Ultimate accuracy of HEQ5 (EQMod) GoTo?

[Macavity]

With my HEQ5 (EQMod), I seem to be consistently missing GoTo targets by ~0.5-1.0 Deg? THAT said, my polar alignment is VERY "ad-hoc" at the moment - Merely "somewhere in the circle"? I sense I definitely need to work on that one...

BUT, I am wondering: With accurate Polar Alignment, I *should* be able to achieve GoTo object centering within a "degree or so"? Half a degree, even better! It is possible with an HEQ5, right?

Aside: Ultimately, I aspire to just "press buttons" (from a warm room!) and DSOs will be (approximately) centered on a 1/2" Video chip / LCD screen, with my 200mm, F4 Photo-Newt.

[Merlin66]

The HEQ5pro and the NEQ6pro can provide better GOTO positioning than that.

I use AstroPlanner to find some obscure stars (Wolf Rayet, Be emission stars etc) and then use the GOTO to locate them.

There's no problem "dropping" them into the FOV of my beamsplitter/ QHY guider.

I only use the Handcontroller connection to CdC...

[ollypenrice]

Yes, it should do what you ask once you have good polar alignment - which is not difficult. In a permanent setup just do a decent drift align.

Although three star alignment should allow for it, cone error is a factor to consider. This refers to the OTA not being precisliely aligned with the RA axis.

Olly

[George]

EQmod providing you have good polar alignment and you do a 3 star N alignment in the area of your target usally puts it in the FOV of a QHY5 guider.....for me in anycase.

[Macavity]

Yes, it should do what you ask once you have good polar alignment - which is not difficult. In a permanent setup just do a decent drift align.

I think I am increasingly persuaded that drift alignment is the way to go.

Although three star alignment should allow for it, cone error is a factor to consider. This refers to the OTA not being precisliely aligned with the RA axis.

Another thing I HAD been wondering about too! As has been implied elsewhere, TS/GSO

F4 Newts are a bit "stick and string"? <G> We try our best, I guess... Thanks, Olly.

[Macavity]

Thanks ALL for providing a bit of (aspirational) encouragement!

Was getting a bit "fed up". But as long as things are possible...

[beamish]

Drift aligning will dramatically improve accuracy of the HEQ5

[chrisshillito]

Once goto alignment has been performed any polar alignment errors are automatically modeled for, and compensated for, by the goto system. Whilst good Polar alignment is certainly desirable form the point of view of tracking accuracy and field rotation it shouldn't have any effect on a "gotos" (if it does then your current alignment model needs more points!)

Chris.

[Donkeiller]

0.5 to 1.0 degree off?! That's dreadful.

One thing I have learnt is that getting the setup correct

Scope balanced

mount level

stars centred.

correct time, date and GPS position.

I have 2 AZ (one very cheap!) mounts, so I don't need to Polar align. That said all Goto mounts need to be setup correctly. I will regularly get 0.2 degree accuracy and usually alot better.

This will always put the object in the field of view- very often right at the centre.

[Macavity]

Once goto alignment has been performed any polar alignment errors are automatically modeled for, and compensated for, by the goto system. Whilst good Polar alignment is certainly desirable form the point of view of tracking accuracy and field rotation it shouldn't have any effect on a "gotos" (if it does then your current alignment model needs more points!)

That's what I believe(d). My infinite commendation re. EQMOD anyway. I do sense (intelligent, wider etc.) point selection might be (at least) part of it. Maybe USING a reticle eyepiece, rather than just guessing...

I just wanted confirmation of what I might expect. I like to think I'm not completely dumb (LOL), but cannot rule out "operator error"! Truth be told, I might be trying to run before walking. A little planning and forethought re. the actual OBSERVING might help, now I have the modest observatory (essentially) complete.

[MikeWilson]

With my HEQ5 (EQMod), I seem to be consistently missing GoTo targets by ~0.5-1.0 Deg? THAT said, my polar alignment is VERY "ad-hoc" at the moment - Merely "somewhere in the circle"? I sense I definitely need to work on that one...

Quick tip: Get Polaris on the circle and then spin the mount around in RA. Polaris should sit tightly on the circle as the mount spins around.

If it does, you're pretty much polar aligned. If it doesn't, adjust your polar alignment and try again.

