Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Meade LX850 silence

This topic has been archived. This means that you cannot reply to this topic.
227 replies to this topic

#126 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 02:45 PM

The three images:

(1) take a relatively short exposure of a bright star field. This will make it possible to measure the exact properties of the optical train exactly as used with a plate solve. The data from this allows scaling of images.

(2) With the mount purposely misaligned to the polar axis, turn off PEC and Starlock off, take a 30 minute image in the exact configuration of (1). This will produce a complete image of several worm revolutions in the star trails.

(3) repeat (2) with PEC turned on.

-Rich


Now let me ask a dumb set of questions.

- Are not the PE plots I generated sufficient for understanding PE error? Is there something wrong with Jason's plots? The current plots look very good.
- What will a picture of the misaligned starfield tell us that a plot will not?
- If I take a 10 minute picture of a star field with everything aligned and operational, and the stars look perfect, isn't this sufficient? Same with a 20 minute picture. We can take all those pictures you mention, however if the image looks wonderful using the LX850 fully operational does it matter if I take a misaligned picture? Isn't this the crux of the issue, that we can take a 10 minute picture using the LX850 system (mount, OTA, and Starlock?) and they look wonderful?

Will I ever use the LX850 mount without Starlock? Highly unlikely. Kind of defeats the purpose of getting the LX850.

#127 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 02:56 PM

Though tracking and PE are not that important for visual, they are for imaging.

The LX850 description seems to indicate that it was designed for imaging, so of course everyone is worried about these factors.

I think that flexure is another important issue that needs to be addressed. Do Starlock suffer from it? That is a real and important question in my book.

So, I know that people do get excited and arguments begin over opinions. This is why real data from several users would help.
Blueman



I would agree tracking and PE are important for imaging. Isn't also the OTA? And in the case of the LX850, isn't Starlock critical? It does handle the PE training, HPP, and guiding. I would think if I had a great mount and a poor Starlock system that would be good to know.

I agree about flexure (Meade has a section about it in their user manual). Key there will be not only the telescope, but did the user take care of issue that can cause this. The 5, 10 and 20 minute exposures should show this effect. I think Meade was wise providing Starlock with a secure mount on the OTA (this was not done on the LX800). The mirror locking feature is also very important. And yes, I will make sure the OTA is properly attached to the OTA and the camera is secured to the OTA to further minimize flexure.

#128 Starhawk

Starhawk

    Space Ranger

  • *****
  • Posts: 6500
  • Joined: 16 Sep 2008

Posted 31 May 2013 - 06:01 PM

I'm sorry- what thread did you post PE plots in? I must have missed it. The images I mention are a really easy way to get plots without having to buy software, which you had complained about in a previous post. If you could point to where your plots are, or if you are willing to repost them, that's faster than taking images of the meridian/ equator intersection with a misaligned mount to get drive images.

Thanks,

-Rich

#129 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 06:29 PM

I'm sorry- what thread did you post PE plots in? I must have missed it. The images I mention are a really easy way to get plots without having to buy software, which you had complained about in a previous post. If you could point to where your plots are, or if you are willing to repost them, that's faster than taking images of the meridian/ equator intersection with a misaligned mount to get drive images.

Thanks,

-Rich


Here is the plot from the PE Utility for the LX850.

Attached Thumbnails

  • 5894909-Screen Shot 2013-05-26 at 1.28.07 AM.png


#130 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 6925
  • Joined: 20 Jul 2007

Posted 31 May 2013 - 07:44 PM

I'm sorry- what thread did you post PE plots in? I must have missed it. The images I mention are a really easy way to get plots without having to buy software, which you had complained about in a previous post. If you could point to where your plots are, or if you are willing to repost them, that's faster than taking images of the meridian/ equator intersection with a misaligned mount to get drive images.

Thanks,

-Rich


Here is the plot from the PE Utility for the LX850.


This is showing raw uncorrected PE? I mean, no PEC in place?
Blueman

#131 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 08:44 PM


This is showing raw uncorrected PE? I mean, no PEC in place?
Blueman


This plot is of the raw, uncorrected PE. No PEC is in place. In other words this is the PE without Starlock, PEC, etc.

Not too bad out of the crate.

#132 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 09:17 PM

Gday Andrew

This plot is of the raw, uncorrected PE. No PEC is in place. In other words this is the PE without Starlock, PEC, etc.



Just being a bit pedantic here :cool:,
That data is the current PEC "model" stored in the scopes Flash memory, and the scope manipulates/averages the raw data before storing,
Ie it may or may not be a close representation of the true PE, ( hence the requests for independent raw data from secondary sources to cross correlate it ).

