Return to the Cloudy Nights Telescope Reviews home pageAstronomics discounts for Cloudy Nights members
Get a Cloudy Nights T-Shirt Submit a Review / Article

Click here if you are having trouble logging into the forums

Privacy Policy | Please read our Terms of Service | Signup and Troubleshooting FAQ | Problems? PM a Red or a Green GuÖ uh, User

Equipment Discussions >> Mounts

Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | (show all)
orlyandico
Post Laureate
*****

Reged: 08/10/09

Loc: Singapore
Re: 10Micron new [Re: Per Frejvall]
      #5781048 - 04/06/13 02:24 AM

Sounds like a good deal for us outside-Europe types. It turns out to be $7500 USD if we remove the MWST (the GM 1000 HPS).

Might be a good choice for me when I need something more than my Mach1.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron new [Re: orlyandico]
      #6004401 - 08/03/13 07:26 AM

Quote:

I'll post back on my impressions once the observatory is finished in April




Ha ha - holiday parks are a law unto themselves. April, May, June passed quietly - always "we have a few jobs to complete first then we do your observatory". Well 3 weeks ago they give the go ahead, two weeks ago they turned up to dig the hole for the pier and pour the concrete. It now at the halfway stage for the concrete curing and robust enough to try out the mount on the pier - briefly - so thats what I did two nights ago. So here is my report:

So the first run was to get used to the new mount, learn how to build the pointing model and nail polar alignment and report back how easy and how accurate the outcome was.

Imaging runs come later once the concrete has fully cured and I've dealt with some shrinkage issues.

So after the HPS1000 was set level on the pier (it has a bubble level) I booted the mount computer*.

After the boot completed, I set up the longitude, latitude and time by hand copying the values off the Losmandy Gemini. I have a StarGPS NX that connects to the Losmandy - would be useful now if I could connect it to the HPS1000!

Once setup, I selected from the handpad screen menu Alignment -> 3-Star Align. My first two attempts to use this resulted in failure.

The procedure starts by offering a list of stars. I picked Altair, the HPS1000 slews close and emits a beep when done, then its over to you to center it in an illuminated reticule eyepiece and press "enter" when you are done (the keypad +> and -> keys are used to change the slew rates at any time the menu selection isn't active - this is a nice touch and very different from the Gemini - much easier in fact).

Next up pick Deneb from the list ... slew, beep, centre and press "enter". My last star was Vega directly overhead. This is were the HPS1000 failed, the slew started but the mount stopped while no where near pointing at Vega and I got no confirming beep. I could hear the motors buzzing like a mosquito (this is OK BTW) and the keypad screen still had it saying it was slewing. Nothing doing.

Not yet proficient with the way out of this I pressed the "shutdown" button on the mount computer, reset the parked position (clutches off, repoint scope by hand, clutches on). Re-booted, cleared any pointing model data via the menu and re-did the 3-Star align using Altair, Deneb and Alberio this time. Same outcome - the mount failed to finish slewing to Alberio (again a star nearly overhead for me).

This time I remembered the "Esc" key on the keypad and was able to backup the menu tree and try a 3-Star again using Altair, Deneb and Alpheratz (over on east side of the sky). This time success!

(later on I did 9 further 3-star aligns and as long as I avoided an overhead star it was successful each time)

At this stage you can use the handpad to get info on how good the pointing model is (as an RMS error value in arc seconds) and what your polar align error is in terms of offset distance (in arc seconds) and position angle or alternatively as Alt/Az off set from the true pole.

A 3-star align is just the start. The next step is to add more stars to the pointing model. From the keypad select 2-Star Refine and pick a star from the list, the HPS1000 will slew to it, halt and beep and you center it and press enter. Keep doing this to build up say a 12 star model from stars spaced widely over the sky - overhead stars now work too (you can build a max of 25 stars into your model). After each additional star you are told what the RMS pointing error is.

Now is a good time to polar align. Pick the polar align procedure from the keypad and pick a star from a list. I picked Altair as it was near the meridian and fairly close to the equator (for no good reason other than the Gemini procedure I used before works best in this region of the sky). This time you must not use the handpad to centre, but use the mount alt/az adjustment knobs to to the centering. Press enter when done.

Thats very nearly it - but not quite. Final step is to 3-star align and refine once more to build a full model - necessary after a physical repointing of the mount.

I experimented for 4 hours to learn how best to do it and how fast it could be done and how good the results are and if further iterations improved things. I performed the procedure 3 times from scratch with the pointing model wiped out at the start.

The optimum solution trading time for accuracy was to 3-star align, add stars to build at least a 12 star model, polar align, then repeat the 3 star align a second time (I added a barlow this time for centring) and again go for at least 12 all-over-the-sky stars and polar align a second time and finally finish off with a third 3 star align and refine to build a final min 12 star model. This took less than 30 minutes and with practice could probably be reduced to 20 minutes

Code:
The results of 3 runs:


RMS error of pointing model over whole sky: 5 (best) to 10 arc seconds (worse)

Error of polar alignment: 16, 17 and 18 arc second offset in the 3 runs I did (2 polar align iterations**).




The PA error in alt/ az was typically 5 arc second in az and ~16 arc secnds in alt.

As an additional test I centered Altair (near equator) and Alkaid (high dec) on the crosshairs (12 mm eyepiece + x2 barlow on 600mm FL scope) and noted that the star remained on the crosshairs for at least 20 minutes.

There are huge differences when comparing these procedures between the HPS1000 and Losmandy GM-8/Gemini. The most striking is with the HPS1000 a slew from E to W (or in reverse) has no initial significant pointing error after the 3 star align. With the GM-8 the intial pointing model is built on one side of the sky first and then when flipping over to the otherside the miss can be a degree or so. Once this initial miss is accomodated into the Gemini model the large pointing error is removed. I suspect this difference must be an absolute encoder thing.

First impressions. Other than the "overhead star" glitch in the 3-Star align, the HPS1000 setup, model building and polar align are extremely easy to do, very accurate with little effort and time and the observed tracking accuracy is superb.

In comparison the best the Gemini can report is to 1 arc minute polar alignment error. The GM-8 unguided tracking accuracy can be occasionally good for 5/6 mins but often only 4 mins as this is dependent on the true polar alignment error.

Note however the if you want to measure the scope/mount orthogonallity error on the HPS1000 you should not use a diagonal (unless its been calibrated). Due to neck craning on a refractor this will certainly slow down the set up. Possibly this measurement can be done after polar alignment has been set up. Then take out the diagonal and redo the 3-star/2-star refine to measure and trim the scopes alignment on the mount.

I got this mount to avoid autoguiding! Looks like it does what it says on the tin.

Next up is testing some imaging runs ....



*BTW someone said earlier on the thread that the "keypad" handles the pointing modelling. The keypad is just a keypad - it is attached to a smallish black aluminium ruggardised industrial strength box fitted with fans running a dedicated version of Linux to which the power, keypad and mount are attached by cables - its the Linux box that is the compute powerhouse.

** Additional iterations did not improve matters - often the PA error increased slightly to ~20 arc seconds only to reduce once again to ~17 arc seconds on a forth pass. It appears that using this approach (iterating 3-star align + refine stars), then once you get below 20 arc seconds error you are effectively done using this method.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Tonk]
      #6006993 - 08/05/13 02:54 AM

Nice write-up!

I have not had any problems with high altitude stars myself... Interesting!

You could try to automate the whole process of building the model, which by the way supports 100 points if you have the newest firmware, and sit back and relax while 100 stars are processed in just a short while.

The software I wrote requires MaximDL and Pinpoint (full version), but I really think that it is worth every penny as you get full automation for the model building part.

I have made the same observations that you have regarding the pier side: it doesn't matter at all. It is always dead on.

As you can see above, I have run me 2000 unguided for one hour with perfectly round stars. I haven't stressed the 1000 to the same level but I figure it will do about the same if loaded to the same level.

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Per Frejvall]
      #6006994 - 08/05/13 02:55 AM

Question: what is your firmware version?

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron new [Re: Per Frejvall]
      #6007051 - 08/05/13 05:10 AM

Quote:

Question: what is your firmware version?




I'll check tonight - its at my "remote" observatory and I'm driving up there to get my first imaging run tonight - weather permitting.

Quote:

I have not had any problems with high altitude stars myself... Interesting!




It was very specifically the 3rd star in the 3-star align step. Never had an issue in the refine steps and did't test an overhead star for the 1st or 2nd of the 3 star align.


Quote:

You could try to automate the whole process of building the model, which by the way supports 100 points if you have the newest firmware, and sit back and relax while 100 stars are processed in just a short while.




How do you "automate" centering a star in the crosshairs? and the alt/az knob turning for PA?? If this involves a CCD I'm out of luck as I only have a Canon 450D and an intervalvometer.

Can you explain the procedure basics and what additional items are needed besides MaximDL/Pinpoint - thanks. Can you avoid MaximDL - I really don't want to use this as I have Images Plus and that enough money to spend on software!!


