285 replies to this topic

[hjd1964]

[One more thing to report to you: First I went to a star in Gemini and found the scope is roughly tracking without guiding. Then I went to M42. Now stars on the MetaGuide screen were slipping to EAST as if it was not tracking. So I checked the turning of the RA pulley; sure enough it was turning to correct direction. At this time I calibrated and guided as I described above.]

 

Did you set the backlash values in OnStep to match your mount?  The slipping behavior described above is typical of not having the RA backlash value set except it happens when doing a goto from west to east.  There are values for the amount of RA/Dec backlash (arc-sec) in the ASCOM driver setup window (and in the Android App.)

 

[I just did angular analysis on the size of elongation to measure angular movement in RA and DEC direction during the 30 sec of exposure.

The best sub frame shows RA~30" and DEC~30" and the worst sub shows RD~30" and DC~60".  These are very crude estimate of angular movements during the worm period.  I will measure again when the autoguiding improves.]

 

Is that guided or unguided?

[photo444]

Good morning, Howard.

1. I didn't tune the back lash yet.  When guide rate was 0.5x the slipping was happening for a long time like 10s of minutes. And the scope was very quickly responding to the Android control pad.  So I thought RA backlash was good. DEC control response was pretty brisk.  I have a plan to tune the backlash in back of my mind.  

2. The values are from the shots I thought guided. But I had only a couple of shots from a relatively satisfied guiding whish lasted only a few minute before the cloud covering.  The values are similar to the unguided ones!

 

I am going to identify and tune out the cone error during day time as much as possible.  By doing so I may better understand the mount behavior. I may adjust the polar axis almost horizontal to aim at an distant terrestrial object, loose RA lock, and adjust DEC using the control pad.   I hope the RA lock does not give extra slack.  Once the cone error is removed, then I have better confidence of my park position, and polar alignment may be easier.

[hjd1964]

Ok, I was curious about having an unguided baseline to compare against.  We have some good weather coming so I'll see about Metaguide here again so I have a bit more experience.

 

Yes, I would expect the RA backlash to be no more than a minute or two worth of tracking. So you attribute the slipping to the clutches?

 

I looked at Metaguide again.  I was thinking that it sets OnStep's rate from those values... But, I bet that was wrong and it actually uses those values to scale its guide response in which case 1.0 would be correct if OnStep is similarly set for 1X guiding ("Slow".)  My ASCOM driver always reports 0.5X and should really report the current rate setting, I'll fix this.

 

I've read that some people mark the ground where the tripod legs are once polar aligned precisely.  That way you can just sit the mount back in the same position again and be reasonably well aligned.  I have a desk right next to a door to my deck,  so I just pick the mount/scope up and carry it out about ten feet and sit it down.  I don't have marks, but the deck is reasonably level so it gets surprisingly close sometimes without even trying.

[pbunn]

Here are some photos of the mount. John Davis did a great job. He made four or five of these before quitting and going to school. I think he transferred the design rights to a company but they never made any more.

 

I think the mount can be made into a nice mount with the OnStep system and some larger motors.

Attached Thumbnails

• 
• 
• 

Edited by pbunn, 29 December 2014 - 09:42 AM.

[hjd1964]

How many teeth on the worm-wheels?  Hard to get a sense of scale, how much can it carry approximately?

[photo444]

Ok, I was curious about having an unguided baseline to compare against.  We have some good weather coming so I'll see about Metaguide here again so I have a bit more experience.

I know some good weathers coming but COLD. I don't like cold.

 

Yes, I would expect the RA backlash to be no more than a minute or two worth of tracking. So you attribute the slipping to the clutches?

It's good to know the upper limit of the RA backlash to be less than a arc minute or two. When I slewing slowly with Android pad I think the RA backlash was about a minute, that is OK.  The problem of clutch was when I mounted heave telescope of 12 pounds.  With the equipment I showed in the pictures seems light enough for the clutch.  Still I have to be careful whenever I loosen and tighten the lock because the scope tends to move when I tighten.

 

I looked at Metaguide again.  I was thinking that it sets OnStep's rate from those values... But, I bet that was wrong and it actually uses those values to scale its guide response in which case 1.0 would be correct if OnStep is similarly set for 1X guiding ("Slow".)  My ASCOM driver always reports 0.5X and should really report the current rate setting, I'll fix this.