The best way to do this is to use something like PEMPro or PHD to gather three or more continuous cycles of the worm on a single overlayed plot, such that data magnitude and repeatability can be clearly seen. No tweaking, no smoothing, no averaging, just a plot of simple PEC off unguided tracking.

Andrew Johansen Melbourne Australia

#133 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 09:36 PM

Gday, back at you Andrew!

So question #1: Are there no value in the results from the Starlock PEC Utility?

Question #2: How far off do you think the error is from the true PE? Even if averaging, is the error significant? For example, even if averaging the value in arcseconds should not be too far off. Looking at the slopes of the curves, things can't be too far off. For example, it would be tough to jump from 0.5 arcseconds, to lets say 10 arcseconds, and then down to 3 arc seconds, yet show a reasonable sinusoidal curve the way it is. Mathematically the curve seems close to the real deal vs. massively swinging all over.

I am loath to purchase PEMPro. I see PHd guiding for the Mac, but I don't have a camera to work with it. So the only other method is the photo method?

I have to believe there is some value in the Starlock PEC Utility.

#134 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 6925
  • Joined: 20 Jul 2007

Posted 31 May 2013 - 09:45 PM

Gday Andrew

This plot is of the raw, uncorrected PE. No PEC is in place. In other words this is the PE without Starlock, PEC, etc.



Just being a bit pedantic here :cool:,
That data is the current PEC "model" stored in the scopes Flash memory, and the scope manipulates/averages the raw data before storing,
Ie it may or may not be a close representation of the true PE, ( hence the requests for independent raw data from secondary sources to cross correlate it ).

The best way to do this is to use something like PEMPro or PHD to gather three or more continuous cycles of the worm on a single overlayed plot, such that data magnitude and repeatability can be clearly seen. No tweaking, no smoothing, no averaging, just a plot of simple PEC off unguided tracking.
Andrew Johansen Melbourne Australia


This is what I suspected. If I were to show the stored data in my AP900 it would show less than 1", but that is not the actual PE for the mount.

This is why I think people are getting a bit confused. They are not getting real data to post.

Download Pempro and take advantage of its 30 day trial. You can then get the real data on your mount.

It is the same thing when I read that the Starlock will give you guiding within 1 arc second. That sounds fine until you figure that if you are imaging with the 12" or 14" SCT you just might be imaging at less than 1 arc second image scale. Then you are looking at a star that could be elogated as much as double or even more. All that this is saying is that it guides within plus or minus 1" PE. This just might not look that good if that is the case.
Blueman

#135 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 09:54 PM

Gday Andrew

Are there no value in the results from the Starlock PEC Utility?



Dont know yet :grin:, thats the whole question in question.
Ie PEC isnt "stored" in the scope as PEC, it is stored as motor speed adjustments.
This data is asymmetric. The curve presented in PEC tool is a theoretical representation of the PEC "calculated" from this data, so it may/may not be good.
The scope and PEC mechanism is so new, no one knows how good/bad this data is.

Doing a secondary plot with raw data via a different mechanism, will allow the two datasets to be cross correlated and hence we get a better idea of the fit.

Looking at the slopes of the curves, things can't be too far off.


I dont understand how you can say something is right by "looking at it" without a datum point.
You could do a PEC ON unguided plot and see how well it tracks, but at the end of the day, you need to test the results against a datum.
Currently, we have seen no datum.

I have to believe there is some value in the Starlock PEC Utility.



And it may be spot on, but its not validated.

I had to believe that Meades serial commands would work as per the book, but then i tested and they dont.
The :gps# bug ( which is still in 1.1g ) will create havoc with your scope if used.
Ie until its tested, i dont believe.

Andrew Johansen Melbourne Australia

#136 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 09:56 PM

Blueman,

Are you sure that the AP900 when showing less than 1" is not a corrected term?

I am very curious on the Starlock PEC Utility from Meade. I will give them a call next week to get the actual inside skinny on all the details of the utility.

Jason - any insight on this tool and the data?

So the claim seems to be (a) the Starlock is not showing real PE data, and (b) the chart is way off from the real data? Let's assume a 20% off the actual. This would put the Peak to Peak still around 7.2 arcseconds. If there is a massive error, or the plot is way off base then this is a rather silly and useless tool.

Well I can't do much until I get home as I am still 1200 miles away. I will look into the 30 day free use. And gosh, another windoze app. :bangbangbang:

#137 EFT

EFT

    Vendor - Deep Space Products

  • *****
  • Vendors
  • Posts: 5071
  • Joined: 07 May 2007

Posted 31 May 2013 - 10:06 PM

Here is an example of what they are talking about in regards to an image that can be used to determine native PE. By using a double star of know separation, you can measure the magnitude of the error as well as get an idea of the shape of the error without the use of any software.

