Topics

New EXOS-2/PMC-8, need advice on tracking


r_hoskin@...
 

Hi!

 

Can I get some help understanding my tracking errors, please?   Sorry if this is a bit long…


Last night, the clouds finally parted for a while, and I was able to take scope and mount out for a test run.  My sole objective was to figure out an acceptable unguided exposure length for future work.  I wanted to know the point at which tracking errors accumulated and my round stars turned into potatoes.


Potatoes happened a lot earlier than I expected it to, but I gathered enough data to see that there isn’t just one thing going on – there is clearly more than one component to the overall tracking error.  I not only have potatoes, I have bidirectional seagulls, and sine waves as well!  I’m looking for good advice on what to do about that.


Equipment and software:

EXOS-2/PMC-8 on 110v, SV 102EDT @ f6.1, Nikon D5300 with t-adapter and a 110v power adapter, Live View focus, Polaroid intervalometer, Parallax USB-to-Serial adapter, Lenovo x230 laptop on battery, Win7 Pro, Ascom w/Explore’s driver, Cartes du Ciel.  Latitude and longitude set manually with values from google earth.


Process:

Did a rough (Polaris near eyepiece crosshairs) polar alignment.  Did a two-star alignment with CdC.  Slewed to M42 and started the intervalometer on its first set of exposures.  While I did not intend to touch the scope while imaging, I found that at 32F/0C, it was too cold for the intervalometer and its display was unresponsive.  I had to unplug it from the camera and take it in the house between shots to reset it.  So there may be some image shift from that.  One bit of serendipity from this is that I ended up accidentally triggering a short 10-second shot that’s one of the nicer ones in the collection.

 

A selection of compressed jpegs of my images are in the attached zip.  You’ll have to zoom waaay in, to see some of the effects.  I’ve concentrated on the line of 3 stars with CA, at 8 o’clock from the nebula, and the smaller stars around them.


DSC_0184_004 is the 10-second image.  If I zoom straight in the three stars are a bit rough, but I’d guess that’s the effects of overexposure and seeing.  The smaller, surrounding stars seem sharp.


DSC_177_005 is a 30-second image.  Potato shapes starting to develop, and the small surrounding stars are becoming oblong.


DSC_185_007 and DSC_186_008 are both 60-second images, and the smaller stars are now seagulls.  In DSC_185_007, they fly left, but in DSC_186_008, they fly right!


At this point, clouds are moving in and starting to obscure the shots.


DSC_0187_009 is a 120-second shot.  The jpeg’s lost detail from the original, but if you look closely, you see that some seagulls are now becoming “W”’s.


DSC_0189_010 is a 192-second shot.  Clouds are almost completely obscuring it, but if you drill in, you see that the seagulls are now sine waves.

 

Thanks for your help with this!



Wes Mcdonald
 

Hi:

I have  no experience using the PMC-8 and EXos II for astrophotography yet, as I am still messing around with my mount/new telescope.  But I have some with other scopes and thus have a few thoughts...free advice, right?

1.  It is important to perfect your polar alignment if you plan to do exposures of any length.  The PMC-8 alignment does not correct polar errors for astrophotography, rather it only sets the mount up to find things in the sky -- objects being tracked should be good to go in the center over time, but the object will rotate in the focal plane of the camera owing to declination and RA error from pure polar.  Thus this will cause the stars to streak or arc on you.  A short exposure, 10 seconds, at F6.1 probably would not show much, even with an alt/az mount.  Longer exposures might, depending upon your errors.  I have no sense of scale for your star smears, can can't say how much polar error contributed, but I am certain it does for the longer exposure times.

2.  I found my mount has a bit of gear backlash after I adjusted the axes to prevent binding- -- there is a bit of eccentricity to the worm and wheel placement, and they are hard-mounted.  Thus one has to release the mesh enough so that over the entire wheel rotation the meshing of the gears is never too tight.  I found this to be a problem with my mount, discussed it with Jerry, and made the adjustments.  But I have some areas where the resulting mesh allows more backlash than in other areas.  With unguided photos however, backlash is not necessarily a problem, if you bias the balance of the mount such that as the thing tracks the object from east to west is drives against tight gears, ie bias the thing to "fall" east in RA and up or down in dec depending upon if the object is rising or setting.  I guess if it is setting I want dec balanced so it falls down to keep lash out, but you should think about that.  So backlash can mess up things.  BUT, if polar is very good, there will be minimal dec driving anyhow, so static balance with bias is what is needed.

3.  If there are mechanical disturbances the picture can get those stars.  Wind for example, or as you noted taking off equipment or whatever.

