Does the level of your mount affect goto accuracy on wake-up?

[g__day]

Last week I pointed out how a German equatorial mount can be polar aligned beautifully even if the mount itself is far from level. This is a consequence of it having two axes of freedom that needed to be pointed to a precise spot in the sky. Once you have this aligned, clever software can get the pointing right – even if the mount is not level.

But there was one point I didn’t think off – does the software build an internal understanding of whether the mount is level or not and retain it when the scope is switched off or put into hibernate mode.

This is obviously mostly relevant to folk with permanent piers – but all users of goto mounts can be affected.

Basically if your mount isn’t level – but is SCP aligned – this situation can be adjusted for in the software – and most likely is accommodated for by most mounts / hand controller combinations.

But what happens if you put your mount into park / suspend mode and re-wake it? Specifically does the mount keep not only the alignment data (e.g. 3 widely spaced stars) – but also the levelling data?

In the case of a Vixen Atlux with a SkySensor2000-PC hand controller – I suspect not. That is I have aligned my mount reasonably well on its pier – I’d say it’s within an arc minute of the SCP on both axes. But I’m beginning to suspect my mount is somewhere between ½ a degree to 2-3 degrees off level. Why – because on power on resume – my pointing is always about 30 arc minutes of its three reference stars in the same direction (mostly RA but also DEC) each day.

Once I re-align on these three stars pointing and tracking is very, very, very good. But on initial power-on resume from the previous nights viewing - all objects are consistently ½ a degree out of position.

Pondering why this could be (on the relevant Yahoo group) lead to the intuition that the level of the mount plays a key part – and this mount level or not information isn’t preserved between power–on – so the mount simply assumes a polar aligned mount is also a level mount!

Has anyone else experienced this type of behaviour? Appreciate your thoughts.

[JohnG]

Can't help you with a Vixen mount but I can tell you the Gemini just requires a Warm Start and a synchronization on one star and all GOTO's will be spot on. All the relevant info is kept in the Gemini unit and power is supplied by a small 2032 battery and used on subsquent startup's.

I must add that my mount is some 10 arc minutes out in elevation and some 13-14 arc minutes out in azimuth since I re-mounted the G-11, so does that effect the GOTO accuracy, no, the Gemini builds an accurate pointing model to compensate

Mine is kept in a Home position with the OTA horizontal, on startup I just put the counterweight down and the OTA pointing south using the Gemini handcontoller, then Align Star, GOTO bright star, Sync, then everything carries on from there, takes all of 5 minutes (if that).

Cheers

[Bassnut]

on restart, accuracy on 1st point depends on the accuracy of your PC clock. I sync the clock with a time server on the net. Is your PC clock correct?, or if your not useing a PC, is the mount clock accurate?.

[g__day]

Yes to 1 sec - every day for 2 months!

[JohnG]

The laptop I use is time synced every startup on the internet, the Gemini is updated every fortnight or so using a Star GPS so it is as accurate as I can get it.

Cheers

[montewilson]

What you are asking is possibly a bit too much for most mounts. There are probably too many variables to achieve it from a cold start. I have a Tak EM-200 and it doesn't have this feature. I have to do a small mapping run of at least six stars from which point it will be sync'd.

[g__day]

Monte,

The Vixen Atlux is a pretty amazing mount once its properly set-up and mastered. It's documentation is lacking, the gear adjustments needed some simple tweaking - but once that is done its pretty close to switch on, check calibration on three stars (avoidable once I figure this out), calibrate PHD and you're straight into imaging.

Given my error is consistent in magnitude and direction - every day, every single time I switch it on - even if its only off for a few seconds its a predictable (i.e. addressable) - not random error.

A consistent error can be modeled out. I did a new MaxPoint model and I found the mount is 33 arc minutes East of the SCP and 17 arc minutes low. This pretty much equates to the pointing error I see on start up (about 1/2 a degree), so maybe its not a level issue - maybe its a celestial pole mis-alignment issue given I've told the mount it is in aligned equatorial mode.

With its current polar alignment error - after a three star calibrate every goto is on my CCD for the C9.25 - never more than 1/3 out from the centre - and that's before MaxPoint kicks in to further improve pointing. Quite simply its an impressive mount. What I might be experiencing is the mount only wishing to be in polar equatorial mode if its better than 15 arc seconds off the SCP. More research is needed!

I've told the mount its in aligned equatorial mode (to allow PEC - Vixen state a PE or +/-3 arc seconds with PE enabled!). In unaligned mode the hand controller correct out polar mis-alignment in both pointing and tracking (which is way cool) - but this make drift aligning effectively impossible if you align on more than one star.

[acropolite]

I'm of the belief that if a GEM is correctly polar aligned that's it, the state of level of the tripod has no relevance. This issue has was recently discussed in this thread (http://www.iceinspace.com.au/forum/showthread.php?t=24093&highlight=level).

[montewilson]

It sounds like the Atlux is quite a decent mount - have fun. Here is a link to post I did showing how I got around the lack of fine adjustments on the Tak.

I use TPoint so I don't have any thoughts on MaxPoint, sorry but with TPoint I can get to within 10 - 15 arcseconds of the pole after a bit of work (about an hour) using, combination, TheSky, CCDSoft, AutomapperII and TPoint.


http://www.iceinspace.com.au/forum/showthread.php?t=22361

[g__day]

Monte - is Tpoint worth it in your view? I went for the far cheaper MaxPoint.

Phil,

Leveling absolutely isn't needed for tracking - with pointing I believe it might be affected on start up by how your hand controller models its sky version. More precisely for pointing lets say the hand controller maps a triangle of three well spaced reference stars onto the sky - that's all you need right? Originally I thought so, until I pondered does it map those 3 stars from a an absolute or relative position? If its absolute - based on assuming a perfectly level mount - things could go awry - if its relative (gear position only) - you have no worries.

Without knowing which sky model is used its hard to say definitely leveling is essential. Certainly with smart enough mathematics - and storing all essential modeling variables - a flat base wouldn't be needed. If your modeling assumed a level mount and didn't store assessed level of the mount you could experience what I see.

So I have at least two variables 1) level and 2) SCP mis-alignment and 3) if you also count backlash which could be playing a part - although I have almost perfect backlash on RA and only a little amount in DEC.

More experimentation is needed - I'm tracking in on a solution.

[g__day]

Where I am up to on this.

MaxPoint tells me I have the mount aligned so that the SCP pointing error is:

i) Polar Azimuth 00 degrees 01 arc minutes 32 arc seconds
ii) Polar Altitude 00 degrees 00 arc minutes 49 arc seconds

PC hand controller On start-up I select ESC on the SkySensor2000 - which tells it uses the last alignment configuration data it had when it was switched off the preceding night. I then do a 3 star confirming alignment on Rigel Kentarus, Achernar and Altair.

I am now measuring my initial alignment pointing error in RA and DEC (which will change as I commence re-aligning on always the same three stars this season).

1. Rigel Pointing Error RA -1 arc minutes DEC -12 arc seconds

2. Acherner Pointing Error RA nil DEC +8 arc seconds

3. Altair Pointing Error RA -1 arc second 9 arc minutes

* * * * *

I'm going to repeat this tomorrow - weather permitting to see if I get exactly the same pointing errors on wake-up.

Given I've gone from 30 arc minutes to under 10 arc minutes on resume, whilst my align has gone from 30 arc minutes to below 2 - this makes me think time accuracy, polar alignment, backlash in gears and the level of the mount are key non random factors.

Time will tell!

[jase]

Following this thread closely, goto accuracy on startup/wake up should not be an issue. Make sure you're not "eye-balling" each alignment star though. I'm assuming you're not doing this anyway if you've used PolarAlignMax to plate solve (i.e. to find the precise coordinates of a given image center). eye-balling is fraunt with danger as you will not produce consistant enough pointing. Automate building the pointing model using your camera (DSI). You may even find that you're already under 1 arcmin from the pole. If you find your mount does drift causing the inconsistencies, its very easy to slew and sync on startup with a quick plate solve.

[ballaratdragons]

Wake Up???????

You mean there is a way to start up my mount and it is all ready to go.

I have been 3 star aligning every time!

Does this 'Wake Up' thing work on an EQ6 with version 2.5a?

[g__day]

Jase - eye balling

I centre on alignment stars using my main OTA - through the viewfinder of a Canon 400D that is mounted to it. The Canon has 9 focus points in its viewfinder and the central focus box has a dot right in the middle. I always put the star I am using right on that and then select align to re-set alignment.

I noted that if I align on 3 stars - pointing is fine. But if I power down and resume 5 seconds later the same initial pointing error re-occurs - so its consistent - and not likely to be a time accuracy issue. Given polar aligning the mount better has helped improve power on alignment - I ponder its not just the mount is maybe 10 arc minutes off level?

PoleAlignMax isn't working for me yet as I have issues getting the FITS header on a Meade DSI.

* * * * *

I noticed doing 20 minute (1,200 second) guided test shots last night I am getting 4 mm (so about 45 arc second) star trails in RA only. Dec is fine but for some wierd reason PHD is note correcting out RA appropriately.

I noticed in a series of 5 consecutive 600 second shots the whole star field was slowly drifting down (RA) the screen each shot - so about every 6 minutes 300 seconds I am experiencing a 30 arc second net drift in RA that PHD is not correcting?

Very strange - I will ping Craig Stark of Stark Labs about this!

Ken

The Vixen Atlux has a wake on resume function - no idea if a standard EQ6 can also do this?

[Aster]

Ken,

I used to have a EQ5 Goto earlier this year. Upgraded synscan to ver. 3.1 as soon as I got it home.

Quite often I parked the scope pointing at an object low down , covered the scope, went to bed, got up a few hours later, told synscan to go to the object I parked the scope at and that was it. Continued from there without going to the rigmarole of a 3,2, or 1 star alignment.

Version 3+ had quite a few extra commands compared to version 2.5.

Maybe some of this will help.

Alexander

[ballaratdragons]

Ahhhh, so it only works if you leave the power to the scope turned on?

I meant from night to night.

[Terry B]

It works with version 3 as long as you park the scope when you are finished and don't release any clutches. When you restart it asks if you would like to use the current alignment. You just re-enter the correct time and away you go. Having said this I have never used it as I don't "yet":D have an observatory to leave my scope in.
Not sure that version 2 can do all this.

[ballaratdragons]

Sounds good.

No, version 2.5a never has asked me, so I'd say it doesn't have it.
Pity V3 is so expensive!

[Aster]

Expensive ??? Are we talking about the same thing. ?

Skywatcher HEQ5 oe EQ6 GoTo synscan used to be a free download/upgrade.

Alexander

[Terry B]

It still is but the version 2 can't be upgraded to the version 3 without a new hand controller.

[Aster]

That explains expensive

Alexander