Offset Init & flip during SGP sequencing


MartinC
 

Hi,
I'm quite new to SiTech having recently purchased a new MESU e200. So I'm probably missing a trick here.

The Mesu has replaced a previous mount with no other changes other than I've built a new pier so that I could take advantages of the 'flipless' option that the e200 provides. It's housed within the same automated dome.

Straight out of the box I was surprised to find that with SiTechexe set to GEM mode it did an automatic flip at both Meridians. Granted there's some elbow room as I was able to configure the Meridian limits up to a point but the mount would always auto-flip somewhere.

After contacting Dan & explaining why I specifically wanted to run in GEM mode but avoid any auto-flips he kindly provided me with SiTechexe version 95Mwin. This gives complete control over the meridian limits. After a little experimentation I found that the attached mount params allow me to track past either meridian without ever flipping. But it will flip for a GOTO to the oposite side of the meridian- just as any GEM would - perfect (I think).

My initial Offset Init was in the EAST & always is if I ever need to re-initialise for any reason. East is much easier for me.

I've been using SGP for a while so I'm quite happy with how to set it up and use it but here's my problem:

My Mesu is parked as normal but has been previously initiallised prior to parking. So it knows whwere it is before I even begin. I start up SGP & connect to all my gear as normal. When I connect to the scope the SiTechexe interface initialises as expected & I can unpark the mount. So far so good.

If I select a target in the WEST and set up my SGP imaging sequence when I run the sequence my shutter opens, the dome slews to the correct position, the scope slews to the desired target, the focus routine runs followed by a plate solve, centering & offset init (maybe twice to get correctly centred). Then imaging continues as expected.

However for any target in the EAST the story is different.
The first plate solve takes place as expected but then the scope flips to the oposite side of the meridian by an amount equal to it's hour angle. By which I mean if my target is 2hrs before the meridian it will flip to 2hrs after the meridian. The declination remains the same. It then plate solves again but of course the dome does not catch up in time to capture an image. So it tries to plate solve an image of the inside of my dome. Obviously my sequence then fails because of the plate solving & the guide star has also been lost.

If the target happens to be in the EAST but close to the meridian - say within half an hour then I may get away with it & the plate solve succeeds. In which case the scope will re-flip back to the correct side, re-plate solve & the sequence will continue as normal & the mount will track past the meridian & keep going until I run out of sky. Which is just what I want.

Alternatively, if I make sure I'm on hand, I can manually slew my dome to ensure that it is in the correct place for the unwanted\unnecessary flips. But that defeats the object of an automated observatory.

I can't speed up my dome. SGP has a minimum 10 second update frequency for the dome slaving. If I slow down the Mesu's slew rate it gives me a slightly wider wndow of oportunity but not enough really.

So my question is why is my mount well behaved in the WEST but exhibits this strange flip behaviour for Targets in the EAST?

I'm wondering if it's related to my unusual mount parameter configuration (meridian limits) or should my initial Offset Init always be in the WEST? I don't know? Or what else am I missing?

I do know that the unwanted flip when targeting the EAST is deffinitely not initiated by SGP because I know this from the SGP log. However I'm not yet familiar with the SiTech log. It took me a while to find it & I realised none of the options ASCOM & Logging tab were selected. So my logs are basically empty.

Can anyone advise me which tick boxes I should have selected if trying to debug this condition?

If anyone has any suggestions as to how to go about resolving this I'd appreciate any help.

Best Regards
MartinC


Don W
 

Hi Martin,
 Yes, I see that your settings are wrong for what you want to do.  I don't have SGP so I don't know what settings there might do.  But in SiTech your current settings allow the flip on goto and your limits are way too narrow (at 2 and 4 degrees for meridian flip limits).  That is why you are getting the meridian flips.

So to eliminate meridian flips in either direction, set the limits to 90 degrees for both directions.  With these settings, you can set the track past meridian limit to zero and uncheck the boxes for GEM autoflip both Goto and track.

With these settings, you can set up your scope either looking East or looking West (using your handpad) and when you do your initialization your mount will remain in that mode all over the sky.

You probably should check or reset your Park positions.  You should choose which orientation you want to remain in (looking east or looking west) and always use that orientation to avoid confusion with parking.
Don W


MartinC
 

Hi Don,

Thank you for your prompt reply. Much appreciated.

In that case I guess I'm misunderstanding the definition of the limits & how they work exactly. To be honest I thought I had this nailed?

Just for clarification I have a pier into which my scope can never collide. Therefore I do not need any track-past limits at all. Therefore I do not want my mount to auto-flip under any condition.

But I would like it to flip as a normal GEM for GOTO's to the oposite side of the meridian. I went into my reasons for this at great length with Dan but in a nutshell the EAST is a good place for me to image. I could configure it as an Alt/Az but in Alt/AZ mode after finishing with a target in the North-West and acquiring a new target in the EAST, the mount simply slews under the pole pointing at my new target with the weights up & the scope so far below my shutter opening that the target is obscured. If you could see how small my dome is & how close the scope is to the wall of the dome it would be easier to understand.

That's why I need to operate it as a GEM and have it flip for GOTO's only; but not auto-flipping at any track-past limits.

Ideally in SiTechExe I'd like to see an option, maybe a tickbox, where the track-past feature can be ignored. After all if you scope cannot hit the pier then I don't think you need it to auto-flip because the thret of damage has been physically removed.

From your reply I'm guessing I have this wrong, but with the 95M version, my understanding was that by setting the underpole limit West to -176, underpole limit East to -178 I've virtually swapped the positions of the Nort & South Meridian lines. With the Over Pole limits set to 4 & 2 and the track past set to 178 I've effectively set the trackpast for the Northern meridian on top of the Southern Meridian & vice-versa.

Before installing the mount in my dome I ran it indoors on a portable pier continuously & it tracked past both Meridians without auto-flipping. Which is why I thought I'd nailed it. However, obviously something hasn't quite worked out so I guess it doesn't work the way I thought it did?

By the way I park the scope in the EAST so that I can enter my dome without colliding with the scope or the weights.

Could you possibly attach a screen shot of your recommended settings as just to make sure I get the over\under, east\west & track-past & polarity's correct as you suggest.

Many thanks
Martin


paul K
 
Edited

Hi Martin,

My contribution is about your system's behaviour regarding imaging before the dome has settled. I wrote my own dome driver and it's ASCOM compliant. My system (SGP  and Sitech) waits until the dome has stopped slewing before any imaging action can take place -  'Waiting for Equipment idle' is the message you see in SGP. Also it's odd that the dome's in place on the west side before a platesolve can take place (this is the expected behaviour). What kind of dome do you have and is it ASCOM compliant?

Paul


MartinC
 

Hi Paul,
I also have an ASCOM compliant dome driver that I wrote myself (it passes the ASCOM conformance test). The dome is an old Pulsar dome that I've done much work on, including building the drive system from scratch. It runs like a dream & I've not previously had any problem with it at all. In fact I don't believe I have a dome problem. I think it's almost certainly how I have the mount configured with the limits. My previous mount ran for 2 years in it perfectly apart from not guiding very well. Hence the Mesu/SiTech upgrade. The only change I've made in SGP is obviously the scope driver, the values for my dome slave settings & unchecking the the "Use Meridian Flip" option. All my other gear is the same.

For a fully automated imaging session my dome behaves as does yours. It stops slewing before anything else happens. I also see "Waiting for Equipment idle".

Any target in the West is fine, over or under the pole. But if I've selected a target in the East everything works just the same (the mount will flip for the GOTO if it needs to), once the scope has slewed to the desired target my filter gets set, then an auto-focus, then a plate solve/offset init.

At this point, unlike in the West however, the mount flips for some reason without the dome slewing first & I don't know why?

When I say flip it's not a genuine flip. It goes to a mirror position on the oposite side of the meridian.

