Solved! Troubleshooting Maxdome alignment parameters request

[g__day (Matthew)]

Hi folks,

I have just purchased and installed a Maxdome II dome rotation kit for a Sirius labs 2.3 metre dome set up. The set up went okay after two modifications Sirius advised:

1. Using a 3mm wedge on the control box's left hand side to prevent the side of the box hitting the homing sensors (the box didn't sit perfectly aligned to the track - it was way too close on the left hand side)
2. Angle grinding out a 1 cm square section of the joins of the two parts of the dome (front and the rear of the dome) - as the join itself sat too close to the motor unit's housing block - it was colliding with the motor gear's housing socket at these two points!

So now it homes and parks well, moves fine and connects over a serial port to the PC absolutely fine. Tics per revolutions is the standard 227 and the auto-calibration doesn't change this value.

What is wrong is the alignment of the dome with the scopes is slightly off - say 5 degrees maximum. I have done my best to triple check all the set up parameters (but expect my guess of home position Azimuth may be a tad wrong).

I wanted to query folks advise on how to best fine tune Dome and scope alignment. I am running three scopes - a C9.25 on the left side saddled to a Williams Optics 115 on the right, with another Williams Optics 90mm on top of this. So my set up is fairly wide and thus sensitive to dome slit alignment!

All slews of the dome seem accurate - but I am yet to try from the ASCOM controller to slew the Dome to 0, then 90, then 180, then 270, then 360 degrees to see that it is moving precisely each time. However it finds the home location with ease each time and parks rather precisely.

It may just be a very bad guess as to where the dome is orientated to in its Home position (I estimated its at 130 Az) when the home Sensory is aligned to the control box.

I also checked the mounts time to internet time - but tired me - I didn't check its pointing was spot on (and I haven't used it in months) - so tomorrow I will have to re-align it to eliminate a rather obvious source of error!

The other settings (for completeness) are:

1. Control Software:

Mount The Sky6 PE -> Tpoint -> Vixen Skysensor2000-PC driver
Dome MaximDL v5 -> ASCOM Dome and linking telescope -> Maxpoint -> The Sky6 controlled Telescope

ASCOM Dome Settings:

GE Mount switched on,
Automatic Home on first Connection switched off
Sync Home to Azimuth switched off

Latitude -33.78
Long +151.12

Dome Radius 45.28 inches (Sirius 2.3 metre diametre dome)

Dome Centre distance East of Mount Pivot Point (defined as junction of RA and Dec) +2 inches East of Dome centre

Dome Centre distance North of Mount Pivot Point - 18 inches (this might be part of my mistake - measured this without a plum bob - it is more likely only -13 inches)

Height of Mount Pivot Point above Dome = -28 inches (the base of my dome sits above the OTA) - must remeasure this to where the wheels of the dome are - it might only be - 25 inches.

Distance of Mount Pivot Point to Telescope Optical axis = 13.5 inches.

So I have some measurement and the simplest of error checking to do tomorrow. Any other folk who have struggled then found the best way to address this - your advice will be greatly appreciated.



When I point the scopes North East to South East - the dome portal is a few inches too South. I think it gets a bit worse the further I head South - which must be a clue to the misalignment error in my parameters!

Many thanks,

Matthew

[Auster1b (Neil)]

Matthew,
If your alignment is varying, then it is the NS EW offset parameters that need tweeking.
Once you get a consistent dome to scope slaving,you can then tweek the ' 'sync home to xxx degrees'.
I did a rough guess, then added 2 x 5 degree steps until satisfied.

Home is a function of the reed switch and magnet so will vary depending on where you actually place the magnet.
Park is a function of the offset parameter 'sync park to xxx degrees' for shut down clearance or in my case gust lock alignment and solar panel azimuth.

I actually relocated the gust locks on my second hand dome to ensure the panel was close to due North. I also relocated the home magnet to the same azimuth so that when doing a manual slew, it is easy to slew to the home position where the dome will stop automatically as it passes the magnet.

It is also very easy to get 'lost' inside the dome so I also added azimuth labels on the walls.

Neil.

[g__day (Matthew)]

Neil,

Many thanks for your suggestions, I particularly like the Azimuth label idea!

