Pointing accuracy of the SynScan Alt-Az

[Macavity]

Folks (me too!) tend to quibble over the "pointing" accuracy of the SynScan Alt-Azimuth mount, so thought I'd try to NAIL it, with (to?) a Skywatcher Pillar mount! I'm sure other folk's DIY skills are better than mine, but basically, I cut a custom-length ALUMINIUM "pier extension", and made a few random "circles" etc., of MDF lite, to interface the various "pole" elements. It is all surprisingly rigid!

If someone is interested, and wants details of precisely WHATIDID... I'd be happy to include a few more photos. <G> Prelimanary results suggest I've STILL to get better than c.a. ONE degree pointing accuracy... but Hey, I am still working on it.

Aside: I do now wonder whether this basic design of the mass-produced Skywatcher etc. mount casting has sufficient ORTHOGONALITY in the axes... Such seems (theoretically) essential to explore the ultimate in both Alt-Azimuth (and Equatorial) "GoTo" mounts? How I wish I could (maybe I can?) "motorise" my Giro III! But I'm always open to ANY (Alt-Az GoTo) ideas...

[rwg]

About a year or so ago I bought a SupaTrak mount (which is mechanically very similar to the SynScan AZ, without the fancy hand controller). Foolishly I did this in the middle of summer, so I didn't get much use out of it until (a) the evenings started to draw in a bit and ( the summer rain stopped.

To keep myself amused in the long, wet, summer evenings, I spent some time decoding the protocol between the hand controller and the mount and managed to put the thing under computer control. I documented it all here : supatrakmounthacking (rwgastro).

Anyway, coming to the point of all this, while doing this I discovered that under software control I could get the mount to point within about 15 arc seconds *as measured by its own axis encoders*. Actually real accuracy was nowhere near this as it is influenced by periodic error in the gears, backlash, etc. I seem to recall measuring the backlash and finding it was about 10-20 arc minutes on each axis.

Depending on how smart the hand controller is at running the motors, you might be able to do better than 1 degree accuracy if you are careful about always approaching alignment stars and targets from the same direction to keep the backlash out of the equation (or are you already doing this?).

I also thought a bit about your comment on orthogonality and it seems correct - if the axes aren't at 90 degrees then pointing (especially near the zenith) will be off by the misalignment of the axes. I think the same applies if the angle between the optical axis of the scope and the altitude axis is not a right angle. This could be even more variable as it depends on the wedge sitting squarely or not.

I now have a Celestron SLT mount (once again same hardware, but with a celestron rather than skywatcher hand control). I haven't really tested its accuracy in detail, but with a fairly cursory 2 star alignment I tend to find objects that i GOTO are within a low power field of view (about 2 deg real FOV). This seems similar to or perhaps a bit better than the results you are getting. If you let me know how you are measuring your accuracy I will try to measure mine more carefully the next time the sky clears.

Robin

[Macavity]

Interim Thanks for all that Robin. There is indeed MUCH to ponder and although I may not have any immediate observations (grin), I'm pleased to find (at least sense) one sometime(?), fellow "enthusiast" re. getting the BEST out of these mounts.

I like the idea of computer interfacing (I use Stellarium/scope) etc. It is anyway quite fascinating to watch the onscreen "meanderings" of the scope as it "hunts down" the object clicked upon! As they say, Darn Clever... etc.

I'm afraid I too haven't done to much (necessary) outside viewing - The weather here has not been particularly conducive to real "measurements" of the ultimate abilities of the WHOLE system. But I live in hope... In haste, C.

[rwg]

Hi Chris, yes I'm definitely keen to get the most out of these mounts - I think they're great from the point of view of being easy to set up and use, as well as affordable. They wobble a bit more than would be ideal, but I've still managed to take some 30s exposures with a nearly 1kg DSLR body hanging off the back of my scope with at least a little success. If the skies ever clear round here I'm going to have another go now I have a proper cable release and have tightened the altitude axis so it slips less freely. cheers, Robin

[jgs001]

There were a few things that were vital with my NexStar SLT to get accurate goto. The first was accurate levelling, the built in bubble on mine was out by quite a lot so I used a spirit level. Then acuracy on the alingment and consistancy on appraoch fo alingment, which also has to match the goto approach direction to ensure backlash was taken out of the equation. With that setup I could be pretty confident of an almost spot on goto each time.