If I had these results I would presume polar alignment, backlash and disturbance.  You can address each of these next time.  I bet things will get better.

Understand I can easily be full of it.

Wes.



r_hoskin@...
 

Thanks, Wes!

No, I get the polar alignment issue.  I'm quite ready to think that it could well be a component of what I'm seeing, but what I don't have is a sense of the degree to which it's affecting those images.  I have research to do on that.  I think I have more than one thing going on, and if some of it is (say) mechanical or a matter of CdC settings, I'd like to clear those so I can get on with the skill-building stuff.

I do wonder about the mechanical side.  This is belt-drive, which is supposed to be better and have less backlash in the overall drive train.  What I have not done is take the mount apart and lube it with something winter-suitable.  (Is sticktion a problem? No idea).  I'm using it out-of-the-box.  I thought to start that and then held back when I saw those belts.  I don't think it would be like pulling off a geared motor.  I think I want some process instruction from Explore before I open it up (and maybe some spare belts).  They're not large, and I do not want to break or stretch one.

There wasn't much wind, and traffic should have been quiet, but I suppose there's alway the influence of me stomping around.  I could add a set of vibration pads, just not sure how they work out in the cold. 


Robert Huerbsch <bobhuerbsch@...>
 

Unguided = you need near perfect PA or as good as possible...

It should be doable at your 600 focal length but use a 0.8x reducer/flatter if you have one. The shorter the focal length and better your PA, the rounder starts and longer subs you will get. I got sub 1’ PA via careful PHD2 drift align, but now use a QHY Polemaster to save tons of time (and even more accurate). Id guess you could get 30-60 seconds unguided depending on the periodic error in your mount and how good your PA is.

AP can be frustrating... an autoguider and great PA makes it much less hassle!


Wes Mcdonald
 

Hi R

I had to take my belts off etc to make the adjustments I made.  The belts connect the stepper motor to the worm.  The backlash exists between the worm and the wheel (this is the big gear actually on the axis, driven by the worm).  I once thought the belt would mean less backlash, but not now.  It's irrelevant, except for the influence a flex coupling might have had in a conventional setup.  

Mount disturbance is what happened of course, we just got to figure out what.  One of these days soon I will get a look at some photos on my EXOSII/PMC8.  I can tell you though my telescope, loaded with all my visual stuff, has a long damping time for a vibration induced by a pulse input (at least one second).  I was thinking of measuring it, but got to come up with an accelerometer from somewhere, and don't want to buy one yet.   Now my scope weighs in at 21.6# without the extra drawtube extensions and eyepiece I have in it.  It is an Astrotech AT8IN, f/4 8" scope, so about 32 inches long.  Basically, if one were to cause it to bounce, say from a gear backlash, it might vibrate.  I wouldn't expect this to happen much or at all if you put in a little tracking weight bias as I mentioned in my previous note.

Field rotation in an AZ/EL mount will be seen in 45 seconds I believe at f/6 or so.  Seems like that was where I had to stop using my MEade LX200 in AZEL mode at f/6.3.  But I have to tell you I usually stopped at 30 seconds.  I guess field rotation from incorrect polar align might be less over the same time assuming a close alignment.  But your stars don't really look like rotation to me , rather the result of disturbance causing the image to be dragged across the focal array (ok, maybe a bit of rotation).

This stuff can drive you crazy.  A friend of mine has an Astro Physics mount and big refractor scope with an astronomy modified Cannon.  He can do some unbelievable unguided images, and now guides which takes things to a new level.  My point is he has $20K invested, not $800.  I don't think we can expect a lot from our little mounts.  I want great goto, and snapshot images to declare success, which I am confident I will get.  Another friend of mine uses a G11 with gemini controller and I would pay good money for his results.  So perhaps once you learn all you can from the EXOSII, it will be time to move up to a G11 plus PMC8.  I think that will be my next stop, but I am far from ready to do that...so much still to learn and enjoy with my LX200 and EXOSII.

Wes.

 


r_hoskin@...
 

Hi, Wes!

Ah, well... the advice (and results) do tend in the direction of bigger and heavier mounts, but I decided that I wanted to stay in the 'luggable' zone.  Not gettin' any younger, and I don't see a spot in the back yard for a dome :-).  If I have to take a bunch of shorter images, so be it. Besides, if I stay luggable, perhaps I'll be able to do some imaging somewhere other than at home!  

Seriously, though, one of the things that struck me (and helped hook me) on the whole idea of AP is the range of gear that people were able to use to produce great images.  It's wide.  The more I looked, the more it appeared that while top-end kit is an enormous help, the real differentiator seemed to be patience in data capture and skill in processing.