Park I just selected so the dome shutter faces the house at a pleasing angle.

I Have to re-read the "sync home to xxx degrees" setting again. What I tried was slewing the scope with the dome slaved to it, then in software disconnecting the dome, then in the dome driver settings altering the Home position between 90 - 130 degrees and re-connecting the dome with track telescope set in MaximDL Dome control tab which would cause the dome to adjust position relative to the telescope, all with varying useful results dependent on which part of the sky the telescope was pointing.

Your Offset North South and East West comments sound rather sensible, so after a coffee I am going to try measuring things more accurately with a plum line.

One question, why did you choose varying the reed switch Home location with the sync Home to xxx degrees rather than the set Home position to that figure - was it just easier to change on the fly or easier to change when the scope is in any specific region of the sky?

Many thanks,

Matthew

[g__day (Matthew)]

Hmmm, two interesting things:

1. I used the dome control software to move the dome from Az 0 in 30 degree increments in a full 360 degrees and marked the position with post it notes. When the dome crosses the home location left or right seems to vary final position by about 2 inches - maybe the home location is slightly off; but I am surprised rotating the dome left versus right gives consistently different positions.

2. Checking with a plumb line the Dome Centre distance North of Mount Pivot Point - is somewhere between 5 - 6 inches (not 13 - 18) this had a big impact on the pointing precision. Its almost perfect now - I'd say its about 3 degrees out in some locations still - but it appears much better centred in all locations and elevations now.

More experimentation!

[Auster1b (Neil)]

Matthew,
Mine does a similar thing, but only 1/4 of a metric inch from either direction.

I suspect it as the reed switch activates under influence of the magnet, in your case 1 inch either side of the magnet.

Try moving the magnet a little higher on the ring (that is where mine is mounted) to increase the distance from the reed; Maybe an eighth or so at a time.

Mine is 15 imperial mm from the top of the switch to the bottom of the magnet housing.

Neil.

[Auster1b (Neil)]

In relation to moving the home switch, it so the dome can home even when manually slewing using the hand controller.
The dome will stop when manually slewing past the home switch. ie no computers powered up.

Neil.

[g__day (Matthew)]

The magnet being a little too strong and being detected too early might make sense if its combined with the dome having considerable inertia and no smarts to slow the rotation as the dome approaches where the software believes the home sensor to be. But I would expect the clock wise versus anti-clock wise alignment error to only be about the length of the homing magnet - say 25mm not 50-60 mm, unless that is interia.

The rotation track itself is only about 18mm wide of which 3mm is taken up with teeth the motor meshes with while the home sensor itself about 10 mm wide and the matching sensor plate is itself part way down over the track - so there is only about a total of +/- 2 mm of play relating to where you can place it on the track! I have mine only 1-2 mm above the teeth.

Well last night I set up and aligned the mount to three well placed stars, refocused my gear and slewed to NGC 2070. The dome auto-tracked to a pretty good position for my main scope, I changed the home position in the software by about 5 degrees and this positioned all three scopes with the viewing portal. I imaged the nebulae for an hour and when I checked at the end of the run the dome was still very well aligned will all scopes; so I would call that mission success.

I don't know if my modification of the starting position will only allow good alignment for specific Az sky locations of all locations yet; so until I tune the mount geometry parameters I guess I can work around mis-alignment by building up a table of offset corrections per target Az. This I figure would allow me to keep great alignment anywhere in the sky for extended periods of time with all three scopes.

The final joy was checking the solar charger this morning. The little light in the control box is glowing amber - meaning with the main unit switched off the solar cell is charging the battery. Does anyone have an estimate on how many hours of direct sunlight it needs to top up the battery for an hours worth of home rotation?

Feeling good this morning!

[g__day (Matthew)]

Well I have raised a support case with Cyanogen to advise next steps:

http://www.diffractionlimited.com/fo...lockwise.1536/

In it I asked what is the ideal distance that the home reader and magnet should be separated by at closest proximity (mine is 2-3mm away at closest approach).

I would be very interested in know what distance folk here place their homing reader to the magnet at closest proximity!