But don't forget SGP's plate solving is still active so SGP just carries on and takes a second plate solving image. In the meantime the dome attempts to repostion itself but it takes time to get there because SGP has a 10sec (mimimum) update frequency + whatever its travel time is. If I'm lucky and the distance is short enough the second plate solve completes and then it will re-flip back to where it should be (on the east of the meridian), re-plate solve and this time continue with the sequence as if nothing has happened.

If I'm unlucky (most of the time) the dome will not reposition in time for the plate solve, SGP will fail into recovery mode.

Last night I watched it go through this loop of events 3 times before I pulled the plug on it.

I'm sure this unwanted 'flip' is purely initiated by SiTech and SGP is totally unaware it's taking place - until the plate solve fails. SGP log files support this.

I'm sure I could just reset my meridian limits to the default, reset SGP to do meridian flips like it did with my old mount and it would be just fine. Except that I wanted to dispense with meridian flips & gain imaging time. Most of my narrow band subs are half an hour so potentially you've lost 30min waiting for a meridian flip. Also I've invested in building a pier & mount that should give me a flipless setup but at the moment I can only image in the WEST.

To be honest I was really surprised to find compulsary track-past & auto-flip in SiTech. I've never had a mount before where you couldn't disable this. Instead I've tried to get round it by playing with the meridian limits & I hope this is be where I'm going wrong. I'm now doubting my understanding of them & just getting more confused.

Many thanks & regards
Martin


Don W
 

Hi Martin,
SiTech controls in ChangeConfig/Mt Parms are simple and do exactly what they say.  The flip controls either flip at the limit or STOP the mount at that limit.  There is no other choice.  Same thing with TRACKING, except that there is also a setting that allows tracking past the meridian limit.  All the limits can be set to zero or up to 90 degrees (from zenith).  Unless you live at Latitude >60° I don't recommend EVER going "Under the Pole".

Th Earth always rotates east to west.  So if you want to avoid a meridian flip, that is ONLY when you start tracking looking EAST.  You CAN NOT track west to east (Except under the pole).

So What should your settings Be??? 
1. You want to have your mount flip to Looking east when you tell it to GOTO for your FIRST Target - so check the box for "GEM Auto Flip GoTo.  Then set "Meridian Limit East to zero (over the pole).  This setting will ensure that you will always start "looking East" for any targets that are east of the meridian.
2. Do not check the box for GEM Auto Flip Track.  This tells the mount to simply STOP at the limit.  This is OK, because you are going to set the limit so the mount never gets there.
3. Set the Meridian Limit West to 90:00:00.  this allows the mount to continue "Looking East" if you want to GoTo a target that is west of the meridian - this will result in "counterweight up" when the target is west of the meridian.
4. Set the Track Past Meridian Overlap to 90:00:00.  Now this will track all the way to the western horizon, never flipping nor stopping.

That is my recommendation for your setup (which is a special circumstance).  The settings for "Under the Pole" will not affect the Over the Pole operations.
Don W


MartinC
 

Hi Don,

Thank you again for your reply. I shall try the settings you suggest but what values would recommend I enter for the Under Pole settings?

I agree that my case is a somewhat special circumstance but perhaps it will become more common in due course, especially with the e200 offering 'flipless' capabilities? For such cases my personal opinion is that any kind of mandatory meridian limit or track past on a 'flipless' GEM configuration, where the scope can never collide with the pier, potentially imposes an operational limitation. Just my humble opinion.

I fully appreciate that for conventional GEM setups it forms an essential safety feature. But it would be nice to have the option to overide\ignore the feature. In my simple mind it could just a simple check box on\off option?

Regarding the pole I do anticipate imaging (slightly) under the pole & in fact I have done several times. I also have a friend setting up an observatory with an e200 on bent knee pier. Due to his site location he will certainly be looking to image under the pole some of the time.

I'll let you know how I get on.

Much appreciated
Martin


MartinC
 

Hi Paul,

In addition to my earlier reply & purely out of interest. I'm assuming you have a conventional, vertical pier & therefore invoke the SGP meridian flip option?
So I guess you set your track past in SiTech to protect your scope & avoid potential collisions? And in SGP you set your 'minutes past meridian to flip' value to something less than your track-past limit in SiTech? That's how I previously operated.

Best regards
Martin


paul K
 
Edited

Hi Martin, yes my MESU 200 is a 2015 mk I model and needs the limits, and yes I have sgp set up like that.

Going back to the previous post, I'm intrigued that SGP will try to capture an image if the dome is slewing. As you know, the slewing property will be set to true, so why hasn't SGP detected that? I think the answer must be because SGP didn't initiate the slew and therefore isn't checking the Slewing property? I've not written ASCOM stuff outside of dome control, but just thinking about it, it's logical for the session management program to initiate program methods. Having just said that, I've realised that if I manually move my dome to an Azimuth other than the target Azimuth, (e.g. by pushing it) SGP does notice it and initiates a slew to move it back. So SGP is checking properties, but perhaps not the Slewing property if it didn't initiate the slew.

Perhaps it might be worth just  trying to link things up through the ASCOM device hub and see if the behaviour is the same?

I also agree with you and Don re the new flipless mount - brilliant to have this as MFlips with an equatorial are usually painful in some way; also because they are so new, it's currently a special circumstance. Perhaps in a future release Dan might have the track past and limits as a tick box option, so those who have flipless mounts just don't see those options.

good luck with trying to sort it.
Paul


MartinC
 

Hi Paul,

Many thanks for your reply & yes you get my point(s) exactly. I agree with your comments. I've checked through the SGP & ASCOM logs including the one for my dome. The only explanation for SGP capturing a plate solve image is because SGP did not initiated the scope slew\flip & therefore is 'unaware' of what's taking place at that moment in time.

After further study I'm pretty sure the settings for my Meridian limits are the cause of the unexpected mount behaviour and that's down to my failure to undestand them properly. I thought I had it nailed but I'll try Don's suggestion next time there's a clear night. Perhaps then it will make sense to me.

I agree MFlips are painful and a 'flipless' GEM is a fairly new innovation. It is a GEM by nature of it's construction but you want it to track past the Meridian(s) as if it were a fork mount. I have asked Dan previously about having the track-past & limits as an option so as you comment hopefully in a future release?

By the way congratulations on writing your dome driver. I hadn't written any ASCOM device specific code before either. A few ASCOM client apps but not a driver so hats off to you as there's a learning curve for sure.

Best regards
Martin


Don W
 

Hi Martin,
On the subject of Under the Pole, I don't do it, so I have not studied it.  However Chuck Shaw created a guide to meridian limits some years ago.  I attached it here for you.


paul K
 

yes I think it would be a nice evolution for Sitech to accommodate the new flipless GEMS. Dan has written some brilliant stuff to get us where we are, so hopefully it's not too much work. Only Dan will know. I've recently purchased a 2.2m Pulsar and just setting it all up in a field with unobstructed sky views (!) but no electricity - so that's my latest challenge trying to keep power consumption down and solar energy available :) I might try to make a wind powered generator - we seem to get plenty of that !

Re dome driver, yes it's a steep learning curve. I used the ASCOM templates and was really fortunate to have some initial pointers from Tom How (ASCOM developer videos) and Rick and Tim at the ASCOM Developers forum. They're a great bunch.

I hope things work out for you and fingers crossed Dan might be able to provide that feature to opt out of flips. There are about 5 folks on the forum who use the eMESU and I think in years to come that will only grow.
best wishes
Paul


MartinC
 

Hi Don,

 

Once again much appreciated many thanks for the guide too. I’ve setup your suggested limits for over the pole & track past so just waiting for clear skies now J

 

Best regards

Martin

 

From: Sitechservo@groups.io [mailto:Sitechservo@groups.io] On Behalf Of Don W
Sent: 07 April 2021 18:58
To: Sitechservo@groups.io
Subject: Re: [Sitechservo] Offset Init & flip during SGP sequencing

 

Hi Martin,
On the subject of Under the Pole, I don't do it, so I have not studied it.  However Chuck Shaw created a guide to meridian limits some years ago.  I attached it here for you.Image removed by sender.