For a (somewhat) jarring experience, take a look at this guy's work:


Then check the notes.  f7 scope, alt/az mount, often 30 sec lights...
The guy's almost  at 90 degrees from accepted wisdom, and he gets amazing results. 
Picked up a book in the Moore Series (Ashley) about grab-and-go AP, and he's not a blip.  

So I have hope that I will be able to both learn a bunch, and turn out some wall-worthy pix with my little EXOS-2/PMC-8.  We'll see how it goes, but I expect that it will work out.  

- Bob


Wes Mcdonald
 

Hi Bob:

It is true, stacking lots of 30-40 second exposures can do wonderful things.  You won't match the deep exposures of the big rigs, but good results can be had -- I know because I got a pretty good Swan with almost no experience but 20 lights at 30 seconds.  My biggest problem then was focus.  Now this was an LX200, but far from optimized.

I, like you, have some expectations for the EXOS!/PMC8.  Also, I like the challenge of trying to do more with less.  So my plan is to :

1.  Obtain good polar alignment with my polemaster
2.  Work to balance the scope as closely as possible to perfect --this is a challenge as the mount is inherently unbalanced owing to the side mounted motor assembly
3.  Isolate the mount from vibrations (isolators on the feet probably, and don't dance around the mount
4.  Implement guiding
5.  Achieve perfect focus with Bahtinov mask
5.  Work camera type and processing.  

Good stuff will happen eventually.

Good luck, let us all know what you find out about your initial troubles.

Regards,
Wes.




Robert Huerbsch <bobhuerbsch@...>
 

You should be able to get 3-5 min subs with the EXOS, probably even longer with a small refractor. I was able to get 3 min subs (EXOS2-GT not PCM8) with a 600fl imaging newt @ 20lb loaded but it was not easy. A small 60-80mm refractor should be perfect for the EXOS. I think the PCM8 could end up being an AVX killer... the DEC and RA was much more smooth and its belt driven for less backlash.

This was shot last night with a 71mm refractor @ 336 FL. There are tons of large DSO to use at shorter focal length which makes AP much easier...


r_hoskin@...
 

Robert,

Thank you for that wonderful image!  Very nice!

Yes, I think it will be ok, just some things going on that need to be understood and sorted out, and folks here have been very helpful with that. 

I'd be delighted to have 3-5 minute subs within reach. Even with small refractors, that's enough capability to keep me out of trouble for a long time. :-)

I had hoped to get out tonight to take another run at it, but the weather flipped on me and I'm clouded in again. 

I've identified two possible sources of focuser/camera motion, and will be doing a better PA next time. We'll see if between those, things settle down a bit.  



Robert Hoskin <r_hoskin@...>
 

Well, a little progress on my seagull issue...

We finally had a bit of clear evening sky (just a bit...).  I was able to get David's drift test done.  Two 5-minute images, no motors, scope repositioned in-between, just to see what the tracks look like.  I thought I had set the camera down to ISO 200 earlier, but evidently not, so these are still at 800.

Then the clouds rolled in as I was trying to get my PA done, and that was that for tonight. 

DSC_0190_013.jpg clearly had *something* cause a jitter mid-track, but it settled down quickly.  I was in the house for the duration, but there's a train track nearby that runs the occasional evening train.  Could have been that, I suppose.

DSC_0191_014.jpg seems like it has fairly smooth tracks.

These don't look bad to me - what do you folks think?



r_hoskin@...
 

Looks like Yahoo isn't taking attachments tonight, either as a forum post or email.
I'll try uploading my .zip again later.


r_hoskin@...
 

Attachment uploads are working again!  :-)


David Pickett
 

Robert,

Some progress, I see. Did you try perturbing the mount (gently!) while the shutter was open. Are you seriously thinking that it might have been the train? If so, repeat the experiment!

Was this just camera on mount, or camera connected to OTA via focusser?

No.15 shows that the camera actually moved (more than once) during the exposure -- pretty quickly across the gaps in the traces. This is either the camera or focusser moving sideways, or some other movement about the RA axis. Was it perhaps caused by backlash in the motor/gears, or were they all locked up solid? The lump in the trace is odd, indicating that the star and camera moved together for a short while.

Of course, if the camera/focusser has moved under the influence of gravity, it will generally stay there.

I think you need to do some more exposures along the same lines (pun intended) until you are convinced that you can guarantee pictures like No.16. Lower ASA and longer exposure would be good. But I know the frustration of clouds!

(By the was, you refer to photos 13/14 and we have 15/16.)

A methodical approach will eventually reap dividends.

Clear skies,

David

At 02:29 09-12-17, Robert Hoskin r_hoskin@... [ESPMC-Eight] wrote:


Well, a little progress on my seagull issue...