Thank God that the creator of the software is helping me!  All I know now from last night's experience is that when I increased the guiding rate the response was increased as well.  I will do some experiments next time.

 

I've read that some people mark the ground where the tripod legs are once polar aligned precisely.  That way you can just sit the mount back in the same position again and be reasonably well aligned.  I have a desk right next to a door to my deck,  so I just pick the mount/scope up and carry it out about ten feet and sit it down.  I don't have marks, but the deck is reasonably level so it gets surprisingly close sometimes without even trying.

That is great suggestion!  I have been thinking about making on the ground but it's lazy me.  I also found wobbling of the tripod as well.  I have an old basketball pole at the edge of my driveway, some day I will chop it down and make it a pier.  As I said before finding my park position accurately is the first step for me.  As the old saying says I may have to check every wit that is involved in this astrophtography.

 

Thanks for your helps!

[pbunn]

I don't remember. I have heard that the worm is the weak point. I'd probably only attempt it as a visual mount - but if I remember correctly - he did some decent photos with the mount during design and testing. It is as large or larger than my G11.

Edited by pbunn, 29 December 2014 - 04:40 PM.

[hjd1964]

I made the software changes here, OnStep/ASCOM now report the active guide rate.  I was replying in "degrees/second" units from the ASCOM driver but switched to "sidereal rate" units - for some reason, the ASCOM developers choose to not standardize on a unit for this parameter so it's a guess.

 

In Metaguide I see no effect either way.  It still stays at a default guide rate of 0.5 when I connect to OnStep's ASCOM driver... I looked at the logs in POTH and it says MG queries CanSetGuideRates (OnStep still doesn't support setting guide rates and so answers False.)  Guess my default of 0.5 just happened to agree with MG's default of 0.5, and MG isn't pulling its value from OnStep.  If I ever go further and implement setting guide rates through ASCOM then I expect it'll work.

 

This isn't a show-stopper, just a minor inconvenience since we can set the value manually.

[photo444]

I am reading the manual of the MetaGuide. It says "For ASCOM, enter desired values and the true values of the mount will be determined automatically." about RA/DEC Rate on page 26.  This explains the use of 1.5x rate I used last night, maybe. 

[hjd1964]

I know that my ASCOM telescope driver you're using doesn't handle setting or getting guide rates directly.  I use SerialCommand in Sky Planetarium to send the :R1# :R2# etc. LX200 commands to set the rates.

 

I just finished adding complete support for setting/getting rates using GuideRateRA and GuideRateDec.  OnStep's ASCOM driver works just like the ASCOM Telescope Simulator for these commands now. Sky Planetarium will set guide rates correctly in both drivers too.  However, MetaGuide still doesn't set the rate, but it does read it back correctly.  I will investigate further.

[photo444]

How about PHD2? Does it have more flexibility of the guide rates?  

[hjd1964]

How about PHD2? Does it have more flexibility of the guide rates?  

I see no interface for setting guide rates in PHD2... but this is splitting hairs.  Guide rates can be set from the Android App, my "Sky Planetarium", or just use the default of 1X and forget about it.  Pretty sure a LX200 protocol connection to CdC allows for a few basic guide rate settings there too.

 

An idea for a possible feature for OnStep that would simplify this somewhat would be to have a separate guide rate for pulse-guiding that gets saved to EEPROM.  This way you don't have to set the rate back down after centering objects, etc. too.  Perhaps have R0,R1,R2 (0.25,0.5,1X) set the pulse-guide rate in addition to the normal rate.  So for centering objects just don't go below R3 and the pulse-guide setting will remain undisturbed (these rates are too slow for centering anyway.)

[photo444]

Guide rates, probably, are not an issue once the guiding is established. My common sense tells R0, R1 and R2 are enough to guide. Storing in EEPROM may be a good idea.

 

I am looking for tonight, they said it will be cleared, so I will bundle up myself for the cold and do something in cold.  I have been setting up in mind many thought experiments for what to do tonight.

