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 Gu… uh, User

Equipment Discussions >> Mounts

Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | (show all)
Ray Gralak
Vendor (PEMPro)


Reged: 04/19/08

Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886275 - 10/28/11 11:52 PM

Hi Jerry,

A couple additions to this...
Quote:


To answer your question Gale, you can generally layer on correction systems, but you have to understand how they work to understand what the interactions might be. It is good to understand the time constants or frequencies each technology is designed to correct errors in. For example, The TDM is correcting errors in the range of 0.2 to 1 second time interval, the PEC is only designed to correct long term worm error in the 480 second range. When you incorporate autoguiding into the picture, I would set it to correct for errors (mainly DEC drift due to innaccurate polar alignment) in the 10-60 second range. Finally, Adaptive optics will correct for errors in the <0.2 second range down to how fast you can get your camera to make exposures.




By the Nyquist theorem you can only correct to 1/2 the frequency. So if sampling is occurring at 5 Hz, the maximum frequency that can be corrected is 2.5Hz.

Also, some PEC systems have corrections that occur 3-5 times a second, so they are capable of correcting frequencies in the same frequency range. But despite what PEC is capable of correcting you are correct that the purpose is to correct the slower worm error and possibly gear errors.

-Ray


Post Extras: Print Post   Remind Me!   Notify Moderator  
gdd
Pooh-Bah


Reged: 11/23/05

Loc: N Seattle suburb, WA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Ray Gralak]
      #4886304 - 10/29/11 12:26 AM

Quote:

To answer your question Gale, you can generally layer on correction systems, but you have to understand how they work to understand what the interactions might be. It is good to understand the time constants or frequencies each technology is designed to correct errors in. For example, The TDM is correcting errors in the range of 0.2 to 1 second time interval, the PEC is only designed to correct long term worm error in the 480 second range.





PEC is the only one of the correction methods able to predict an error and begin the correction in time to prevent it. As Ray points out it can correct errors much shorter than the 480 second range, for example gear box errors that are harmonic with the worm period. The problem is most modest mounts have non-harmonic errors. So PEC can take the already small error of a precision mount and reduce it by a larger percentage than the less precise mount.

The TDM corrects at a much higher frequency, so maybe the lack of a predicting capability does not matter. Perhaps the biggest advantage of the encoder technology is that it will improve over time and become cheaper because it can be adapted to many other products in addition to astronmical tracking mounts.

So given the interactions with the mount, if TDM is installed on both a precision mount and lower tier mount of the same weight class, will both track equally well unguided? Will both guide equally well? I would think the precision mount would have an edge - but how much I don't know.

Gale

Edited by gdd (10/29/11 12:37 AM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jerry Hubbell
member
*****

Reged: 02/16/09

Loc: Locust Grove, VA, USA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Ray Gralak]
      #4886312 - 10/29/11 12:31 AM

Quote:


OK, my 30-min AP105 unbinned is about 2.0 arc-sec/pixel, and my 5-min AP155 barlowed is about 0.55 arc-sec/pixel. I think that in both cases the software corrected tracking has better (rounder) stars when you account for image duration and image scale. Would you not agree?

-Ray




Hi Ray,
I have not measured the FWHM and elongation of the stars in my 600 sec image, or in your images, so it is hard for me to agree that your stars would be rounder without looking closer at the images, although you may be correct. Also, based on your description of the software tracking correction you are either modeling the correction for flexure and refraction based on previously acquired imaging, or it is being done in real time. Either way it sounds an awfully lot alike to using a guide camera to adjust the pointing of the scope to stay locked on the stars. The point of the TDM (for me) is to not have to rely on feedback from the stars in order to minimize the amount of extra equipment used to make excellent images. What is your software tracking correction using as a reference to correct to?

