SynScan polar alignment bug

[Steffen]

It appears that the latest version of SynScan for EQ mounts (v.3.32) has a bug in its polar alignment procedure.

The routine seems to be assuming a northern hemisphere setup with the mount pointing north. During polar alignment the altitude ("elevation") error is being corrected nicely and can be diminished after just a few steps. The azimuth error however roughly doubles with each run of the procedure.

This can be worked around during the polar alignment routine by bringing the alignment star as close as possible to the crosshairs via the altitude adjustment screws first and then using the azimuth knobs to not bring the alignment star into the crosshairs but doubling its distance from it.

After several public betas it's disappointing that this bug made it into the release and I hope it gets squashed soon. If I can find a feedback link for SkyWatcher I'll report it to them.

Cheers
Steffen.

[whitek2h]

Thanks Steffen,
I thought something stange waas going on will check out your method tonight

Kevin

[LewisM]

I noticed it too some time back. If I did a poor alignment, leading to an average GOTO, I would run the polar realignment routine, which was FINE for alt, but made it significantly WORSE in az. Where I was close to a DSO before, after the realignment it was WAY off.

We need to get Synta to fix this.

[garymck]

Hi,

had the same error two nights ago. The error seemed to appear in the release prior to the current 3.32.

I had no issues with it working correctly 2 versions back (ie the first one with the polar alignment routine.)

cheers
Gary

[Rob P]

Thank you Steffen.

I thought I'd just have a look on the forums while I waited for the download. Saved me banging my head against the ground tonight when it all went wrong!!!

Cheers

[Steffen]

I've reported the bug via the "contact us" mailer on the Australian SkyWatcher site. I hope it gets the right attention soon.

[Steffen]

Yes, I also remember it working correctly in one of the betas.

Cheers
Steffen.

[blink138]

do synta also make the celestron cg5 gt?
my brother got the latest one and he has been having an awful lot of trouble polar aligning it
pat

[Steffen]

I've just packed up after a frustrating night of fighting with the mount. There is another new "feature" I didn't notice before.

After successful 3-star alignment (and ignoring a small polar alignment error), the mount slews to objects I tell it to go to, and upon arriving there (pretty much spot on) slews away by about half a degree, slews back to the object and goes back and forth a couple of times until it stops half a degree away from the object. What's that about? It only appeared to happen for objects in the north to north-east (around Orion and Taurus tonight).

I thought it might be some sort of backlash compensation, so when it overshoots the target it goes back and tries to come again (in order to approach from the right direction). But it should eventually make it to the target and not stop way away from it.

What is a SynScan firmware version people are happy with? I might re-flash my controller with that one to get some reliability back.

Cheers
Steffen.

[Steffen]

I've decided to go back to v.3.27. I've lost the polar alignment routine but I've gained reliability that way.

Cheers
Steffen.

[2stroke]

Yes they own celestron. Have a look at synta website :)

[pluto]

I'm on 3.30 beta and I seem to remember that they said they'd fixed the northern hemisphere issue that was present in the 3.28 beta... obviously not though as the first time I tried the polar align routine I had a similar experience to you. I gave up and went back to drift aligning.
I've not noticed the slewing issue though.

[Juanmisterio]

Hi, I'm from Argentina and I've the same problem with the v3.32. I also wrote to "contact us" and they just send me a V3.33 beta to try.
As soon as I try it, I'll inform you. Right now I'm on 3.28b15, and the polar alignment works ok (if anyone wants the file, let me now).

Agustin

Dear Agustin,

Thank you for your inquiry.

Could you please take a look at the attachment (for firmware version 3.33 beta 01) ? We had updated the firmware regarding the polar alignment in the south hemisphere issue. Could you please test the firmware again with the new attached code and see if the polar alignment function works properly? Could you please also let us know whether the result of your test?

Thank you very much again for your interest in our products and inquiry!

Kind Regards,

Sky-Watcher Support

[Steffen]

Hi Agustin, I didn't receive any reply from them but I would be keen on trying 3.33b1.

Cheers
Steffen.

[raymo]

hi everybody,
I recently bought a used HEQ5 Pro GOTO, and it is running
version 3.20. This seems to be working fine, except that some of the alignment star choices it gives me are either just above the horizon,
and hidden behind low trees, or have obscure names, and are not listed in Norton's list of 156 named stars.
raymo

[ReaPerMan]

I have just tried the Synscan 3.33B01 everything looked fine for the 2 star align. It was pointing nicely and then I tried the polar align... miaplacidus was the target of choice and it slewed to within a short distance of the star needing minor correction. When the routine ran it immediately tried to slew to 354 degrees and bury itself to the north :eyepop:. NOT the few Minutes of correction that it needed :mad2:. It would seem that 3.33 is a bust as well. I will now go back to 3.28 which works.

all the best

Paul

[LewisM]

Even though my dec issues continue, I did run 3.33B01 last night, and for a 3 star align, it offered me a LOT of stars! Nice change, compared to the few it used to offer in 3.28 etc.