We finally had a bit of clear evening sky (just a bit...). I was able to get David's drift test done. Two 5-minute images, no motors, scope repositioned in-between, just to see what the tracks look like. I thought I had set the camera down to ISO 200 earlier, but evidently not, so these are still at 800.

Then the clouds rolled in as I was trying to get my PA done, and that was that for tonight.

DSC_0190_013.jpg clearly had *something* cause a jitter mid-track, but it settled down quickly. I was in the house for the duration, but there's a train track nearby that runs the occasional evening train. Could have been that, I suppose.

DSC_0191_014.jpg seems like it has fairly smooth tracks.


r_hoskin@...
 

Thanks, David!

No, it was absolutely stationary; mount axes locked, motors off, DSLR on OTA, OTA on mount. I'd moved indoors by the time the intervalometer fired. 

I thought the point of this exercise was to isolate any physical vibration component (cars, trucks, whatever) of my seagulls and "S"'s, so this was hands-off except to set up, and move the mount between exposures.

I was pretty pleased with it from that point of view.  *Something* showed up, but whatever it was did not produce the un-starlike shapes I saw before, and it did not persist.  There was an event in the middle of one shot, and then it was over.  So, if this happened when I was imaging, I'd lose the shot it was in, and that would be that. 

Those star trails are 5 minutes long, so that makes the event (or series of events) just under one minute long and apparently confined to that time window. Could have been anything, but a freight train 200m away might do the job. We do get some long ones.  No guarantees that there *was* one during the shot, as we do not always hear them, but that's my guess until something better comes along.

When the skies clear again, I will do another stationary star-trails shot, but also want to move on to a tracked shot, trying to do a better PA, and seeing if some focuser tightening I've done helps.

Sorry about the number typo!

- Bob


 


r_hoskin@...
 

Here's a selection from tonight's images.  No seagulls, but still some problems. Thanks for any assistance!
 
Cold out: 13F/-10.5C
Camera focus not as good as the first time -> fuzzy stars.  Live view, scope nearly vertical, cold...
General area is Cassiopeia.  D5300/ISO200.
Focuser rotator ring tightened. Camera cables tethered to the scope and not weighing on camera.
Better PA than before, but not good.  Tried alignmaster, and it did not go well.  Gave up and eyeballed Polaris. 

DSC_0193_017.jpg is a 5 minute star trail.  The mount's off.  Trails are fairly smooth this time, although some trails are individually perturbed briefly.  Seeing, perhaps?
DSC_0177_018.jpg is a 10-second tracked shot. Round (fuzzy) stars. 
DSC_0214_019.jpg, DSC_0215_020, DSC_216_021.jpg are a series of 3 60-second images in a row.  Notice that a) the stars change shape, and b) the image shifts slightly from shot to shot. 
DSC_0224_022.jpg, DSC_0225_023.jpg, DSC_0226_024.jpg are a series of 120-second images in a row.  As before, the stars change shape, and the images shift.  Curiously, DSC_0224_022 shows elongated S's (Safeway logos :-) when enlarged, but these fade, with 0225 being a little hooked, and 0226 smoothing out to streaks.

What do you think about the overall image shift, and the trailing in the 120-sec images?  Is that all polar alignment, or do I also have a tracking rate problem? 

The star shape variations from shot to shot are interesting.  Opinions? 
FWIW, Cleardarksky.com rated seeing 'bad', for a site a few miles away from me.  But... if most of the stars are varying at the same time, I guess that means that there's still movement somewhere. 

Thanks again!


hubbell_jerry@...
 

Hi Robert,

Sorry I haven't jumped into the conversation until now.  I was looking at your images and I think that the polar alignment is off for one. I submitted one of your images to nova.astrometry.net to see what the orientation of your image was to make sure that the apparent drift was in the Declination axis, and the solve verified that it was. 


I submitted the image 


and UP was 196 deg E of N or South UP 16 deg to the west. The drift was in that orientation.
The Seeing was bad tonight with high clouds (I was doing some imaging tonight also) causing the blobby stars. Your focus could probably be tweeked a little also. I didn't measure the FWHM for your stars.

I don't think the tracking rate is off per se, just the normal periodic error you would get with this caliber of mount.

Jerry Hubbell
Director Electrical Engineering
Explore Scientific, LLC.

 


r_hoskin@...
 

Thank you both!

Instinct was telling me that the shapes I was seeing were the result of a combination of problems, but I wasn't sure how to isolate them. This has been a great help, and thank you for the confirmation on the tracking rate, Jerry!

Last night's PA was supposed to be done with Alignmaster, but it didn't work out.  Looks simple and quick to use, and some folks do love it, but the stars it really wants to use are near the horizon.  Who, from a municipal back yard, can see the horizon?  I did try tuning its star file, but no joy. It just dug deeper and found more stars blocked by houses and trees, so I gave up on it, eyeballed Polaris, and captured what I could. [David, I'm not using ES - that requires Win10.]

I can see Polaris, so that's a help.  A Polemaster would be lovely, but that would set me back almost CAD$500 with shipping and taxes for a single-purpose tool.  Sharpcap looks to do the same sort of Polaris-neighborhood plate-solve with existing hardware, so that's a plus (even if I don't have the hardware it wants to use, yet).  There is always drift, but between winter cold, portable gear, and probable short observing windows, I wonder if that would work out.

This is the point where I think I need to involve additional software or hardware in the process to help with alignment, so some thinking and research required.  Somewhere in that mix would be BackYard Nikon or equivalent, to help manage the camera, and that will give me a better solution to focusing than I could ever have with Liveview.


Robert Huerbsch <bobhuerbsch@...>
 

Get SharpCap running properly during the day so you dont waste time fumbling at night. Should take 5 min to PA via sharpcap. If you have little cone error, you can get v good PA with sharpcap. I would then spend another 10min to quickly check drift in PHD2. When you get it perfect you can guide without DEC! This is on a hypertuned EQ6 but I got the same results with a EXOS-2GT on a pier...



M31 with 3Ti and EXOS-2GT in a LP location


Wes Mcdonald
 

HI Bob:

1.  Focus can be made near perfect with a Bahtinov mask and live view, or at least a real time download and display of an image from your camera.  You can make such a thing, just search for focusing aids.  MAking a Bahtinov is probably too hard, but there are others, and also buying a mask is not too bad, and it is not a one time tool.  Not shilling for Highpoint, but here is a link....https://www.highpointscientific.com/astrozap-full-aperture-aluminum-bahtinov-focus-mask-for-5-5-to-8-5-od-telescopes-az502?utm_source=google&utm_medium=cse&utm_term=AZP-AZ502&gclid=CjwKCAiAvMPRBRBIEiwABuO6qaWWyn2ITPW4NB2rS9IlO0BS-6mJX7kpA6k_xfvijB3yAgXyVpM5zRoCli0QAvD_BwE.

A Hartman mask is easier to make, but not as effective it seems.  Here is a link that has a link to an online Bahtinov mask plan for self making...https://photo.stackexchange.com/questions/23972/how-to-nail-focus-for-dslr-astrophotography, link to generator is....http://astrojargon.net/(X(1)S(lgbxxovml1fn2a454rea4p45))/MaskGen.aspx?AspxAutoDetectCookieSupport=1


2.  Every time you go to do astrophotography, you are going to need to PA.  Thus a Polemaster is one of those things, about as important as an eyepiece...think of it as an investment.  And remember, "if you get your child interested in astrophotography, they will have no money left to get into trouble" applies to adults too!

3.  One of the things that has to be perfect to prevent declination movement is the level of the telescope, I think.  While I think it is true that when you do the ES alignment step ES computes your mislevel and makes adjustment, such mislevel will lead to image rotation in the camera field., thus cause streaks with long exposures...at least I think this is all true.  Anyhow, I see you have not used ES....if this is true then you don't get any benefit from the ES mount model, this would equal more rotation, I think.

4.  Jerry's plate solve really is the best clue here.  What you got to prevent is declination motion.  And that takes a really good PA.  So PA is the way forward.  BTW, don't feel bad, I still am a clod when it comes to PA, even with a polemaster.  Just not quite as bad a clod as without.

5.  Note also Jerry's mention of periodic error.  We have not addressed that yet, but I assume there is some, maybe a lot.  The fix for PE is guiding, so break out your wallet....again.

Thanks for including us all in this, in my case you are the leader of the band, and I am benefiting greatly from your efforts.


Regards,
Wes.




r_hoskin@...
 

Thanks, Wes!

Yes, better PA is next, now that that's been isolated.  I'm still marvelling that a focuser rotation feature I never cared about (and hence ignored) was such a factor in those crazy images before. Much better with that out of the picture, as it were.

The whole -er- focus... of the exercise was to establish some bounds on the capability of the mount unguided, before trying to do anything serious.  Thirty seconds, a minute, two, three?  What's the practical limit?  Turns out that it's too soon to tell, because there's more to learn before I can have that answer, and (probably) more kit to buy.  I had intended to work unguided (minimalist here...), but it may be that the simplest route to an alignment process I'd be happy with, is to change that plan.  Things to think about...