Thanks

Jerry Hubbell
Lake of the Woods Observatory (MPC I24)
Locust Grove, VA


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

Reged: 08/10/09

Loc: Singapore
Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886325 - 10/29/11 12:41 AM

as an aside.. the notion of using a high-precision encoder on the RA axis for corrections is not unique to the TDM (it is a $5000 option on the El Capitan).

so.. i tried looking for such encoders, the idea being "hmm.. maybe i can roll my own." only Gurley and Renishaw seem to make encoders of the required precision (>1 million pulses per revolution). maybe Johannes Heidenhain does as well but all their docs are in German.

the usual encoder makers (BEI, US Robotics, HP, Allen-Bradley..) don't make million-count encoders, they top out at around 200K PPR. and, the million-count encoders cost at least $500 (used).

so I don't really see how the TDM price can go much lower. Maybe (just maybe) it can hit $1000, but ultra-high count metrology encoders aren't exactly a mass-market item, they've been around for decades, and the astronomy market isn't really a mass market.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Ray Gralak
Vendor (PEMPro)


Reged: 04/19/08

Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886330 - 10/29/11 12:50 AM

Hi Jerry,

Quote:

Quote:


OK, my 30-min AP105 unbinned is about 2.0 arc-sec/pixel, and my 5-min AP155 barlowed is about 0.55 arc-sec/pixel. I think that in both cases the software corrected tracking has better (rounder) stars when you account for image duration and image scale. Would you not agree?

-Ray




Hi Ray,
I have not measured the FWHM and elongation of the stars in my 600 sec image, or in your images, so it is hard for me to agree that your stars would be rounder without looking closer at the images, although you may be correct. Also, based on your description of the software tracking correction you are either modeling the correction for flexure and refraction based on previously acquired imaging, or it is being done in real time. Either way it sounds an awfully lot alike to using a guide camera to adjust the pointing of the scope to stay locked on the stars. The point of the TDM (for me) is to not have to rely on feedback from the stars in order to minimize the amount of extra equipment used to make excellent images. What is your software tracking correction using as a reference to correct to?

Thanks

Jerry Hubbell
Lake of the Woods Observatory (MPC I24)
Locust Grove, VA



The reference is plate-solved images precessed to JNow coordinates. The rate corrections are made by offseting the tracking rate in RA and Dec slightly from sidereal... enough to compensate for refraction, polar alignment errors, flexure, etc.

Although you could try FWHM it is probably not a good comparison because our sites are completely different and you said your image was a little out of focus.

Instead, please take a look at the stars in the AP155's 5-minute image at 0.55 arc-sec/pixel, which would be equivalent to a 10-minute image at 1.10 arc-sec/pixel. Then compare them to the stars in your 10-minute image (at 1.38 arc-sec/pixel). As I mentioned before I think I can see a slight bit of trailing in stars in your image (from upper left to lower right). It's slight, but it's there. I don't think there is any trace of trailing of stars in my AP155 image.

Now I realize that is just one image so if you have better images then my conclusion might be different.

-Ray


Post Extras: Print Post   Remind Me!   Notify Moderator  
gdd
Pooh-Bah


Reged: 11/23/05

Loc: N Seattle suburb, WA
Re: Telescope Drive Master - The Truth of the Matter new [Re: orlyandico]
      #4886341 - 10/29/11 01:03 AM

Quote:

so I don't really see how the TDM price can go much lower. Maybe (just maybe) it can hit $1000, but ultra-high count metrology encoders aren't exactly a mass-market item, they've been around for decades, and the astronomy market isn't really a mass market.





My thought was true the astronomy market is not a mass market, but there exist other markets which may be much larger. Also the encoders do not appear to require much modification other than a shaft adapter to re-purpose them from one industry to another. They must have come down in price over the decades or they would have been applied to telescope mounts years ago.

Gale


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

Reged: 08/10/09

Loc: Singapore
Re: Telescope Drive Master - The Truth of the Matter new [Re: gdd]
      #4886381 - 10/29/11 02:12 AM

they have been applied to telescope mounts for years.. just not at the $1500 level.

it's the same case with harmonic drives, they're used everywhere in industry (probably to a much greater extent than million-count encoders). but there aren't any mass-market mounts with harmonic drives.

the bottom-line is, the 200K PPR encoders are probably good enough for the vast majority of applications out there. Note that the 10K PPR encoders cost $20 - and are used everywhere.

somehow i don't see million-count encoders for $20... or even $200.


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

Reged: 08/10/09

Loc: Singapore
Re: Telescope Drive Master - The Truth of the Matter new [Re: orlyandico]
      #4886398 - 10/29/11 02:50 AM

Heres another point: most mass market scopes lose sync if you move the mount manually. The cgem and atlasas well as most of the vixens definitely fall in this group.

But what would it take to ensure that the mount never loses sync? A couple of lousy $20 encoders, one on each axis. I believe these encoders are options on the G11, and standard on the Taks. $20! And yet the mass market cant be bothered to add these.


Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: Telescope Drive Master - The Truth of the Matter new [Re: orlyandico]
      #4886483 - 10/29/11 06:26 AM

Quote:

But what would it take to ensure that the mount never loses sync? A couple of lousy $20 encoders, one on each axis. I believe these encoders are options on the G11, and standard on the Taks. $20! And yet the mass market cant be bothered to add these.




The inexpensive encoders on mounts are able to be high precision because they are in front of the gearbox and aren't directly reading the axis angle as the TDM encoders do. But this also means they can't "see" periodic error and gearbox errors. In order to have high precision to move the 'scope by hand, you would need much more expensive encoders on the axes - unless all you want is an approximate sky location.

I'm not sure how inexpensively these encoders can be made, but I do expect them to appear more - but maybe not for the low to mid-range mounts - at least for a while.

Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886494 - 10/29/11 06:41 AM

Quote:

I have read through these comments with a lot of interest, and indeed there is a lot of misunderstanding about how the TDM works, and how it is useful to the average amateur with the good but inexpensive mount i.e., EQ6 Pro, Orion Atlas, etc.




Hi Jerry-

I think if you look at the comments I made in the other thread, you'll see that I agree with parts of your assessment, but not all. I think you have demonstrated unique performance at a low cost for the specific goal of short, unguided images for photometric and astrometric work - but I don't think you have demonstrated a cost-effective solution for long exposure deep sky imaging - either with or without autoguiding. As I said, I have yet to see an FWHM value for a long exposure image with this system - and that's the main thing I go by.

You allude to having tried to autoguide your mount, but I don't think you explored the full space of software and techniques for optimal results. In my view, if you use OAG and low latency guiding with video, and rapid corrections (a la MetaGuide, which I wrote, and which is free), you can achieve much better results than by more conventional methods. This involves no costly hardware add-ons. I work with cge and cge-pro mounts and achieve 2" fwhm and below from unexceptional skies in the northeast U.S., and I believe others have done well with more inexpensive mounts via EQMod and similar methodology.

You also allude to the lack of need for 10-20m exposures given typical light pollution - but that ignores narrow band work at f/5 to f/10 - which is growing in popularity. You need long exposures for that.

I think high-res encoders have the potential to smooth out RA errors due to gearbox and bearings - and that is a key limitation to obtaining good results with a non high-end mount. But you will still need to autoguide to make corrections in RA and Dec to achieve optimal fwhm in long exposures - and the end result may not be much different from tight autoguiding at high frequency and low latency with no special encoders involved.

So - I expect high-res encoders to make guiding easier and more casual with cheaper mounts, and they will smooth out errors at arbitrary frequencies that are unreachable by PEC at any update rate, but they will still require tight guiding with a good centroid to obtain optimal fwhm.

Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jerry Hubbell
member
*****

Reged: 02/16/09

Loc: Locust Grove, VA, USA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Ray Gralak]
      #4886573 - 10/29/11 08:18 AM

Quote:

Hi Jerry,

The reference is plate-solved images precessed to JNow coordinates. The rate corrections are made by offseting the tracking rate in RA and Dec slightly from sidereal... enough to compensate for refraction, polar alignment errors, flexure, etc.


Now I realize that is just one image so if you have better images then my conclusion might be different.

-Ray




Hi Ray,

So your system DOES use a guide camera to acquire images for correcting PE. The only difference between using PHD and your software guide correction is that yours uses multiple stars to calculate a plate center as opposed to using an individual star to adjust the drive rate.