Posted Image

#138 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 10:06 PM

Gday Andrew

Are there no value in the results from the Starlock PEC Utility?



Dont know yet :grin:, thats the whole question in question.
Ie PEC isnt "stored" in the scope as PEC, it is stored as motor speed adjustments.
This data is asymmetric. The curve presented in PEC tool is a theoretical representation of the PEC "calculated" from this data, so it may/may not be good.
The scope and PEC mechanism is so new, no one knows how good/bad this data is.

Doing a secondary plot with raw data via a different mechanism, will allow the two datasets to be cross correlated and hence we get a better idea of the fit.

Looking at the slopes of the curves, things can't be too far off.


I dont understand how you can say something is right by "looking at it" without a datum point.
You could do a PEC ON unguided plot and see how well it tracks, but at the end of the day, you need to test the results against a datum.
Currently, we have seen no datum.

I have to believe there is some value in the Starlock PEC Utility.



And it may be spot on, but its not validated.

I had to believe that Meades serial commands would work as per the book, but then i tested and they dont.
The :gps# bug ( which is still in 1.1g ) will create havoc with your scope if used.
Ie until its tested, i dont believe.

Andrew Johansen Melbourne Australia



Gday (or is it Good Morning?) Andrew,

Yes, I saw the values in the CSV file is stored as motor speed adjustments - I was hoping to see the units differently.

And I understand what you are saying about cross correlating with two different sources.


You other point, if I understand you correctly, since there is the :gps# bug there may be other bugs in the software, which in the end may impact the results.

The :gps# bug doesn't seem to be impacting my LX850. Can you again tell me what it is.

However my one concern about the bug claim - I see LOTS of software with bugs (Windows? iOS? Android? Heck I am always finding bugs in flight software!). Still, the products and tools are being used. On the flip side the issue is the LX850 is so new, the claim is major errors have not been flushed out.

Still, in the end the proof in the pudding is will the LX850 take pretty pictures.

What a time to have this discussion and to be far away from my baby! And in an area expected to be hit with tornados in the next 24 hours!

#139 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 10:18 PM

Ed (and Rich),

Well it looks like the picture method is another way I will try. Ed, I will use the same target you picked.

My alignment does not need to be far off from the polar axis to pull this image off, does it?

Does it really need to be 30 minutes? Wouldn't 20 or so minutes be sufficient?

#140 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 10:27 PM

Gday Andrew

So the claim seems to be (a) the Starlock is not showing real PE data, and (b) the chart is way off from the real data?



Not at all.
Starlocks PE utility appears to be showing true PE,
"as calculated from the motor speed deltas".
But again, it is a "model", it is not raw data.

The question is how close a fit to reality is the data.
As i said, it may be spot on, but it may not.
We just dont know yet.

Andrew Johansen Melbourne Australia

#141 EFT

EFT

    Vendor - Deep Space Products

  • *****
  • Vendors
  • Posts: 5071
  • Joined: 07 May 2007

Posted 31 May 2013 - 10:29 PM

Ed (and Rich),

Well it looks like the picture method is another way I will try. Ed, I will use the same target you picked.

My alignment does not need to be far off from the polar axis to pull this image off, does it?

Does it really need to be 30 minutes? Wouldn't 20 or so minutes be sufficient?


I was provided this image in order to explain how PE was being measured on a particular mount and did not take it myself (but can use it). I'm not an expert at this, but this graphic helped me to understand the method and goal so I figured it would be helpful here. The method is something that you will have to play with a little bit. You want the off-set to be sufficient to show the "curve" and you want the time long enough to show a few worm cycles. If the offset is insufficient then the "curve" will be all jammed together and if the exposure isn't long enough, then you won't get a good feel for the actual periodic error (i.e., repeatable error) vs. the random error.

#142 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 10:41 PM

Gday Andrew

since there is the :gps# bug there may be other bugs in the software, which in the end may impact the results.



Geez, where to begin.
There are so many unresolved bugs, its safer to assume there is a bug vs not.
These bugs are in the firmware, the ASU Updater, the old PEC tool etc etc.
I dont assume anything works till tested.

The :gps# bug doesn't seem to be impacting my LX850. Can you again tell me what it is.



There is a serial command ( :gps# ) that asks the scope to send a GPS string back to the PC.
If you send this command, the handbox isnt blocked properly, and hence picks bytes out of the GPS data stream at a set interval and processes them as tho you had pressed a key on the handbox and held it down.
This can cause some really interesting results, including runaway slews.
You may not use :gps# and hence to you its not a problem, but my PEC editor checks it every time you connect, and hence can cause runaways if anyone uses my app. I dont like that.

Still, the products and tools are being used.



Again, until a problem hits you specifically, it's never a problem.
If you go look at the Yahoo LX200GPS moderated site archives, you will see lots of people who get caught over time doing something specific.