Many thanks,

Matthew

[w0mbat (Ian)]

Hi Matthew,
As I told you when I sent you pictures of Maxdome I also have had alignment issues which I have not resolved yet. However a couple of things I have learnt on the way may be helpful.
Dome setup refers to the dome equator. This is not at the bottom of the dome. Sirius advised me that the equator of the dome is 100 mm above the bottom edge.
Have you accurately determined the intersection of the RA and Dec axes? This of course is inside the mount. All measurements refer to that invisible point.
Somewhere in the Sirius or Maxdome instructions it says that the dome drive does not stop till it reaches "the next" change of reflectivity on the shaft. I guess this will mean a slight discrepancy between the stopping points depending on which way you approach home position. I think you should always home in a clockwise direction.
I have had a lot of communication with Diffraction Ltd. My initial theory was that there was a problem with the Maxdome algorithm in the Southern Hemisphere but they absolutely insist that the system is verified to work in both hemispheres.
They were absolutely no help with trying to find where I am going wrong. They just kept saying my measurements must be wrong. I have remeasured over and over again trying different methods and always get the same results.
In the end I moved on to other issues so have still not solved why my scope points properly through the slit in most areas of the sky but significantly wrong in others.
Good luck and keep us informed,
Ian

[g__day (Matthew)]

Hi Ian,

The Dome equator being 100mm above the bottom edge makes sense to me if they are referring to the mounting ring I use. To me Dome equator would be were the dome wheels touch a surface to roll on - this is what I used, then entered a negative number because my my mount sits well below the wheels rotating the dome.

I have tried (just last night again to) use the exact mid point between the RA and DEC axes (which as you say is inside the mount), and checked when I measure from the dome centre I used only the North / South and East / West components of the mis-centring. My lab is square and faces North East, so when you are inside it its easy to think the house is East of me instead if South East! I fixed this up on the second set of measurements. Last night was my third set - this time I used string and levels to mark the exact dome centre at the height of the mount axes centre point. Turns out I was 4" not 2" East of the Dome centre and my South measurement was off by an inch too. So hopefully this time I have nailed it better.

Homing in a set direction sounds like a good work around to me. If the motor turns thru another change in reflectivity mark that should only be the maximum of half a revolution - which to me would be say 10 - 12 mm.

With only one imaging session (NGC 2070 for two hours) managed successfully (using the old geometry parameters). My approach was to slew the dome, check for dome / scope mis-alignment, then alter the dome home position Az reading in software by a few degrees - and let the dome re-position itself to the scopes. After two goes this had the dome perfectly orientated to the scopes, and after two hours they were still well aligned. Is that the approach you and others use for imaging to centre the dome if your geometry is a bit off?

Another mistake I had made was trying to use the scopes Az when pointing to the middle of the dome slit in its home position to determine the home Az; but as the dome and Mount rotations points are mis-aligned I presume the domes position for say East with be a degree or two further left than the scopes bearing for East!.

[w0mbat (Ian)]

Strictly speaking the dome equator is the line around the dome where it is half a sphere. In the Sirius 2.3M that seems to be at the level of the black internal wheel covers. (I'm not sure if you have these)
I'm not the person to ask about long term usage habits as I am still in the commissioning stage. My impression is that correct slit alignment is achievable in all sky positions IF you get the parameters correct. One experienced user told me he had to fiddle the numbers (away from the measured values) till he got it to work but that it did work.
I do my tests in the day by shining my collimation laser through the viewfinder of the DSLR mounted on the telescope which projects a good image of the aperture on the dome.
My mount has a polar alignment laser in it so I use that to align the centre of the slit due south to get the azimuth position correct.
Another issue I am not happy about is the sign of the North South offset parameter. Logic suggests to me that as an equatorial mount in the Southern Hemisphere faces south rather than north the N/S offset sign should be reversed. Diffraction Ltd initially answered that they were unsure about this. Then the software author wrote that it definitely should not be reversed. Doesn't make sense to me. If anyone understands this I would love to know.
Somewhere I found the maths involved in calculating dome position. There are pages and pages of complicated looking maths! It is not a trivial problem. So the definition and locations of things like the dome hemisphere and mount intersect are critical.
If you set the dome to home on start up it always rotates clockwise. You also have to add the dome/motor inertia which carries it slightly past the switch off position which will add to the directional discrepancy. Somewhere in the manuals it says homing is always done clockwise and that if it is not happening that way then the motor wires should be reversed.
Ian