Well, isn't this cheating a little on our discussion here? The premise I presented was that the EQ6 coupled with the TDM performed as well as an expensive mount WITHOUT GUIDING on STARS. That was the basis for comparison. Further my contention is that with an excellent polar alignment and the TDM the results are comparable to mounts in the $5000-$10000 range for the type of imaging I am doing, UNGUIDED. I don't believe we can have an honest conversation until we see the results under the same requirements.

Let me ask an interesting question, Don't take this wrong, I am honestly curious about why this is:

If Guiding is absolutely necessary, why don't the manufacturers make large mounts (load capacities in the 20-50Kg range) that have mediocre (30-40 arcsec p-p) PE. It seems they could make mounts cheaper if they didn't have to worry about trying to achieve unguided PE <5 arcsec p-p. I understand that everything needs to be beefier but I am sure a lot of time and effort is put in to make sure the PE is low on these mounts resulting in the higher cost and price for the buyer. If external guiding with a guidescope is necessary anyway, why not create a mount that has an equipment capacity of 50Kg and offer it for say $2500?

I think I know the answer, When customers pay >$5000 for a mount the expectation that not only the capacity increases, but that all the performance parameters improve including PE. That's reasonable.

But to expect the performance improvements when paying a large sum of money, and then come back and then say it doesn't matter that PE has improved on the expensive mounts, we have to use guide cameras anyway, is a bit disengenuous.

Jerry

Edited by Jerry Hubbell (10/29/11 08:51 AM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jerry Hubbell
member
*****

Reged: 02/16/09

Loc: Locust Grove, VA, USA
Re: Telescope Drive Master - The Truth of the Matter new [Re: freestar8n]
      #4886594 - 10/29/11 08:38 AM

Quote:

Quote:

I have read through these comments with a lot of interest, and indeed there is a lot of misunderstanding about how the TDM works, and how it is useful to the average amateur with the good but inexpensive mount i.e., EQ6 Pro, Orion Atlas, etc.




Hi Jerry-

I think if you look at the comments I made in the other thread, you'll see that I agree with parts of your assessment, but not all. I think you have demonstrated unique performance at a low cost for the specific goal of short, unguided images for photometric and astrometric work - but I don't think you have demonstrated a cost-effective solution for long exposure deep sky imaging - either with or without autoguiding. As I said, I have yet to see an FWHM value for a long exposure image with this system - and that's the main thing I go by.

You allude to having tried to autoguide your mount, but I don't think you explored the full space of software and techniques for optimal results. In my view, if you use OAG and low latency guiding with video, and rapid corrections (a la MetaGuide, which I wrote, and which is free), you can achieve much better results than by more conventional methods. This involves no costly hardware add-ons. I work with cge and cge-pro mounts and achieve 2" fwhm and below from unexceptional skies in the northeast U.S., and I believe others have done well with more inexpensive mounts via EQMod and similar methodology.

You also allude to the lack of need for 10-20m exposures given typical light pollution - but that ignores narrow band work at f/5 to f/10 - which is growing in popularity. You need long exposures for that.

I think high-res encoders have the potential to smooth out RA errors due to gearbox and bearings - and that is a key limitation to obtaining good results with a non high-end mount. But you will still need to autoguide to make corrections in RA and Dec to achieve optimal fwhm in long exposures - and the end result may not be much different from tight autoguiding at high frequency and low latency with no special encoders involved.

So - I expect high-res encoders to make guiding easier and more casual with cheaper mounts, and they will smooth out errors at arbitrary frequencies that are unreachable by PEC at any update rate, but they will still require tight guiding with a good centroid to obtain optimal fwhm.

Frank




Thanks Frank for your thoughtful discussion. I agree, for the absolute highest quality deep sky (long exposure)work with the lower cost mounts, there is no substitute for using a guide camera. And the system that Ray has talked about in this thread (using multiple stars to do plate solves) is probably the best way to do it. But trying to use a guide camera for high frequency (<0.5 hz) gets a little dicey depending on the field you are imaging.