 

As I was reviewing what I did the other night, there have been several mistakes in entering scope (guide) info.  The screen shot showed irrelevant  numbers like f# and focal lengths and so on.  There are so many things to worry about for a new comers to MetaGuide.  I will try to iron them out as much as possible.  I am machining adapters now to use different focal lengths for guiding; I will use a zoom (50-200mm or 400mm fixed) to experiment with wider choices.  This is my plan.

1. Null out cone error at the PARK position: Watching the rotating image out of the guide scope through LN300 (real time) while rotating RA with hand, see if there is a center of rotation on the screen: Adjust DEC to center the center of rotation to the sceen center.  

2. Adjust mount (latitude and azimuth) to get the NCP on the screen center.

3. Mark this position as my PARK position and set the setting circle to 90 degrees.

4. Do Star Alignmet using GOTO to further fine tune the polar alignment.

 

This is my plan that needs a lot of prayers.

Please let me know your wisdom and suggestions.

[hjd1964]

1. Cone error is due to a non-perpendicular condition between the Dec axis and the Optical axis.  This requires shimming or other adjustment to correct, without doing this it's quite possible that there is NO Declination orientation which will bring the center of rotation close to the screen center. What you're calling the PARK position i would refer to as the HOME position (counterweight down) and anytime you start an align, OnStep assumes your scope is at the home position.

2. A reasonable plan.

3. Ok.

4. Doing a One/Two/Three Star Align in OnStep doesn't have any effect on tracking accuracy.  A full three star align does provide information about where the pole is, and that is available in Sky Planetarium's Telescope dialog.  To the right (PZ) and below (PE) the pole are negative (values are all in arc-seconds).  If internet connected, you could also send an image up to Astrometry.net and plate-solve it to see where the pole is.  Just plate solving any image from the camera/lens you intend to use will give you the field size which might be a useful bit of information to have at hand.

 

Looks like I'll be testing MetaGuide tonight too.

 

Best of luck and stay warm

Edited by hjd1964, 30 December 2014 - 03:10 PM.

[photo444]

I was wrong about the cone error. Thanks for correcting. 

I saw "Plate Solving" when I was scheming through the Sky Planetarium.  It may be a good chance to use it.  There are so many things to learn and get used to.

 

Have fun tonight 

[hjd1964]

Astrometry.net's web interface provides graphic overlays which show coordinate grids so you can find the pole.  You can just upload an image to them on the web for this purpose.

 

Sky Planetarium's plate solving is more for helping with Goto accuracy (it connects to Astrometry.net's web based JSON API so you need an internet connection.)  Naturally, this is much more relevant if your camera is directly supported.  To see it work, I recommend setting up a test during daylight.  One can then use the ASCOM telescope "Simulator" and an image file camera (telescope sim pointed near where the object imaged is in the sky) to do some trial solves and make sure it's working for you.  The "Goto corrected location" setting is normally the one I use (then with every goto you do a plate-solve is done and another goto takes place), or just force a solve to see the marker come up on the map.

Edited by hjd1964, 31 December 2014 - 02:14 PM.

[photo444]

That is powerful.  The other night when I first goto a star I wasn't sure where the scope was pointing.  I think it is a perfect solution to get oriented when disoriented.  Thanks a lot!

[photo444]

I have been in a machine shop making camera adapter for the LN300 to any NX camera lens.  The first picture shows a 1/2" C-mount extention tube, a plastic donut coupler, and a body cap for Samsung NX camera (a hole was cut to accept the extention tube for focusing.  The secon picture is an assembled adapter. Samsung NX leses are very sharp and light and cheap. I will use it with 50-200mm zoom, first. I am ready for tonight's session. 

Attached Thumbnails

• 
• 

[hjd1964]

I checked MetaGuide out again last night with my Tak EM10B/OnStep.  The guide-scope is a 50mm finder with my Lumenera LU070M camera (about 1/8 sec. exposures.)  I didn't update the mount's OnStep with the recent changes to enable getting/setting guide rates. Quick calibration went without a hitch.  Guiding appeared to be working so I ran a series of 5x30s exposures both guided and unguided near Messier 1.  After reviewing the data this morning I have to say that the results are inconclusive, guided and unguided look too similar.  I'll try again with more aggressive settings and a shorter guide period.  MG was also reporting a rather large seeing fwhm which is puzzling, perhaps my focus is poor.  I need to read the MG manual!  Comments/suggestions are welcome.