[w0mbat (Ian)]

Matthew, I just thought of another problem I had which you should watch for. I assume Maxdome still comes with a Kemo charge regulator. Make sure only one LED is on at a time. If my battery needed charging the charging LED would be on and the battery full LED would come on when it was charged. The problem was the charge light stayed on. I thought this was normal until one day I found the battery was dead from overcharging. I think the regulator was faulty from day one. Sirius would not cover it as I had had the observatory kit for about two years. They quoted $36 plus postage. Then I found Jaycar sells the exact same regulator for $29-95.
Ian

[g__day (Matthew)]

Thanks Ian,

The regulator is showing only one light on at a time. I know the regulator items that Jaycar sell - they look identical. It's almost 5am on Sunday morning in Ottawa - so I hope Diffraction guys have a simple answer for me soonish. It might just be a case of move the magnet further away. Given the magnet is stuck onto the metal track - maybe that is causing a widening of the magnet field too!

Matthew

[Auster1b (Neil)]

Matthew,
Hard to get a good shot, but here is how mine is set up although perspective is skewed.

Reed switch ~ 7 mm sq, attached to track cover.
Magnet ~ 7 mm sq double sided to ring.

Vertical distance ~15 mm.

I took the equator to be the centre of the wheels, which agrees with drawings.

Neil.

[g__day (Matthew)]

Some great new here - I am glad to report it the dome is now very consistently landing on the right Az bearing when slewed from the software; regardless of direction traveled to reach the compass bearing every time!

My correction and testing methodology was simply:

1. Inspect the motor and sensory housing for slippage - there are two bolts holding the sensor and it seems to be firmly locked in position
2. Check where the dome stops when it slews past the home sensor manually or automatically - it always stops with the sensor that is stuck on the track about 1/2 inch to the right (dead in the middle) of the read sensor on the motor box
3. Blow any excess dust remaining from the bottom of the toothed track being initially worn by the gear wheel, or fibreglass dust from the fit out away and leave everything clean - then dust it again with a fine paint brush
4. Set the dome park and home location to 130 degrees - my best guess
5. Move the track around from the home location in 10 degree increments and mark exactly where that landed and go back and check several previous readings every 20 - 40 degrees.
6. Randomly slew the dome from multiple locations to multiple other locations - close by, left or right, far away.

Only one slew seemed to confuse things - the dome did a complete 360 plus to get the home sensor right again then it nailed the goto. In about sixty slews it seemed to nail the landing to with 1-3 mm every time!

So put the probable cause down to one of three user errors in case anyone else has this same problem in the future:

1. Didn't clean out the dome and sensor well enough after running the track in for a few days
2. Wasn't methodical enough on setting Az locations and looking for variations
3. Confused the setup park and set up home location at some point.

Bottom line - delighted with the domes pointing. So onto hopefully the simpler part of set up.

* * * * * * *

I presume the next task I have will to really nail the Az bearing where the home location is - without relying on my telescope which is several inches away from the domes' pivot point. So sending my scope to point due East at the height of the base of the dome might have it point to a slightly different point to where due East would be from the Dome rotation point centre - about 2-3 inches out - which would translate to 3-4 degrees out. Without access to a marine grade compass can you think of a way I could get the correct bearing for the home sensory position? My best idea at present is to judge the angle of the centre of the dome slit to the dome rotation axis against a really bright star!

Once I have the home setting nailed I will leave park bearing = home bearing for a while (to avoid any possible confusion) then judge how well the scope and the dome align for all bearings in 10 degree increments from 0 at 30 and 60 degree elevations.

Many thanks for everyone's help and suggestions!

Matthew

[g__day (Matthew)]

So the next major discover for me today on the Cyanogen forums. All the bearings they wish you to take:

1) Home position
2) Park position
3) Dome rotation centre versus Mount RA / DEC pivot point offset North / South
4) Dome rotation centre versus Mount RA / DEC pivot point offset East / West

...all of these factor refer to Celestial coordinates - not magnetic or geographic, and of course all are relative to the dome's rotation point not the mount!