Didn't try the realignment routine, for frustratingly obvious reasons.

Seeing it has all new firmware now, I thik I can rule out hand controller issues, and that leaves the motor or the board (AGAIN).

[ReaPerMan]

After reverting to the 3.28 beta, I performed a 2 star alignment successfully and then went to the polar align. This time the error was smaller than stated with the 3.33b? After adjusting I placed it back to home and then did another 2 star align. I expected it to nail Sirius in the cross hairs but for some reason I had to make adjustments. After the align I once again performed the polar align and instead of minor moves I had to make a significant az alt align?? I spent some time getting it right and watching for drift. I eventually was happy and confirmed the adjustment. I put it back to home and asked it to goto a star and it missed again? Apart from the strange goto problem the polar align seems solid and whilst imaging Jupiter it stayed near the center of a 640 x480 pic on a 1/3" chip for more than 10 minutes with no guiding at all just the mount tracking. Not perfect but then it is a beta!

I hope skywatcher can resolve these problems as a fully working version of syncscan would be fantastic.

All the best

Paul :thumbsup:

[ReaPerMan]

First time use in guiding! No other polar alignment used than the sysnscan from 3.28b. 4 hours of guiding with a near perfect result. I did not even do any drift alignment or anything.

[meade expert]

Anyone find 3.33B01 firmware any good?:confused2:

[LewisM]

Nope, it's carp too.

Darned thing gave me an 11° error to the east after running the thing. I figure maybe if I run it again, and correct it 11° WEST instead, it may bring it back onto target LOL.

I think the Chinese winged the wong numbers.

[ReaPerMan]

The 3.33b has the same problems as 3.32. Whilst the Star selection etc is better than 3.28b The prime function of Polar adjustment is still screwed up. I have no idea why they can get it right in an earlier version and then throw out what works and replace it with this piece of Poo!

The only other thing I have noticed is that (as stated in my previous posts) the handpiece reports odd things even after you do the polar align when you rinse and repeat it then gives you a whole lot of other error numbers. I would have thought that the numbers would get progressively smaller until 'perfection' but even after a 3 star align and the 3rd iteration It was telling me i was something like 17 minutes off in Alt (but only 3 secs off in Az). After correcting this making sure that the star was centered in my illuminated eyepiece I watched the star for 5 minutes and it remained rock steady in the center of the crosshairs. I asked the handpiece to do a Goto and it missed by miles. I then disconnected the handpiece and connected the mount to the computer. When I fired up Maxim's observatory and asked it to go to Acrux It was right there almost dead on the crosshairs in the camera picture. I adjusted and synced acrux's position in the observatory software and asked it to go to another target and it nailed it again. In fact every target was spot on! These were my best goto results ever, but notabley not from the handpiece. I'm at a loss for these inconsistancies, I plan to use the handpiece to do a 3 star align, Polar adjust, 3 star align, Polar adjust and then send it to home and use my computer software and EQMod from then on. If I do this I'll cut down my present setup times by half. I can only hope skywatcher get it right eventually as if it all worked as advertised it would be terrific.

regards

Paul

[cosmophoton]

Hi Paul and LewisM,

I have noted the same thing with 3.33B01 firmware (thanks Agustin for the download). The alignment seems good, but the goto function is not working at all. Hope that they fix that soon!

Clear Skies,
Luiz

[Steffen]

I wonder whether they're even aware. I don't know how much feedback they're getting from the southern hemisphere. I encourage everybody to report these bugs.

For want of a better channel I've used the "contact us" form on the skywatcher.com.au website.

Cheers
Steffen.

[RobC]

The following is an evaluation of the polar alignment carried out by a friend of mine in December 2012. This was fowarded to Skywatcher through one of their dealers but as yet we have recieved no response from Skywatcher

" Last night I had a chance to test the Final release version of the HEQ5 firmware Ver3.32.

The new firmware is now designed to work with AZ and EQ mounts and will auto detect
what mount that it has been connected to, which I guess could be a good thing, maybe..

Here is my evaluation it might as well be a beta version, as once again the polar alignment
procedure dose not work. which is consistent with all previous version other then Ver 3.28b21
which was the first version to have it implemented, and works just great although not 100 percent
it is at least 90 percent accurate.

