PinPoint

[Big Bang!]

Chris

Something unexpected has happened this evening...

I have done a cable reroute in order to tidy up what was a bit of a mess - I had always been concerned that cable snagging might be causing a problem and wanted to be sure this had been eliminated.  So I did a new model; Multiple iterations of the PA got it down to 0' 30" and then I put in 69 points as before.  When I went back to check the model the PA had drifted out to 2' 16", requiring a 0.05 Rt to correct. I haven't bothered to do this as I want to see what the tracking is like but  can you give any insight to why that should happen? 

 

Gus

 

[ChrisLX200]

54 minutes ago, Big Bang! said:

Chris

Something unexpected has happened this evening...

I have done a cable reroute in order to tidy up what was a bit of a mess - I had always been concerned that cable snagging might be causing a problem and wanted to be sure this had been eliminated.  So I did a new model; Multiple iterations of the PA got it down to 0' 30" and then I put in 69 points as before.  When I went back to check the model the PA had drifted out to 2' 16", requiring a 0.05 Rt to correct. I haven't bothered to do this as I want to see what the tracking is like but  can you give any insight to why that should happen? 

 

Gus

 

Difficult to say but perhaps the cables are still pulling and inducing a bias into the model (which will also affect the calculation of the PA position). Another useful feature of ModelMaker is that you can pull up a map of the model points and view (in the form of a scattering of coloured arrows) all the offsets/position errors which each of the model points contributes. If there is a bias it shows up well there. I was hoping to get my scope out tonight but the forecast changed at the last moment, otherwise I would post an example.

Chris

[Big Bang!]

Chris,

In order to minimise timing errors in my modelling I have today installed NTP on my Laptop.  Seems to be updating UTC ok and I am using Per's ASCOM driver so the mount gets an update on each slew or every 30 mins. 

However, for a reason I am not able to get to the bottom of the UTC and Local time are two hours apart with one hour set as the Time Zone difference. How does that work?  When I sync UTC correctly, Local jumps to two hours later; if I correct Local in the handpad, UTC jumps two hours before.  I don't think Local has any impact on the modelling?  So as long as I have UTC correctly Synced I should be ok?

But do you have any idea what will have caused this or how to correct it?

Gus

[perfrej]

Dang, I missed out on this thread entirely. Sorry about that!

You need to set the correct time zone - the "real" one that is used during winter, not DST. After that, select daylight saving on or off (on now, then) and then set the clock. When autumn comes, just turn off daylight savings and you're done.

/pe

[Big Bang!]

Hi Per,

Thanks, I got that now.  I was a complete numpty:  I never did set the time zone in the first place and assumed that it was worked out by the mount after getting site Lat/Long off the GPS.  So I ended up with the time zone set to plus one hour and DST on top of that - 2 hours from Local.  It was right in front of me and I couldn't see it!

But now I've got your attention for a moment...

I have an inconsistency.  Last night: 82 point model, 4.8 RMS.  30 minute frame through a green filter - stars absolutely as round as a round thing.  Next frame, immediately afterwards through red, I've got about a 60% diameter trail and the same or slightly worse after that through luminance.  I have assumed the model to be good given the quality of the green frame so I am thinking it is either the filters (bottom of the range, cheap) or the fact that they were all binned x1, giving a way oversampled scale of 0.38"/pix! (The seeing was excellent) However, I also took a 20 minute red control frame at x2 and although it was better than the x1 red it wasn't a patch on the x1 green that was 10 minutes longer. I never touched the focuser so I don't think its that.  

Hazard a guess for me where the error is.  Could cheap filters cause aberrations like that or is the oversampling more likely? But if it is oversampling why was the green frame absolutely perfect?  Attached some zoomed pics of the same star.

What thinkest thou, O wise one?  Should I be looking at better filters?

Gus

[Physopto]

Well that is strange! I would like to know what has happened there as well. Only thing I can think of is wire pulling? as your green looked OK.

[Big Bang!]

Yes, it is odd isn't it.  I doubt it is a pull because I have followed Per's example and got my cabling all tied together in a loom that doesn't snag.