All the best,

Mike

[chrisshillito]

That's what I believe(d). My infinite commendation re. EQMOD anyway. I do sense (intelligent, wider etc.) point selection might be (at least) part of it. Maybe USING a reticle eyepiece, rather than just guessing...

Yes, use a reticule eyepiece if you have one. If your using "N-Point + nearest" and the pointing isn't great then you might try switching over to "nearest" and see if things improve. The problem with 3-point transformations is that a "bad point" can have an effect in an area you might not suspect (You can use the the points editor to see which alignment points are active).

Also you might consider using the point filter to ensure that only those points on the same side of the meridian as your target are considered. If you have an observatory based mount then I'd go for a 12 point model, with a triangle of 3 in each quadrant and set the point filter to use "local quadrant".

If setting up each night I'd just sync on stars local to my targets rather than spend time building an all sky model up front.

Chris.

[Macavity]

...The problem with 3-point transformations is that a "bad point" can have an effect in an area you might not suspect ...

Thanks, Chris. I think that the "bad point" explanation is most likely! I'm actually using the HEQ5 on top of Skywatcher PILLAR - Albeit, jacked up, on three fairly solid concrete brick plinths in the observatory.

Of course, without a (homebrew) "pillar top" there is no easy access to the mount securing KNOB, which I had *loosened* slightly to allow for the polar alignment. I had hoped that the azimuth setting screws and "spigot" would hold the mount in position, but I sense (as some point) a small rotation might have occurred...

MEA CULPA (Partly? Mostly?). I might wriggle inasmuch that so-often it seems you need to add "custom machined" products to get a typical working system. I do know some tame CAM/CAD machinists, but they are busy and like cobblers tend to forget one-off jobs until you arrive to collect.

On the positives. Thanks ALL for encouragement and advice on Polar Alignment. As a sometime software developer, I do have faith in EQMOD's (Chris'!) ability to cope with polar misalignments. <G> The "point map" may have been instructive? Many of synched "new" points showed a somewhat regular, systematic, shift between predicted and measure points - Suggesting something had moved physically?

I HOPE noone feels time has been wasted on this... As ever, a few interesting FACTS emerge, to become anecdotes that others might benefit from too? A more complete understanding rarely goes amiss either:

EQASCOM Alignment Part 4: Proximity Range & Point Filter

Hmmm... A lot of my "triangles" WERE quite acute too!

P.S. As ever, I'm sure there is a READY market for Skywatcher pillar tops.

[SlipperySquid]

Hiya mate,

I to use HEQ5 and EQMOD and find more often then not my first star is way way out, even with what I think is a pretty accurate polar alignment.

Though once I have found the star I originally wanted the mount to point to and centered it, all other stars/objects are close to the center of the field of view and once a three star has been completed it's more or less bang on central.

I so need an obsy

[Macavity]

Though once I have found the star I originally wanted the mount to point to and centered it, all other stars/objects are close to the center of the field of view and once a three star has been completed it's more or less bang on central.

Heheh. Well, I think, next time, I'll FIRMLY CLAMP the blooming thing, irrespective of the Polar Alignment.

I suspect the tracking is (demonstrably) OK-enough for Video Astronomy. I do have faith in the software.

[George]

Once I have perfect polar alignment I'll spend 10 minutes or so creating triangles in the areas I'll be imaging. What I found was first time you generate a star list is the first star will be off and what I did was slacken the clutches and centre the star then tighten up, all GOTOS after that were pretty much bang on.

This was proberbly down to cone error.

[chrisshillito]

Hiya mate,

I to use HEQ5 and EQMOD and find more often then not my first star is way way out, even with what I think is a pretty accurate polar alignment.

An accurate initial goto (when there is no correction applied at all) would require not only accurate polar alignment but also

accurate time

accurate location

perfectly level

power up in the perfect home position

Scope mounted perfectly orthogonal to the mount

Perfectly aligned optical system.

No atmospheric distortions

Of course if you have all that you needn't to bother with an alignment model

Many folks are happy to zero the time/location and home position errors by slackening the clutches and manually centering the first alignment point (or getting as close as they can before fine centering with a gamepad/keyboard//mouse). The alignment model will cope with these errors but by reducing them in this way any subsequent pointing corrections will be smaller which is no bad thing.

Chris.

[MikeWilson]

Many folks are happy to zero the time/location and home position errors by slackening the clutches and manually centering the first alignment point (or getting as close as they can before fine centering with a gamepad/keyboard//mouse). The alignment model will cope with these errors but by reducing them in this way any subsequent pointing corrections will be smaller which is no bad thing.