My alignment here is so close that I only needed to do the re-Polar Alignment procedure once
with minor adjustments, so I slewed the scope to Betelgeuse which at this stage was pretty high up
and not far from the Meridian then come back in an hour to check the drift and it hadn't even moved,
that is, no guiding.

So I'm sure there are lots of issues still with the firmware, although I didn't check the Sidereal rate problem,
at a guess I'm sure the issues with that hasn't been resolved."


Rob

[AdamH]

I've just upgraded from 3.27 to 3.32 primarily to use this feature. My experience was that the elevation and azimuth errors doubled after each iteration of Polar alignment adjustment. I'm glad to discover that the problem wasn't solely my incompetence.

[JQ]

I purchased a Skywatcher 100ed pro with an HEQ5 mount. I could not believe I could align the scope so easily on the second try. Then I upgraded to 3.32.

I emailed Skywatcher and received a very prompt response - next business day. They said they are aware of the problem and that it's Southern Hemisphere specific. They also said they are working "flat out" to fix the bug.

It can't come soon enough. Back to 3.27.

Regards and thanks for the great community support.:)

John

[HunterGeo]

I'm also having this issue (v3.32 firmware) as well as the issue where the P.A. makes things worse rather than better. :(

[Stardrifter_WA]

What I find difficult to accept is that Skywatcher allegedly knows about the issues with the 3.32 firmware, whilst still allowing their product to be sold. :question: One shouldn't have to wind back to an earlier version of software to get a new mount to work properly. This is just wrong :sadeyes:

It has certainly put me off buying their new EQ6 Az/Equat. mount. But, what are the alternatives; the Meade LX80? I don't think so. :mad2:

[MrB]

Yes it is very confusing. I don't understand how it has been so long now with no fix when I have read reports of the PA working in earlier versions.
Get one coder working on it fulltime and the issue would be solved in no time. Their priorities obviously lay elsewhere, like getting the EQ8 to market.
I've not gone past v3.27, and not likely to anytime soon.
Maybe they think we Southern Hemisphere customers are not worth the effort? :lol:

[Stardrifter_WA]

Maybe the ACCC could do something about it, since the product clearly doesn't conform to advertised statements. :mad2: Maybe they should pay all owners having trouble compensation, since they appear to know the problem exists.

[AdamH]

I contacted SkyWatcher Australia here http://www.skywatcheraustralia.com.au/?page_id=1767

On my first contact at the end of January I was told:
"Thank you for your email. The factory R&D team are currently working on it."
(That's the reply in full)

I've recontacted them today to see if they have any dates, and will provide an update here if they get back to me.

I'd encourage everyone to politely contact them asking for information.

[Stardrifter_WA]

I did, and got no info :sadeyes:

[Astroman]

I am hopefully getting a mount soon with v3.20 does this have the polar alignment feature on it or would I need to upgrade? Not sure I wan to if there are known issues with the firmware.

[Visionoz]

That site belongs to Tasco the Oz local distributors for Skywatcher - perhaps it is not getting through to the "right channels" - may be better to contact the original Skywatcher HQ - unfortunately there isn't any email contact address on their site as their "Contact Us" & Customer Support page points to all their distributors world-wide

However I have gleaned this email address which should be received by someone in their company: administrator@skywatchertelescope.n et

HTH
Cheers
Bill

[AdamH]

There's a new version of the firmware, but it doesn't fix the PA issue.
http://www.iceinspace.com.au/forum/showthread.php?p=946322

[raymo]

Hi Andrew,
My v3.20 does not have the P.A. facility, and, like you, I am wary of upgrading. I have several issues with it already, and don't want
to add more.
regards raymo:(

[2stroke]

looks like they fixed it in v3.37 LMAO ;)

edit that's v3.34 other wise that sarcastic joke doesn't apply rofl

[seeker372011]

Is there any update on the PA routine with Synscan that anyone knows of?

Hate wasting clear nights drift aligning

The v 3.35 is supposed to have an improved routine..does it work for the Southern Hemisphere?

[LewisM]

The PA feature of 3.35 when I tried it simply refused to work. It tells you to centre a star, I do, and hit enter, and it goes back to the Setup page all over again. In other words, it doesn't do ANYTHING.

I just drift align. Less frustrating.

[Steffen]

The polar alignment routine in 3.35 works well, even in the southern hemisphere. You get pretty good alignment very fast. Not sure what Lewis is doing… ;)

Cheers
Steffen.

[LewisM]

Well I never! ;)

For me, when I run PAE, it asks to centre on a star, say Achernar. Fine, off it goes. I centre, and press enter, and all it does it revert to the first screen of the PAE. In other words, it does NOTHING.

I have run the pAE before with 3.33 etc, so know what I am doing, but for some reason, it just is not working on mine. So, I just do the "Lewis Drift Align" - I use 3.28B21. I do a 2 star align, and I observe the MEL/MAZ readout after a successful alignment. I then adjust the mount "by eye" in az and el, and redo the alignment using the same stars. I then do this for several iterations until I am under at least the minute of error (0*00'XX") for both axes. I then check it again as a goto assignment, and if it nails it (which it usually does), I will then slew to a globular or similar and see if it nails that.

I will then run the mount in sidereal for 20 minutes with PHD "live view" on and watch in the autoguider. If the star does not stray (NO GUIDING, just watching with gridlines on), I will then start my imaging for the night :)

[Astro_Bot]

Firstly, I haven't used the SynScan polar alignment feature yet, so bear with me, but ...

PAE is Pointing Accuracy Enhancement, and it sounds like it's behaving correctly from Lewis's description.

Polar Align is a different feature altogether - it's accessed from the Setup menu, whereas PAE is accessed from the Utility Func. menu.

Could it be as simple as that?

[seeker372011]

B****r. Why, when it works well on Celestron, can they not get to work on the EQ6, I can not understand.

[seeker372011]

Oh ....

[Steffen]

Sounds like it is, just a little PA vs PAE confusion. PAE does indeed work like that. And yes, it works well, too, although I haven't found exactly how big around the PAE star the effective radius is. All the manual tells is that there are some 80+ PAE regions.

PAE has bugs, but they don't seem to interfere with its function when ignored. For some objects, after calling up the PAE function, it doesn't show the object's name. For some other objects (esp some SAO stars) it shows the name of an IC object (always the same IC number, forgot which one).

Cheers
Steffen.

[LewisM]

No wonder it didn't work...:screwy:

I tells ya, I am getting stupider as I get older. At least my drift alignment works ;)

[chiaroscuro]

Should you routinely do this PAE after doing a rough polar alignment and a 1-3 star alignment? (I need to change my username to "Confused beginner".)
Does the software constantly correct alignment errors as you uses it - i.e.. if you dial up an object, and then need to adjust to centre it, does the software integrate this correction for each subsequent object? That would seem like a sensible feature.
I've had a bit of a whinge elsewhere about Synscan firmware being not fit for purpose. The ACCC would be interested in any complaint about a product being sold which cannot perform the function which it is designed for. I'm going to email skywatcher about it.

[Steffen]

No, PAE should not be routinely required.

The pointing accuracy also doesn't depend on how well polar aligned the mount is, only on how accurate the 2- or 3-star alignment was. After doing just a 2-star alignment it also depends on the mount/scope's inherent cone error (the whole gamut of geometric errors hiding under that term).

If you take care with 2- or 3-star alignment (use a reticle eyepiece, be as accurate as you can, enter the time correctly to the second, enter your location as accurately as it lets you) then subsequent GOTOs should be good.

I find PAE useful (and used it sometimes) when hunting for objects that are on the verge of visibility. I want to be sure the object is in the centre of the field before straining my eyes, trying eyepieces, filters etc. So I do a PAE step on a nearby star before going to the faint object.

The hand controller cannot automatically adjust the pointing accuracy after a GOTO and manual correction. It cannot know whether you're centring the object after the GOTO, or just panning around.

Cheers
Steffen.

[chiaroscuro]

If I was smart, and the synscan software developer, I'd put in an option for this. It seems obvious to me that the accuracy of the "map" in the algorithm would be incrementally improved with each GoTo and centring.

But Synscan needs to learn how to crawl before it can walk.:rolleyes:

[LewisM]

Why these days with GPS accuracy so refined that a mount developer cannot automate polar alignment based on GPS coordinates. A GPS will tell you the lat and long, and ergo by the lat and long what the magnetic deviation is to correct for true South/celestial south (but a GPS should be reading true south anyway)

I am sure they will do it eventually, but I cannot see the hold up unless - as I suspect - I know absolutely nothing about this subject (considering for me PAE was PA... d'oh!)

[alocky]

Unfortunately at the level of precision we are looking for in polar alignment, the geomagnetic field varies too much in time. As well as an accurate map of the local geomagnetic field - available from your friendly government geoscience agency, your computer would also need the IGRF model stored internally. Not so trivial. To get this kind of pointing accuracy also requires a very expensive vector magnetometer.
What's wrong with eq align?
Cheers,
Andrew.

[Steffen]

I suppose you'd call them stupid, too, if they messed up pointing accuracy every time you slewed to M65, then panned a bit to fit in the whole triplet ;) That's not to say future SynScan version won't have mind reading abilities :D Until then we have to live with a few extra button pushes.

Cheers
Steffen.