Gus

[Physopto]

Weird to say the least. You didn't say something detrimental within the mounts hearing did you?

 

[Big Bang!]

[Physopto]

I had something like (not same ) it once but I doubt you made my mistake. A very slightly slack clutch.

 

[Big Bang!]

Well, I'll check that but I don't think so.  Also, if you look at the direction of trail on the bottom two you will see they are significantly different and that makes me think it cannot be the model/tracking as the frames were immediately subsequent; I think it must be something that is distorting the light - made me think filters, which are not the highest quality, or the oversampling.

Gus

[Physopto]

I would love to help but cannot think of anything else. To be honest though unless the filters are really bad cannot see it being them. If so surely you would have noticed that before. I would still look at something either loose or having moved somehow.

I will watch with interest.

Derek

[perfrej]

Gus,

I browsed through the thread... You're at 2000 mm FL? That sounds like a reflector to me. Describe the setup, please  Focuser, camera, wheel, scope...

 

/p

[Big Bang!]

Hi Per,

I have a permanent set up in a Pulsar Observatory.  1000 HPS on a pier bolted to over a ton of concrete. Altair RC250TT (with a TS70 APO sitting piggy back), Moonlite focuser, Lakeside AutoF, Baader Filters, ATIK One 9 camera, your Stickstation.  

I now use SGPro and PinPoint for modelling, which has improved things greatly.  (I have Model Maker but not Maxim so I'm in the queue for the upgrade).  Typically I now get below 5" RMS and can achieve 30 mins unguided with what I consider to be perfect stars (and I call a spade a spade) but it is not consistent.  You can see from those pics that over the space of 2 hours I have one perfect 30 min frame and two with trailing at different angles through different filters.  I recognise that at 0.38" they are over-sampled but that green frame is way better than a red I did at x2 (0.76).  Didn't touch the focus between frames  so I don't think its there.  The cable loom is tied to the OTA, then hangs clear of the pier before being tied again to the top of the pier.  I don't think the cable loom can be snagging. I am not aware of slack anywhere in the rig - have certainly done my best to eliminate any movement.  The only place I have not looked is the internal filter wheel in the camera - I'm a little bit reluctant to go messing around in there - but that is the only thing that I moved between frames.  So the mechanism or the filters themselves must merit an investigation.

Before I recently upgraded my modelling techniques I was typically getting, what I call, good frames at about 10 minutes - 15 at a stretch.  I know the mount is capable of much better - obviously, something is not quite right. I was just wondering of cheapish filters could cause this kind of problem.

As always, very grateful for your input.

Gus

 

 

[perfrej]

OK... After loosing the piggy I'd start with the focuser and the main mirror cell. Inconsistency usually comes from things moving inconsistently. I have a Moonlite myself and I would definitely try something else after looking at the main mirror.

Still, given that you have a rather beefy scope and even a pig on top - and the smallest of the 10Micron mounts - your results are excellent. It's just the inconsistency... [removed word]!

 

/p

[Big Bang!]

Evening Per,

You feel the piggy back is asking too much of the mount then?

I am not familiar with examining the assembly of the scope.  I have never checked it over but rather assumed it would be correctly set up at the factory.  How does one ensure that the main mirror cell is secure?

I feel I have been banging my head a bit on this so thank you for your kind encouragement - most grateful.  

Gus

 

[perfrej]

Hey Gus,

A piggy-back tends to flex some, so if you temporarily take it off that would be a good test. Next, the primary mirror cell - I have yet to see one that doesn't shift at a cost below crazy, so just open the back and look at it. Shift is OK for imaging but not while recording the model. Then grab your camera and try to move it back and forth. It needs to be rock steady.

 

/per

[Big Bang!]

Hi Guys,

Dismantled everything yesterday, including the mount off the pier and started from scratch in my usual meticulous fashion.  Now no piggy, both primary and secondary felt rock solid, cleaned them both, tried moving the camera/focuser around a bit, couldn't feel anything.  

This evening, a 45 point model; 20 terms and RMS 4".  However, two pics, same star, first at 20 minutes and the second at 10 seconds.  They have the same shape, which surely means it is not a tracking error because if it were, it would be far greater in the longer frame and, unless it was huge, you would not expect it to show up on a 10 second frame ?  Suggests to me this is optical in some way.  

Collimation looks a little off to me but far enough to cause this kind of error? Laser on order but I've never done it before. How do you know which screw to turn or is it trial and error with all three?

And finally, many of the brighter stars had double diffraction spikes! Suggesting two bright spots - consistent with what is visible in the second pic.

This is driving me nuts.

Gus

 

[Physopto]

For some reason that looks like you have two optic axes. It is as if you have something causing birefringence. I can't see how unless you have a calcite type element in the optic train, highly unlikely as I have never heard of that in a standard scope and can't see any reason for it. There what looks like two focal points. Could be a reflection maybe as the second point is reduced in size.

Actually thinking about this a bit more I am wondering if the mount is jumping between two points during its tracking. The movement may be so small that you cannot tell it is doing it. May still be something like a loose electrical connection internally. 

I will be very interested in the explanation, but cannot offer any help, sorry.

Derek

 

[perfrej]

That's got to be either a mirror shift or one of the older, less reliable firmwares. What version are you on?

 

For the record: reflectors that cost less than one umptzillion usually have mirror shift and need to be improved in the main cell area.

To check for shifts, point the scope at north horizon, loosen the dec clutch and swing it south. Hear a "clunk" sound? Then you have a BIG shift. If you don't hear a sound you will have to dig further by taking it out and pushing the mirror back and forth with your hands. It needs to be firmly in place and not move.

 

/per

[perfrej]

In your double-spike image, I interpret the lower spikes to be somewhat less exposed, implying that the shift happened right before or right after the half-way point in the exposure. Sounds reasonable?

 

/p

[Big Bang!]

Hi Per,

I updated to the latest version last week 2.13.6, however, I believe since then I have taken one 30 minute frame that I consider as good as I could ever hope for (raw attached). 

Yes, the exposure is not even between the spikes and that does suggest either side of mid-way is the shift point. But since the exposures are of very different lengths it means that the shift occurred after about 10 minutes in the first frame and after about 5 second in the second.  I wonder if that is random or driven. Why would software change anything mid-exposure.

I am running your Stickstation but I believe it only updates on slew so I don't think that can be it.  I do have your driver set to update mount continuously using the laptop system time which is NTP connected but that is on slew only too?

I am going out to the Obs now to really get to grips with the mirrors but I have to say I have already tried really hard to move them and cannot. Nonetheless, I will try again.  I will have a laser collimater by this time tomorrow and I will immediately put that to work. If it is not a mirror shift then it has to be a software issue of some kind so my plan is to remove the Stickstation, stop the NTP service, un-check the continuous update in your driver and run another test - I will add back each one until the error reoccurs.  If it is still there with that lot deactivated then I would have to guess firmware.  

Need some clear nights.

Gus

[Big Bang!]

OK guys I've had a play around doing Per's horizon test.  There was no clunk as such but on each swing there was the sound of something shifting very slightly.  But it wasn't the mirror - even with real force the only way I can get the mirror to shift is if I slacken the bolts off.  It was the thin metal cylinder that shrouds the mirror shifting as the scope went through the zenith.  As this is metal it must have a reasonable mass and therefore momentum once moving.  So, in three places I've squeezed a rolled-up piece of electrical tape in between this shroud and the frame and that has stopped the sound.

But surely, during the course of tracking a target this shroud would only shift once and then ought to be secure in its new position until the scope is next slewed through the zenith?  Yet, I have a series of frames with the error in it. I am doubtful this is the fault but I cannot find anything else.

 

Gus

 

[perfrej]

Well, you may be onto something there. I had to do exactly that to my Skywatcher 190MN in order to reduce mirror shifts. As for firmware, I suggest you grab the 2.13.16 beta - which is very, very good. It's on the site.

 

/per

[Big Bang!]

Thanks Per, I'll get that now.  Forecast to be clear over the weekend so I'l report back next week.

Gus