For Sydney Celestial South is 11 degree East of magnet South, so this changes my bearing measures by 11 degrees and my off set measurements by an inch or two!

Amazing when there is a choice of measurement systems how Diffraction Limited didn't document which to use, nor include and pictures or schematics to help understand all their terms and assumptions!

Once the weather changes I will see how this affects all my all sky pointing. Stay tuned!

[g__day (Matthew)]

I am making progress! The Dome hits the mark each time in that its reported Az matches the Az marked in 10 degree increments on the track covers, and it it reads the telescope's position correctly.

My challenge now are the geometry / slaving coordinates must be wrong somewhere even though I have checked them all so carefully. At multiple selected Az bearings, taken between 30 - 50 degree elevations the software shows the dome and slot aligned to give a fine view at all times - even thought the dome position varies between fine to out by 5 - 15 degrees! To me this says the geometery calculations are wrong which says the geometry parameters are incorrect somewhere...

I am back to pondering the best methodology to identify my mistake. I start by presuming my N/S (-10") and/or East/West (+4") must be out somewhere (regardless of how many times I have measured and counter-checked things). Is there a reasonable methodology to follow to rapidly identify where the geometry is out?

I will mention my mount is 27" below the dome equator - I presume using -27 in the field height of mount pivot point above dome is fine?

I was thinking of say aiming at elevation 45 degrees then move from Az 0 in 30 degree increments and see what the scope alignment on the screen looks like and measure and report here how much mis-positioning occurs (tomorrow as 3:30am when there is work tomorrow is too late for me). I presume if I know what bearings are best and what are worse this may indicate which parameter is wrong?

[Auster1b (Neil)]

Matthew,
The software will calculate the best alignment of slot and aperture.
Because of your mount offset, this will mean that there will be a difference in the actual dome azimuth to ensure slot alignment. The further the mount offsets the greater the difference to achieve alignment.

So, even though you notice a difference in dome scope azimuth, is the optical axis aligned with the slot ?.

Neil.

[g__day (Matthew)]

No they dont for about half the sky. Around east is best alignment, as you head South to West things get more misaligned. So the closer to the home position the better the alignment of optical axis and dome slot.

I will test Az from 0 to 350 in 10 degree increments and report how far misaligned the dome slot is at say 40 degrees elevation in the next 24 hours.

[g__day (Matthew)]

Latest update - (copied from what I have just posted to Doug on Diffraction Limited's support forums.

I ran a dowel from one join of the dome to the other, then from the dowel's measured mid-point I hung a plumb line of sufficient length so it was adjacent to the mount's pivot point. Next I rotated only the dome through 180 degrees back and forth checking that the plumb didn't move at all - which shows I did indeed have the centre of the dome's rotation identified correctly (see attached pictures). This gave me a simple way to measure the required offsets; giving:

• Offset East 4.25" (later discovered this was the error - East should be entered -4.25 for the Southern hemisphere)
• Offset South -8.25"
• Dome radius 44.9"
• Height of mount pivot point below dome -26.5"
• Height of mount pivot point to scope Optical axis 13.25"
• Home position sensor Az bearing 130 degrees.

With these figures the dome / mount synchronisation is showing I estimate about half the maximum mis-alignment error now (about 10 degrees or about 8") - which unfortunately is enough to cut a few inches off the scope's field of view. I measure misalignment using an small thin dowel rod reaching along my scopes to almost touching the dome - to see how much it protruded past the viewing slot's edge. Every degree of error at the dome circumference is approximately 1.9cm of travel. So 1-3 degrees out would be livable - 8 to 10 needs adjustment.

Testing at Elevation 40 degrees - if I change the bearing so the East / West dome slot matches the mount position well (about a 8 - 10 degree home position adjustment) then the North / South positions of the dome are out by 4" - 8". Vice-versa if I align for North or South by altering only the home position Az bearing - then when I point the mount East or West the Dome is about 8" - 10" out.

Keen to hear anyone suggestions of what I should try next. I am tending to believe its just the geometry solve going askew now - as the mount generally tracks to around 30 arc seconds of target anywhere, and the dome keeps hitting the Az marks within a millimetre or two of slew every time.