[rwg]

Now that's interesting... I have never bothered much with getting my mount dead level, since in theory it shouldn't matter. As long as you have at least 2 alignment stars, it's possible to work out the deviation of the mount from level and adjust for it. I did it when I wrote the ASCOM driver for the SupaTrak mount, although to be honest I have no idea if Celestron do it in the handset firmware - they have rather less in the way of resources to work with. If anyone is interested, there is a rather mathematical treatment of how to calculate a two star alignment without levelling or polar aligning a scope here : http://www.asahi-net.or.jp/~zs3t-tk/aim/aim.htm . It even has some details on how to calculate and correct for mount fabrication error.

Where do you use your spirit level? The mount head itself has very little in the way of flat surfaces, so I guess you'd be using the accessory tray.

cheers,

Robin

[Macavity]

At the risk of a little pedantry, V3.07 of the Synscan Alt-Az firmware claims:

SynScan AZ-Mode Firmware (Ver.03.07) Release Note

Date: Aug. 10, 2009

===============================================================

1. Improve the initial setup of the system -- The telescope no longer needs to be level and pointed to the North as its initial position in order to accurately perform the star alignment procedure.

I did flash the EPROM (whatever) with the above - It seemed like a good idea. Another area I hope to explore (understand!) is this "PAE"(?) facility - where you re-centre the last object, and hopefully gain better "local" accuracy.

[jgs001]

Sorry Chris, I think we're possibly diverging a little... the pier looks the business.

Robin, I calibrated the accessory tray but removing the mount head, leveling across the top of the tripod till that was level, then adjusted the accessory tray till that also read level, then only used the accessory tray. Whilst in theory the alignment ought to be able to work it out, I found, and in my turn read that on here from others, that having the mount level to start with certainly helps a lot, if nothing else it minimises another opportunity for error.

[rwg]

Ok, so finally the skies cleared last night and I got outside for a couple of hours and managed to check out the goto accuracy.

I did a fairly careful 2 star alignment on Polaris & Betelgeuse (not exactingly careful, but fairly careful), then looked at Mars for a bit. After that I did a goto to the moon. The goto finished well within the moon's disc, although closer to the edge than the center, so I'd say that the inaccuracy was of the order of 10 minutes of arc.

I'm afraid my excitement at actually seeing something after so much poor weather stopped me from doing any more tests.

One data point for you, anyway

Robin

[mariaone]

ok im about three years too late on this, but, do you have any drawings on how you mounted the ALT AZ to the pier, as i am now in the process of trying to do the same thing and i keep hitting snags.

[Macavity]

ok im about three years too late on this, but, do you have any drawings on how you mounted the ALT AZ to the pier, as i am now in the process of trying to do the same thing and i keep hitting snags.

No diagram, I'm afraid - I tend to make these things up as I go along.

Although useful, this setup was eventually replaced by an HEQ5.

For what it's worth (from memory) the underside of the Synscan mount proper is retained inside the tripod "cup" by a single (M5?) screw. This engages with a threaded hole in a short bar, retained by screws either side. The basic idea is to replace this by a similar construct, capable of taking a larger M10 Skywatcher retaining screw. I suspect the best way is to make a similar thing from steel flat-bar. Drill and tap the holes appropriately. You the have to think how to mate the base of the mount with top of the Skywatcher-type head...

The first obstacle is probably the *post* that fixes "North". For an EQ3-2 / (H)EQ5 tripod, it may be easiest to simply saw it off! (It can easily be replaced by an M8 bolt). On the Skywatcher Pillar, the post is conveniently retained by a screw. With post out of the way, you can then hopefully envisage how the mount and head might be brought *flush* together? I don't have a lathe, so made "rings" / disks from thin "MDF light" (cut with a hole-saw etc.) as SPACERS. Obviously MDF won't take a threaded hole, so you have to design some scheme of holding a "captive" nut.

You then have to work out how to access the retaining screw from below - The "Owl Box" challenge! I cut a small section of 4" (1/4") Al tube so I could simply reach inside this to tighten the bolt, before assembling the whole Pier. Unfortunately, you just have to "wing it" really? An MDF prototype is useful though. In the absence of "Engineering Drawings", you can take it along to a tame CNC MACHINIST to have the bits made up from Al or Steel. Any half-decent, skilled person will appreciate the [gentle] chuckle... And, with a bit of luck, advise on: "how to do the job properly"!

[mariaone]

many thanks. i know it was a while ago but it gives me a few ideas on how to proceed. apart from chucking it out and getting a HEQ5 or 6