Anyone got any ideas about the 3-star alignment/overhead star issue? I guess that one to ask 10Micron folk


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Tonk]
      #6010793 - 08/07/13 02:37 AM Attachment (23 downloads)

Automating the model making involves using MaximDL, Pinpoint full and my Model Maker. Any camera Maxim can control works - DSLR or CCD.

Maxim is old school and considered the "Gold Standard" in AP by many. I, for one, really do not like it at all. It is, however, what people who write software normally choose as an interface to the observatory and I walked down the same path when I wrote the Model Maker. I figured that since I was about to run one rig with CCD Autopilot and had ACP on the horizon I needed Maxim anyway. Another interface is CCDSoft (which Bisque discontinued). If ASCOM camera drivers were more common I would probably go straight to the source and just control it myself. Pinpoint is pinpoint, even though there are alternatives. I wanted something that was independant of an internet connection so that left me with... Pinpoint

When you do venture into some degree of automation you will most likely be "forced" to get Maxim...

Below is a screen dump of the Model Maker (which I offer for free at the site in my signature).

I now have at least nautical twilight and have the GM1000 on the balcony pier with a 190MN/ST8300M. I need to run a model anyway so I can look at the high altitude situation next time the skies are clear.

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron new [Re: Per Frejvall]
      #6011116 - 08/07/13 09:40 AM

Hi Per - thanks for this - food for thought.

I'd done my first imqging run and came away very very impressed. This time I only needed to do 3-star + refine, a PA and a final 3-star + 9 refine stars to get a PA error of 18 arc seconds. I selected comet C/2012 F6 which was overhead in Cepheus and tried 8 minute exposures (at ISO 800). 8 to 10 minutes is nearing the maximum I can do for my sky conditions (SQM measure was 20.75 bpsas) to retain good dynamic range.

The result was perfect round stars in every frame for 4 hours! The meridian flip was spot on. Flipping the post meridian flip images and overlaying on the pre flip images showed a very close match up.

So for the FL I like to work at (480mm) I can do a fresh by-hand pointing model/polar align that is more than good enough in <20 mins now. Its still an exciting novelty that thoughts of automation are a little way off.

I forgot to get the version number(s) (doh!) The first one was 1.7 right at the start of booting but there was another one further along the boot sequence with 3 digits (x.y.z) which I have not been able to recall. Which is the one you are interested in.

Oh - one final note - I checked the errors for each star in the pointing model. I had put Vega in (overhead) as the 4th star in the model. It had an error of >200 arc seconds whereas most other stars had error <20 arc seconds. I deleted Vega from the model and gained a slight improvement in the pointing error and each remaining star in the model also gained a very slight improvement. So the interesting thing was that the one overhead star in the model had a high error. The other 11 stars in the model dominated such that the one "rogue" didn't degrade the model by any significance. Still there appears to be some sort of issue with overhead stars that I'm picking up.

I have read somewhere that some goto's have issues pointing with accuracy at the polar region around Polaris while others have a similar problem around the zenith. Is this familiar with anyone?


Right small sidebar for Per ....

I'm very interested in the automation game. What functions of MaximDL and PinPoint are you needing. I'm a long in the tooth software developer (since 1977!) and can see that I could utilize/adapt the open source AstroTortilla in place of PinPoint but what else is needed to use your automation tool (I like the idea of resisting MaximDL and also developing open source for the astro community - something to do when I retire in a few years time )

http://sourceforge.net/projects/astrotortilla/


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Tonk]
      #6011293 - 08/07/13 11:41 AM

Good numbers, good results. Some say it can't be done but they have only read about these mounts, not tried one in the flesh

All polar coordinate systems have difficulties near the pole. When I landed a helicopter on the North Pole in 2004 I had real difficulty in determining what the GPS actually said. It finally shut down a few meters from the pole and had to be restarted...

Anyway... Automation. There are only a few things that I want that Maxim has. It has to do with camera control, and since so few suppliers write ASCOM drivers for their cameras you are stuck with a software package that can do full COM and handle a camera.

Same for automation. It's Maxim or CCDSoft behind it all. Nebulosity doesn't have aproper automation interface, I do not know about the Tortilla stuff. I have considered writing an alternative to ACP and may go ahead with that. Such a project would be written from the start to utilize different camera control platforms. We'll see...

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Per Frejvall]
      #6011336 - 08/07/13 11:57 AM

I found a movie I made in order to show the model optimization that can be done from the latest version of Model Maker. Thinking of you Vega deletion, check this out for cleaning up a model. The film doesn't contain the model run, just the cleanup functions.

/per

http://filer.frejvall.se/Optimize.swf.html


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron new [Re: Per Frejvall]
      #6011404 - 08/07/13 12:31 PM

Quote:

... a movie I made in order to show the model optimization that can be done from the latest version of Model Maker.




