Setting a more accurate "park" position (HEQ5/NEQ6)

[JamesF]

(For clarity, we're talking about kit permanently mounted on a pier in this instance.)

I'd like to be able to get "acceptably close" to targets on the first slew from the park position, specifically in those cases where plate-solving isn't feasible such as slewing to the Moon (and the Sun, come to think of it) or doing so during the day.

Tracking/polar alignment is pretty good already.  It's just having the telescope actually point where the mount thinks it's pointing at the start that is the issue I believe.  I'd guess the error is perhaps around a degree or so.

What are good ways to improve this?

James

[Thalestris24]

1 hour ago, JamesF said:

(For clarity, we're talking about kit permanently mounted on a pier in this instance.)

I'd like to be able to get "acceptably close" to targets on the first slew from the park position, specifically in those cases where plate-solving isn't feasible such as slewing to the Moon (and the Sun, come to think of it) or doing so during the day.

Tracking/polar alignment is pretty good already.  It's just having the telescope actually point where the mount thinks it's pointing at the start that is the issue I believe.  I'd guess the error is perhaps around a degree or so.

What are good ways to improve this?

James

Hi James

I don't have a pier (I wish!) but I have limitations when it comes to alignment. My kit is semi-permanently set up. What I do is platesolve and then at the end of a session, I slew to a convenient position which isn't far from 'home'. Then I use eqmod to 'park at current position'. I find that it's then quite close to any targets next time I do some imaging. Having said that, I'm surprised you're a degree out - there must be an error in your home position? If you can correct that then I'd expect you'd be pretty much spot on from the get-go.

Louise

[JamesF]

2 minutes ago, Thalestris24 said:

Having said that, I'm surprised you're a degree out - there must be an error in your home position? If you can correct that then I'd expect you'd be pretty much spot on from the get-go.

I think that's exactly the issue, Louise.

Specifically I want to get it more accurate so I can do things like hit the Moon (or Sun) first time with a Mak or SCT when I'm running things remotely.  At the moment I can get sufficiently close with the Mak that by removing the camera and looking down the tube I can see the reflection on the inside which means if I'm right next to it I can nudge things in the correct direction.  It's a touch more tricky to do that from the office though, and I'd not be keen to do it that way with the Sun (though alternatives exist if you're at the mount). 

James

[Thalestris24]

3 minutes ago, JamesF said:

I think that's exactly the issue, Louise.

Specifically I want to get it more accurate so I can do things like hit the Moon (or Sun) first time with a Mak or SCT when I'm running things remotely.  At the moment I can get sufficiently close with the Mak that by removing the camera and looking down the tube I can see the reflection on the inside which means if I'm right next to it I can nudge things in the correct direction.  It's a touch more tricky to do that from the office though, and I'd not be keen to do it that way with the Sun (though alternatives exist if you're at the mount). 

James

Hi James

Why do you think you have the problem in the first place? PA error? Or star alignment / pointing error? Something else? You can obviously correct PA error by drift aligning. I'd have thought that once that is accurately set, it should remain accurate. I only redo my PA occasionally and/or if I change or move something. If it's pointing error, then either creating a more accurate pointing model in eqmod (never done it myself). Or using platesolving should sort it - but you then have to make sure you use 'park at current position' rather than going back to the saved 'home' as that will lose the platesolving position - if you get my drift! I suppose if there's a problem as a consequence of returning to 'home' then lat/long settings could be made more accurate?

Louise

[JamesF]

The PA is fine -- less than an arcminute error.  You're idea about "parking at the current position" may perhaps give me a way out, if I tell the mount to move to a position that is very close to, but immediately south of, the NCP.  If at all possible I'd like to avoid solving this in software so I have the ability to switch between controlling PCs and have things work just the same (handy for developing software so I don't mess up a system that I want to rely on continuing to work :)

I guess the question comes down to:

Given a mount with good PA, how does one ensure that in the park position the OTA is also, as close as is practical, pointing at the NCP with the additional constraint that the optical axis of the OTA should be vertically above the RA axis?

James

[Thalestris24]

3 minutes ago, JamesF said:

The PA is fine -- less than an arcminute error.  You're idea about "parking at the current position" may perhaps give me a way out, if I tell the mount to move to a position that is very close to, but immediately south of, the NCP.  If at all possible I'd like to avoid solving this in software so I have the ability to switch between controlling PCs and have things work just the same (handy for developing software so I don't mess up a system that I want to rely on continuing to work

I guess the question comes down to:

Given a mount with good PA, how does one ensure that in the park position the OTA is also, as close as is practical, pointing at the NCP with the additional constraint that the optical axis of the OTA should be vertically above the RA axis?

James

Do you normally use an eqmod pointing model? I think eqmod saves that,so should be accurate, and I don't think there would be any problem using the same model data across different computers. It also saves the current position if you select it rather than home. The thing is, if there's a pointing error, where is it coming from? I reckon that's the puzzle you need to solve. If from the home position, you slew to somewhere else and then slew back to the NCP, is that then still the home position as far as eqmod is concerned?

I'm sure you can solve this!

Louise

 

[Thalestris24]

This might be useful: http://www.awrtech.co.uk/telerrr3.pdf

 

Edited November 13, 2019 by Thalestris24

[michael8554]

Can you not centre the moon with the image from your guidecam? 

Or centre by the scope using the finder?

I suspect I'm missing your point...... 

Michael 

[kens]

I use the Static Polar Align (SPA) tool in PHD2. If your polar alignment is already good you use SPA to find the centre of rotation of your mount. Then adjust in declination to bring that point to the centre of your field of view. Any offset that is at right angles to declination is due to cone error so you may want to adjust that too.

Once centred set your mounts home position there.

 

[R26 oldtimer]

If I am getting James question right, it's about more about setting the mount in an accurate home position (weights down-scope pointing North), rather than polar aligning (which comes next).

If that's the case I've found this tutorial quite helpful: https://www.myastroscience.com/homeposition

[JamesF]

3 minutes ago, R26 oldtimer said:

If I am getting James question right, it's about more about setting the mount in an accurate home position (weights down-scope pointing North), rather than polar aligning (which comes next).

If that's the case I've found this tutorial quite helpful: https://www.myastroscience.com/homeposition

I think that has to be worth a try.  At least if I put a spirit level on the counterweight bar I should get some idea how far out the RA is.  If it isn't then I may have some head-scratching to do.

James

[Starflyer]

1 hour ago, JamesF said:

I think that has to be worth a try.  At least if I put a spirit level on the counterweight bar I should get some idea how far out the RA is.  If it isn't then I may have some head-scratching to do.

James

This works, I used to do something similar but in EQMOD. For this to work you'll need to clear the pointing model first in EQMOD.

Get the mount roughly in the home position, connect EQMOD and unpark. Click on the + sign, top left, so it shows the clock dials for RA & DEC. Slew the RA axis by 90 degrees using the clock dial to get it as close to exactly 90 degrees as possible. Use a spirit level on the RA part of the mount, unlock the RA clutch, set it level and lock the clutch.

Park the mount, unpark and repeat for DEC, using the scope / tube rings etc to get your leveling surface.

This, together with good PA, accurate location and time, and low cone error will get you very accurate go-tos.

[kens]

10 hours ago, R26 oldtimer said:

If I am getting James question right, it's about more about setting the mount in an accurate home position (weights down-scope pointing North), rather than polar aligning (which comes next).

If that's the case I've found this tutorial quite helpful: https://www.myastroscience.com/homeposition

I assume you are commenting on my post. Whilst it is easy enough to get a weights down position with a spirit level it is harder to get the declination aligned to the pole. The tutorial I linked shows how to use the SPA Tool to help point the OTA at the pole rather than for polar alignment. Any tool that calculates the centre of rotation could be used. Indeed, one could set a longish exposure time and rotate the mount to get an arc and eyeball the centre of rotation. If the mount is already polar aligned then the centre of rotation is close to the pole and adjusting the OTA in declination can bring that point to the centre of the FOV at which point you are at dec +/-90 within a few arcminutes. Any residual offset that is orthogonal to the declination movement is due to cone error. If that is substantial you would need to shim the OTA to bring the centre of rotation to the centre of the FOV.