Chris.

TOP TIP! Thanks Chris :-)

Mike

[Macavity]

Just discovered another potential facet to this problem - A significant cone error, due to non-orthogonality of the tube rings and the dovetail bar. Although blaming myself for movement in between the pillar and mount seemed initially "attractive" (LOL), I'm more in favour of "cone errors" now. I suspect theses are indeed corrected by EQMOD, within say a triangle of alignment points, but might not be so well controlled when the GoTo's depend on nearest stars. Clearly more (well chosen) points might have helped though.

A rather obvious pointer (maybe?) was via today's daylight white-light solar observation. Having Synched and tracked the sun successfully, during the morning, I "parked" the scope, then tried a GoTo the Sun AFTER local solar noon. On the other side of the meridian, it was quite fair a bit out! Something to work on, tho'? But perhaps a can o' worms, I never envisaged...

[MikeWilson]

On the other side of the meridian, it was quite fair a bit out! Something to work on, tho'? But perhaps a can o' worms, I never envisaged...

If you find a solution to this, I'd love to know what it is.

The blasted meridian line drives me nuts. If I align on one side of it, alignment is always way out on the other side. So I tend to stick to targets on one side of the Meridian or the other.

Making an all sky map (i.e. aligning on both sides) just makes things worse.

All the best,

Mike

[Revs]

Can't you just correct it with the 4 dovetail screws? I've been meaning to do this on mine.

[chrisshillito]

The blasted meridian line drives me nuts. If I align on one side of it, alignment is always way out on the other side. So I tend to stick to targets on one side of the Meridian or the other.

Making an all sky map (i.e. aligning on both sides) just makes things worse.

Are you using EQASCOM?

If so you could create an all sky map (say three on each side of the meridian) and then set the points filter to only consider those points form the model that are on the same side of the meridian as your target.

Chris.

[MikeWilson]

Are you using EQASCOM?

If so you could create an all sky map (say three on each side of the meridian) and then set the points filter to only consider those points form the model that are on the same side of the meridian as your target.

Chris.

I can do that?

Are you the chap who did all those EQMOD tutorial videos? If so, I've seen a couple and they're very helpful. I'll have to take a look at a few more to see if I can find this feature.

All the best,

Mike

[Macavity]

Far greater success last night. Perpendicularly aligning the tube rings with the dovetail (by set-square) seems to have helped quite a lot. Aligning on a (small) "Summer Triangle" (East of the meridian) found Albireo nicely centered via the first GoTo. Moreover, THIS time, slewing to Arcturus (West of meridian) was much more accurate. With one or two "anomalies" (I always have to synch on a "keystone" star re. M82 for some reason!) the familiar DSOs were (albeit rather faintly) in evidence. Clearly my MAK150 (f=1800mm) is not the ideal scope for rooting out (initial!) alignment errors, but (as the sky model built up?) I was beginning to hit GoTo targets within the (~0.25 deg) field of a 10mm reticule eyepiece. ALL this without ever aligning on (the real) Polaris too.

As noted above (elsewhere!), Dovetail Bars have extra holes for aligning tube rings... Sadly none of my (Skywatcher or GSO) combos match, position-wise? BUT at least the idea of "roaming the skies" with an (f=800) GSO photo Newt and a Watec (1/2") Video chip (TFoV <=1/2 deg) begins to look feasible now!

Aside: I'm really starting to *like* EQMOD, though, in fairness, it (realistically) takes up all the resources (USB sockets, notably) of my humble NCxx Netbook. I now need another one to run any (Video) capturing! Also, remote, ELECTRIC focussing seems almost an essential for this idea... Never blimmin' ends, does it?

[chrisshillito]

I can do that?

Are you the chap who did all those EQMOD tutorial videos? If so, I've seen a couple and they're very helpful. I'll have to take a look at a few more to see if I can find this feature.

All the best,

Mike

Yes, you can do that - you can even filter on local quadrant if you like as well (but then your going to need 12 alignment points, three in each quadrant, which is getting to be a bit of a chore unless your mount is permanently sited.

Yes I'm the guy boring everyone to death with EQMOD videos - I also write the majority of the software nowadays. The video that mentions the points filter is here YouTube - ‪EQASCOM Alignment Part 4: Proximity Range & Point Filter‬‏

Chris.