Andrew Johansen Melbourne Australia

#143 galaxy_jason

galaxy_jason

    Vendor

  • -----
  • Vendors
  • Posts: 273
  • Joined: 22 May 2007

Posted 31 May 2013 - 10:41 PM

Andrew I already posted a comparison of the starlock data to pempro. Apparently it got deleted. The PEC tool data with the latest version is almost identical to pempro

#144 blueman

blueman

    Photon Catcher

  • *****
  • Posts: 6925
  • Joined: 20 Jul 2007

Posted 31 May 2013 - 10:50 PM

Blueman,

Are you sure that the AP900 when showing less than 1" is not a corrected term?

I am very curious on the Starlock PEC Utility from Meade. I will give them a call next week to get the actual inside skinny on all the details of the utility.

Jason - any insight on this tool and the data?

So the claim seems to be (a) the Starlock is not showing real PE data, and (b) the chart is way off from the real data? Let's assume a 20% off the actual. This would put the Peak to Peak still around 7.2 arcseconds. If there is a massive error, or the plot is way off base then this is a rather silly and useless tool.

Well I can't do much until I get home as I am still 1200 miles away. I will look into the 30 day free use. And gosh, another windoze app. :bangbangbang:


You misunderstand, I am saying that anything stored in the mount would be PEC, or corrected PE and has nothing to do with real raw PE.

#145 galaxy_jason

galaxy_jason

    Vendor

  • -----
  • Vendors
  • Posts: 273
  • Joined: 22 May 2007

Posted 31 May 2013 - 11:07 PM

PEC tool is taking the PEC table numbers and displaying a wave form. I believe it is very close to actual raw PEC

Attached Thumbnails

  • 5895311-SL_PEMPro_comare.jpg


#146 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 11:11 PM

Gday Jason

Andrew I already posted a comparison of the starlock data to pempro.



I remember seeing something, but ( IIRC ) it was a calculated curve, not a raw error plot.
Plus it sounds like you have a really good mount :jump:

I personally want to see raw data from multiple sources to also see how consistent it is.
After spending 8 years looking at LX200 data plots i know the one consistent point is nothing was consistent.

Andrew Johansen Melbourne Australia

#147 galaxy_jason

galaxy_jason

    Vendor

  • -----
  • Vendors
  • Posts: 273
  • Joined: 22 May 2007

Posted 31 May 2013 - 11:22 PM

Agreed, but many here doubt the engineering concept. Meade has shown it can be done.

#148 Spacetravelerx

Spacetravelerx

    Vanguard

  • *****
  • Vendor Affiliate
  • Posts: 2495
  • Joined: 23 Dec 2012

Posted 31 May 2013 - 11:32 PM

Jason,

That is what I thought, the new utility data from Meade is matching up with PEMPro. Too bad your data got deleted.

So near as I can tell, there is some doubt even though you presented some clear data, and folks are still concerned my plot from the Starlock PEC Utility may have significant errors.

Next steps:
-- Compare my Starlock PEC Utility plots with PEMPRo. This will give results from 2 users. Ahopp and others, you are next! ;)
-- Take a picture for another view of the PE data with Starlock to be even more sure of the results.

However, if we want to really do this right we will need a large sample to see what the standard deviation is from the average of results to see the overall trends. Understand, I have no problem adding to the pool of knowledge and my experience using the LX850. However I am 100% convinced my results will still not sway many (2 data points). Of course if everyone holds off purchasing the LX850 until they are convinced the average PP PE is under 10 arcseconds we may never get a good data set. I will say, due to the critical eye of some folks here I have to believe several folks will be scared to even do this process. Some will want to just play with the telescope and not bother.

I am convinced if I post a really purty picture of M51, it will get many others excited outside of the die hard AP pool. Heck, a picture of the Moon still gets lots of likes from my FB followers, lol.

#149 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 11:45 PM

Gday Jason

Agreed, but many here doubt the engineering concept.



I have no problems with the concept, i just wonder if its implemented correctly, and how consistent "mechanically" the different setups will be.
That requires time and a range of datasets.

Ie Even to this day, in the LX200s, PEC plays back inverted if you are in the southern hemisphere. The model is correct but the implementation doesnt work.

Andrew Johansen Melbourne Australia

#150 OzAndrewJ

OzAndrewJ

    Mercury-Atlas

  • *****
  • Posts: 2731
  • Joined: 30 Nov 2010

Posted 31 May 2013 - 11:50 PM

Gday Andrew

and folks are still concerned my plot from the Starlock PEC Utility may have significant errors.



Where has that been said? ie "significant" errors.

Andrew Johansen Melbourne Australia


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics