pempro polar alignment tutorial:
http://www.siriusimaging.com/Help/PEMProV2/polaralignwizard.htm
In step 10 and step 13 it tells you exactly how much to adjust each axis.

[Tandum]

I'm a little confused with the numbers now. sidereal is about 0.00418, John's gemini guiding speed is 0.00208xxxxx.
Isn't that half sidereal rate? Does the mount add the move speed to it's current tracking speed?
If that's true then I can ignore star movement.

Hi Pete, this is just another way to do it for free.

[scagman]

Hi Peter, I now have pempro and spent a few hrs last week trying to sus it out. I tried the PA wizard but am having problems with it at the moment and need to spend some more time playing with it to sort out why its not playing with me. (or maybe the problem is it is playing with me:D).

The CCD-align method works well, and is simple to work out:).

Robin,
I thought guide rate was 5x sidereal and centering was 20x. I dont know how that translates to your numbers.

Cheers

[peter_4059]

What problem are you having with it?

[scagman]

When I get to step 6 I think it is. Pempro is having difficult in finding a star. Its suppose to increase the exp. length and take another image but the exp. length doesnt change. It also uses a 32x32 pixel subframe and doesnt increase or move the subframe to a differant part of the image while looking for a star. I tried moving the mount so there was a star in the subframe but still Pempro just keeps looking, saying its increasing exp. time but never does.

I'm just waiting for some clear nights so I can have another go. I have been trawling thier forums and I think I have to play around with the settings in one of the earlier steps.

Sorry for the crude description as to what is happening.

Cheers

[peter_4059]

I had a similar problem with finding a star. I found taking darks helped and there is also a setting somewhere for the exposure duration that I increased - can't find it at the moment though.

Found it - it is step 2 of the polar align wizard under camera information advanced settings I'm using fixed exposure duration, dark subtract and 1 sec exposures. You can also change the tracking rect size here. I've also got the min/max star brightness set at 1000/30000.

[Tandum]

Sidereal is 0.00418 deg per sec, according to that dump you did, your guide rate is 0.00208903731114552 deg per sec which is half sidereal rate. So the mount must be tracking at sidereal rate and adding the rate in the move command.

[Tandum]

Just looking to see if my new Avitar works .. It's Tiny :( ... it fixed itself :)

[scagman]

yes it is a bit small.

[Tandum]

If v1.0.4 is too fast for you, I uploaded 1.0,5 which uses slow speed and no offsets.

I have not archived any of these versions, what ya see is what ya get.

[alistairsam]

thanks Robin

We're hoping to have a clear night on fri or sat after a long wait, so will try and let you know

Cheers
Alistair