 

The resulting image stack (760mm FL) at about 2.2"/px, no registration, focus was poor, stretched horizontally since the CCD pixels aren't square, and finally E/W is vertical:

 

Here's what PHD2 looked like, notice what I assume is reduction gear noise... it's pretty bad.  PHD2 supports my ASCOM camera driver for the LU070M which enables long exposures so I can actually find guide stars near the object of interest:

 

The resulting image stack (760mm FL) at about 2.2"/px, 10x60s, no registration, stretched horizontally since the CCD pixels aren't square, and finally E/W is vertical:

Edited by hjd1964, 31 December 2014 - 11:37 AM.

[Geo.]

 (I should be be able to repair the RA lock without too much work)

I've got some SP parts. PM me if they'll help. The SP is pretty rugged. Mainly you see bent worm shafts (which can be straightened with care  ) , broken saddles and clutch levers or missing parts. This looks like a good setup for an SP. Here's some motor bracket ideas. 

Attached Thumbnails

• 
• 
• 

[hjd1964]

I had better luck early this morning with MetaGuide, now confident that OnStep works with it.  I increased the aggressiveness (0.8,) dropped the guide period to (0.75s,) and discovered that integration needed to be turned on... so I could even find guide stars near an object.  I imaged M64 with my DSI-II Pro like before (2.2"/px) except 5x90s, registration, stretched horizontally since the CCD pixels aren't square, and again E/W is vertical.  I hacked it up a little messing around with dark frames.  No way can this EM10 make anything like a round star at over seven minutes w/o guiding:

Edited by hjd1964, 02 January 2015 - 04:42 AM.

[photo444]

That M64 look great! It surely looks like a black eye!

BTW Happy New Year again! I had a couple of hours of guiding last night as well.

 

When I made the calibration work, GuideRates were 2(RA) and 1.5(DEC), rather big numbers, with aggression of 0.8 and 0.9.  I will show you the result soon. There still are many things to be straighten out on my end.  Guiding showed typical under damped feedback problem due to over correction, but it was  a step improvement.  Star patterns look like "Lissajous Pattern" of 30sec samplings between RA and DEC oscillations. I am currently studying the log file examining collected data.

 

After having guided I have more confidence to tune it to a critically damped case to have clean star shape. I am going back to it tonight.

[photo444]

After fighting several nights, the mount is beginning to behave.  I had hard to time to calibrate, so after fooling around with guide rates and aggressiveness, I was able to calibrate at RA rate = 1.0, DEC rate = 1.0, RA aggression = 0.8, and DEC aggression = 0.9. And I have to fight with poor polar alignment which consumed much of my time, you will see the poor alignment which is evident on the plots below.

 

The first picture is a screen shot when I got the calibration and guiding working in the beginning: A lot of oscillations and overshooting went on. The next is PEC plot which was recorded by MetaGuide (It generates a lot of useful data that challenges me to digest) showing gradual improvement of the behavior of the mount as I was changing guiding parameters.  The final values are RA and DEC guide rates of 0.5x, and surprisingly enough both aggression are 0.04, yes 0.04. Even a change of 0.01 made substantial response!  I also had to increase the frame rate of 16/s.  Still the poor polar alignment gives a big headache, a remaining home work.  My plan is to use longer focal length for the guide scope, I still can sacrifice the f-number because of the high sensitivity of LN300. 

 

The third picture is a plots when I was using above set of parameters.  It still is not idea but it has been improved a order of magnitude.

Attached Thumbnails

• 
• 
• 

[photo444]

Here are shots for comparison. Stars are much smaller but could be better.  Both were taken with 400mm F5.6 NX300M ISO1600 30sec.  I am going to stack good ones to compare with unguided best shots in the past.  I was able to slew to several different objects because of the GOTO function.

Attached Thumbnails

• 
• 

[hjd1964]

You're making progress, the second image with some background subtraction would be a keeper in my book.

 

Are you sure you had OnStep in 1X guide mode?  Needing very low aggressiveness numbers in MG could be caused by OnStep using fast guide rates.  When using the Telescope Pad in Sky Planetarium to center objects be sure to switch back to "Slow" mode (which is 1X, assuming that was the mode you were in when you calibrated MG) before attempting to guide.