Crikey that is a nice piece of work!!! maybe I'm going to have to shake out my piggy bank for maximDL etc.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Tonk]
      #6011606 - 08/07/13 02:28 PM

Thanks!

/p


Post Extras: Print Post   Remind Me!   Notify Moderator  
frolinmod
Carpal Tunnel


Reged: 08/06/10

Loc: Southern California
Re: 10Micron new [Re: Per Frejvall]
      #6012141 - 08/07/13 07:10 PM

Quote:

It's Maxim or CCDSoft behind it all.



In the case of CCDSoft, it has now been retired and replaced with the Camera Add On to TheSkyX.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: frolinmod]
      #6012834 - 08/08/13 02:05 AM

Yes, I noticed that. It was quite good and some people even reported that SBIG images from CCDSoft were of better quality of MaximDL ones shot with the same camera... I wonder what goes

Maybe I should hide in a closet for a few months and write a lean and mean alternative...

/p


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron new [Re: Per Frejvall]
      #6017898 - 08/10/13 04:16 PM

My overhead star issue when doing 3-star align seems to have gone. I suspect it was something to do with the default slew/tracking limits.

Two questions.

1) Do I have to explicitly save the pointing model before turning off the mount controller? I noticed when I turned it back on this morning it had a zero star model! This seems to be counter intuitive - the Losmandy Gemini preserves the model automatically (you have a "warm" vs "cold" restart option meaning keep or wipe the prior model)

2) How do you synchronise an existing model with the sky? The gemini had a nice "sync" menu option. Can't find the equivalent on the 1000HPS

Thanks


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Tonk]
      #6018012 - 08/10/13 05:36 PM

1) Nope. It stays there. Do remember to turn off the mount with the switch, not by cutting power!

2) Set it to "Sync refines OFF" and a sync will just move the existing model to the sync position. If it is ON it will just add another refinement point to the model, which is not what you want.

How's that for straight answers on a Saturday night?

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron new [Re: Per Frejvall]
      #6018528 - 08/11/13 04:09 AM

Quote:

Nope. It stays there. Do remember to turn off the mount with the switch, not by cutting power!




Er yes I do use the switch and only the switch - and the model definately gone when I power up again. For example going to the "Delete star" list you are informed that there are no stars in the list. Before powering off I had 14 stars in the list. So far I've had to rebuild the pointing model from scratch on each reboot (thats 4 times now). Hunting around the menu system I found a place were models are saved, just wondered if I should have used that.

Quote:

Set it to "Sync refines OFF" and a sync will just move the existing model to the sync position.




And my problem is how do you do the "sync" bit. I see the "mode" setting but I see no menu item that is "do a sync now"


Post Extras: Print Post   Remind Me!   Notify Moderator  
Pogo30
journeyman
*****

Reged: 02/23/10

Re: 10Micron new [Re: Tonk]
      #6018545 - 08/11/13 04:55 AM

Quote:


And my problem is how do you do the "sync" bit. I see the "mode" setting but I see no menu item that is "do a sync now"




Just center the star and push and hold the enter button. It will then do the sync.

Regards
Martin


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Tonk]
      #6018547 - 08/11/13 05:00 AM

Hello again!

Number one really puzzles me! That sounds totally wrong. Number two is simple, just slew to any bright star by means of the handset, center it wit hthe arrow keys and then hold enter depressed for a few seconds. This is the "sync here" command you seek!

Can you please go into menu item Settings, then Version and report the firmware version for me?

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Per Frejvall
sage


Reged: 09/28/12

Loc: Saltsjöbaden, Sweden
Re: 10Micron new [Re: Per Frejvall]
      #6018574 - 08/11/13 06:12 AM

New ASCOM driver (6.5.1) is up on my site. Try it out!

/per


Post Extras: Print Post   Remind Me!   Notify Moderator  
Tonk
Postmaster
*****

Reged: 08/19/04

Loc: Leeds, UK, 54N
Re: 10Micron [Re: Per Frejvall]
      #6018587 - 08/11/13 06:50 AM

Thanks Martin and Per re sync'ing - that was a huge help

Nice to see they made it really obvious! . I spent ages searching the menu system for that!

Per the two numbers that I see during the boot up sequence are 1.7 and then a bit later 2.9.8

Can't try out Ascom just yet - still using my simple intervalvometer for camera control and its just the handpad for the mount - no additional computer involved.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | (show all)


Extra information
21 registered and 35 anonymous users are browsing this forum.

Moderator:  Dave M, richard7, bilgebay, iceblaze 

Print Thread

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled


Thread views: 22387

Jump to

CN Forums Home


Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics