Still struggling with GOTO alignment

[Jessun]

I would happily argue for both !

The home position is just a calculated number of ticks away from last known position - a star. For a one star alignment: find a star, loosen the clutches - or move by handset, center star, enter and tick back home . You'll end up with the same home. (Two or three star alignments help to compensate for cone error, if I'm not mistaken, and become quite useful if you do meridian flips - I personally never bothered).

So I don't see a difference between clutch on or off for the first star - the only certain thing is that the star is where it is and once centered and entered it doesn't matter how you got there. Then to park, the mount will calculate the ticks required to go back from that last known position to the weights down home position. If your mount is leaning a bit it won't affect the home position - a spirit level on the weights should be pretty spot on anyhow. (Your setting scales will appear be off though). Power down, then fire up the mount and it will perform its only guess of the day - 'I guess I started from the home position'. Chances are then good though that you'll be much closer to the first star on a subsequent alignment if you left the mount where it was, but inaccurate timing as Per points out is the next evil...

What I'm interested in Neil is that you mention a lot of drift during imaging after you finally get going. Now that is a whole different problem, totally unrelated to the GOTO accuracy you battle. Pointing and tracking are two different things. (OK it get's a bit blurred with absolute encoders and pointing models but that's not the case here).

Flimsy OTA designs aside, only three things can upset traditional tracking: Polar misalignment, atmospherical diffraction and finally that the mount actually doesn't move at the correct rate - and there are sadly many sources for errors in the mount's rate. I can only assume you track at sidereal rate and not one of the other, like solar etc. I can't remember how clear that is on the SW handsets. Could you have a slip in the clutch?

Polar misalignment can be big if the polarscope has had a knock, but you say that it's spot on.

I'd spend all effort in looking into the drift, and then come back to the pointing accuracy. Good luck!

/Jesper

[chrisshillito]

So I don't see a difference between clutch on or off for the first star

The only difference is that with "clutch off" you remove the reliance on the mount controller / handcontroller to apply suitable corrections for home position error to every goto it subsequently does. We should remember that this discussion is in response to someone who is already having goto issues and that those suggesting "clutch off" are only doing so as it removes at source one "variable" that could be causing problems. The worst that can happen is that "clutch off" has no impact on the original problem - it can't make things worse. By eliminating one variable though it might just make the problem easier to diagnose.

Chris.

[Jessun]

The home position is not used to make any corrections at all. It's just a way to conveniently park the scope in a position that is relatively easy to find in the field using a spirit level, and if it hasn't been moved it's quicker to start from that position to find the first star. Then it's out of the equation. It's just there to give you a sporting chance to get you going in the rightish direction as you start aligning.

So it starts off as a guess, and is then forgotten.

/Jesper

[chrisshillito]

The home position is not used to make any corrections at all. It's just a way to conveniently park the scope in a position that is relatively easy to find in the field using a spirit level, and if it hasn't been moved it's quicker to start from that position to find the first star. Then it's out of the equation. It's just there to give you a sporting chance to get you going in the rightish direction as you start aligning.

So it starts off as a guess, and is then forgotten.

/Jesper

I can only comment from the perspective of skywatcher EQ goto mounts where the home position on most goto mounts is much, much more than just a convenient place to park. For these mounts at least the mount controller expects the mount to be powered up with the telescope pointing at the pole and at the home position. The initial initial slew is then calculated on that basis. If it the mount was not set at the home position when powered up then the initial slew will be certainly be out by a corresponding amount. Alignment is the measurement of this error(+ others) and gotos are the result of the subsequent correction of that error. The home position is never forgotten, it is the coordinate base from which all motor movements are made.

Chris.

[Jessun]

The initial initial slew is then calculated on that basis.

Yes, the initial slew is based on the guess that you've started at the home position - at least that is what you tell the handset so it can only trust you. Once you have centered and entered an alignment star in the handset the mount really knows where it is. It will now know that it is pointing at a known pair of coordinates. To move to the next star or object it will work on starting from those known coordinates, and move in RA and Dec according to the database. The home position is not part of that simple equation, and there is no correction to be made on that basis.

The stars are where they are, defined with arc second accuracy in the database, and a rough home position, impossible to get spot on in the field - perhaps off by many degrees - is completely useless from here on. The first time you hit enter on the first alignment star after slewing across the sky to find and center it, you are effectively saying 'sorry about that...' to the handset.

/Jesper

[Jessun]

Anyways, I think Neil has other underlying problems with the setup that make alignment seem inaccurate.

/Jesper

[Peter Drew]

And to think the majority of beginners buy GOTO telescopes because they think they will be so much easier to use!

[perfrej]

Jesper, I cannot agree with you. Again, introducing many compensation factors in the model to calculate where to go is stupid. Yes, of course everything can be corrected, but how good is the math of the model then, and in what way do accumulated errors affect it? Get rid of the polar alignment error. Get rid of the cone error. Get rid of the flex. Then, establish a good home position and everything will be very nice.

Since I discovered the de-clutch method, my NEQ6 has had a GOTO-performance that rival considerably more expensive mounts.

[themos]

I read the first post and it clearly says that there is drifting while imaging. If that's declination drift then the cause is bad polar alignment and that will throw off GoTos (unless you are using a proper mount-model pointing software)

[Jessun]

Per, what is it that you don't agree with?

That the initial home position you had on the night isn't a factor once you have aligned star one, or the argument that you can move to alignment star one with the handset? What I'm saying is that the home position is not a factor - since it's most likely wrong in the first place - so it's of no use, it's not an added compensation factor.

The handset will however protest if your first star is off by a mile, but that's most likely a way to protect the scope against a flip into the tripod or someting. A big time error will have the same effect.

/Jesper

[dph1nm]

The home position gets reset when you do an alignment - so it doesn't matter how precisely you try to set it up to begin with it will get overwritten! Also, the default home position (i.e. pointing near the pole - I believe you can change this with firmware 3.32) is not the ideal place to align on, so personally I wouldn't bother trying to be precise. I would say that loosing the clutches can be a very convenient way of starting the alignment - saves all that random slewing around with the handset buttons!

NigelM

[Jessun]

Regardless of schools for star alignment, I really do think you have something else going on there Neil. (Picking the wrong second star btw is something that I hold my hand up to doing a few times ... One star works fine for me, and I pick one close to the target I want to image).

It's the drift...

/Jesper

[broadsword]

All,

Hoping to get out later to try all these but the other half is protesting. Will update if I get a session in.

Thanks all for the debate

[Jessun]

Tell her it's URGENT

/Jesper

[broadsword]

Just clouded over so that b......s that.

[DaveGarland]

What firmware version are you using? I've just updated mine to the latest version and am having all manner of problems! I had no trouble at all with the last version but now I'm getting the Alignment Successful message but when I ask to se Jupiter say, it's way out! Also I understand the scope with the latest software always slews in from the same direction to cancel backlash but sometimes the scope now seems to slew back and forth several times before settling. Anyone else getting this???

Cold, frustrated and wishing I never did the update

[Astro_Baby]

By the way the OP did say he was usimg the de clutchimg method.

Have you tried just doing a two star align and see if the situation mproves ? The third star is for cone error and if there is no significant CE it might make no difference.

[Astro_Baby]

Ps I dont use the latest software ever. Thought ai'd wait and see if others had aggro....it seems they have. You can always go back to a previous version. If the SW site doesnt have it hit me up on PM and I will email hou a copy of the older version.

[DaveGarland]

Um, looking at the website I'm worrying I might have been a bit of a div :-S

I have an EQ5 Synscan and have installed the 3.32 (400kb) but there is an EQ Synscan version 3.27(288kb). Have I installed the wrong darn software???

[broadsword]

Broadsword - if you are after a "guru" look no further than Steppenwolf who has already responded.

Oops, that'll teach me to speed read through the posts. Apologies Steve, really appreciated the book and the views on the current topic.

[chrisshillito]

By the way the OP did say he was usimg the de clutchimg method.

Broadsword is doing that now, but to get the full picture of broadsword's goto adventures you need to also read the thread

Chris.

[knobby]

Um, looking at the website I'm worrying I might have been a bit of a div :-S

I have an EQ5 Synscan and have installed the 3.32 (400kb) but there is an EQ Synscan version 3.27(288kb). Have I installed the wrong darn software???

3.32 is the latest, so no, you're jot a div...

You are putting in 2013 aren't you...

[DaveGarland]

Yeees! Not that much of a div lol. Im going to backdate the software tonight. I dont think the slewing back and forth so many times can be right. Is it possible theres something wrong with the encoders? Has anyone ever taken them apart to clean them? Is that difficult to do?

[knobby]

You never know more than just a few of us have done it!

Maybe we should start a poll

3.27 vs 3.32...

[chrisshillito]

I dont think the slewing back and forth so many times can be right. Is it possible theres something wrong with the encoders? Has anyone ever taken them apart to clean them? Is that difficult to do?

That is extremely difficult to do as there are no physical encoders on an EQ5 synscan mount . The "encoders" are actually just internal counts that the mount firmware maintains and that correspond to the number of microsteps the stepper motors have been intructed to move - there is no actual feedback from the motor or the RA/DEC axis so say where the are currently positioned. This count is held in "volatile" memory and is lost on power down, and reset to zero on power up.

Chris.