Don W
 

Hi All,
Your comments wanting special settings for flipless GEMs are un-necessary.  The settings I suggested for Martin do exactly what you want.  Nothing more is needed regarding modifying SiTech.
Don W


Russell R
 

Don W.,

I don't ever deal with meridian flip limits, but curiosity got the best of me. What are the colors blue, green and yellow referring to in that diagram? 
Ol Chuck is quite the artistic engineer!

Russell R.


Don W
 

Hi Russell,
The colors are the limits set  for the meridian limits, and they reflect the colors of the limit lines in SkyView.
Don W


MartinC
 

Hi Don,

 

I’ve not had an opportunity to image yet but in preparation I have the settings configured as you suggested. I am initiated in the East and parked in East as I always do.

 

I can perform GOTO’s via Skyview, SGP, Cartes Du Ciel or Stellarium and the mount behaves as I would expect a GEM to behave. I.E. it flips if select a target on the opposite side of the meridian; either going from East to West or West to East. It goes through the pole if my target is in the opposite sky quadrant. And it re-parks correctly from anywhere in the sky.

 

It does not auto-flip or stop tracking when the scope tracks past the 90 deg track-past setting you suggested or if the scope tracks past either pole.

 

This is great but all of the above was also true of my previous settings.

 

The problem occurs during plate solving in the East, when the ‘InitPoint’ window appears during the adjustment (East radio button active). That’s when my mount would flip – but as I said to Paul – not a ‘genuine’ Meridian flip. It wouldn’t be pointing back at the target. It flips to the same declination but mirrors the RA position on the opposite side of the Meridian. I.E. If the target was 3hrs East of the Meridian, the mount would ‘flip’ to the same declination but 3hrs West of the Meridian.

 

If my scope was in the open air & not in a dome then it would actually work out because having performed this strange flip SGP re-plate solves, with the InitPoint window showing the West radio button active this time (I assume because the scope is now looking West). The plate solve error is obviously massive but the mount then returns to the correct position, plate solves again (East radio button active again when the InitPoint windows appears) and this time my imaging sequence continues as if nothing has happened.

 

If my target was in the West the plate solve takes place (West radio button active when the InitPoint windows appears), the mount remains where it should, it takes one or two small plate solve adjustments as you might expect and my imaging sequence continues as it should.

 

So Don, although I’m really hopeful that your suggested settings resolve this behaviour I’m struggling to understand how the Meridian & track-past limits can influence the plate solving\InitPoint routines in this way?

 

If you have any further thoughts please let me know. If not then I’ll get back to you once I’ve had a chance to actually image.

 

Many thanks & regards

Martin


paul K
 

Hi Martin, so if I have it right, what you're saying is if you decide to image a target anywhere in the east and the mount platesolves in order to centre on the target, your mount will flip. Is that right?

I don't image in the east due to trees, so never tried a platesolve looking east.

Paul


paul K
 

I respect your position on this Don, but I can just hear folks saying - why am I getting involved with instructions for driving a car when I drive a motorbike.... :)
Perhaps a minor mod to help those folks who don't need to get involved with the not inconsiderable detail of meridain limits and flips, might be to provide a tick box in Sitech config which auto populates the boxes with the setttings needed for flipless working. Just trying to help....
best wishes
Paul


MartinC
 

Hi Paul,

 

Yes that’s exactly what I’m saying J

 

So just out of interest you always initiate in the West? And park in the West too?

If so then you are initiating, parking & plate solving all on the same side of the pier?

I’m just trying to get a handle on how others are working & if this could be relevant?

 

Many thanks for your support.

 

Best regards

Martin