Finding a bright enough guide star is always problematic. Plus the trend is towards lower focal length, smaller guidescopes. This creates a problem in that the plate scale for the guiding image does not have enough resolution to correct to sub arcsecond movements, especially when the seeing is not that poor. So the answer of course is to use guidescope similar to the old days with long focal lengths and smaller fields of view which only aggravates the issue of finding a bright enough guide star.

There is absolutely no doubt that for narrow band imaging, long exposures (20-60 minutes, if anyone goes that high) are necessary to beat down the noise and not be overwhelmed with readout noise because of the number of frames necessary with shorter exposures, that external guiding systems are necessary. I agree with you 100%.

I have used PHD with an OAG and with external guide scopes, and for me it's not a matter of tweaking my equipment and refining my skills, and finding the best software available... It's about how much time do I want to spend messing with all that. That's all, that's the bottom line for me.

I think the recent development of Adaptive Optics for amateur level systems is a response to the need for very fast corrections due to bad seeing and scintillation, and may take the place of guide scopes also. These system's cost are the same, or more than the TDM, and I do not see the same level of discussion, but I have not been looking for it either.

I understand fully that the TDM is not for every amateur, just like refractors are not for everyone, or DOBS or RC's, Every tool has it's strengths, and for me and my observing program, the TDM saves me time and effort at a reasonable cost.

I appreciate the discussion, I think it is valuable.

Jerry Hubbell
Lake of the Woods Observatory
Locust Grove, VA

Edited by Jerry Hubbell (10/29/11 09:21 AM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
gdd
Pooh-Bah


Reged: 11/23/05

Loc: N Seattle suburb, WA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886618 - 10/29/11 09:13 AM

I have wondered why someone does not put a mid-precision encoder on the output of the gearbox so PEC will have a cleaner worm error to deal with. That should be a lot cheaper than a high precision encoder on the RA. There would still be uncorrected noise from the worm and RA bearings, but there should still be an overall improvement.

Gale


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

Reged: 08/10/09

Loc: Singapore
Re: Telescope Drive Master - The Truth of the Matter new [Re: freestar8n]
      #4886628 - 10/29/11 09:17 AM

Frank,

The encoders on say the CGEM or LXD75 are on the motor shaft. These are used solely for counting pulses (to determine position) and for determining motor speed. These usually are only hundreds of PPR (pulses per revolution), perhaps even dozens - the Meade motor boards have these 24-odd tooth encoders on the motor shaft before the gearbox.

The encoders on the Tak mounts (the later ones, e.g. EM200) are indeed on the RA and DEC axis. They don't have to be million-tick like the TDM, since they don't need to measure periodic error, they are solely for pointing. A 10000-tick encoder is probably plenty for this application (and costs $20 in bulk) -- all DSC's out there use either 2000-, 4000- or 10000-tick encoders on the RA and DEC axes. Even 50000- to 100000-tick encoders aren't crazy expensive like the Gurleys and Renishaws. Of course you won't get sub-arcsecond pointing but for GoTo that is probably not needed.

The FS2 Goto system can also take encoder inputs from RA and DEC to lock the controller's pointing to the axes.

My point is - putting 10000-tick encoders on the RA and DEC of a mount is an "obvious" feature, but none of the entry- to midrange mounts have it (you have to go to a G11 or Tak EM-200 to get it).

Hence - I really don't see TDM-style million-tick encoders migrating to the mass market anytime soon, if 10K-tick encoders aren't even there. I assume TDM-style technology would appear on the $5K class mounts first (CGE-Pro, etc.) in fact doesn't the iEQ-75 come with such an encoder?


Post Extras: Print Post   Remind Me!   Notify Moderator  
Gord
Carpal Tunnel
*****

Reged: 01/06/04

Loc: Toronto, ON, Canada
Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886631 - 10/29/11 09:22 AM

Hey Jerry,

One of the things that you pointed out that I think is important to re-iterate is the simplicity involved with the TDM solution. This was described in the S&T article, and to me at least, it's a big factor.

You install it (bolt it on), plug it in, and turn it on. Instant result. No software, no configuration, no playing around, no extra steps. It just works. Yes there is a cost, but in the grand scheme of things, there is a cost to everything in one way or another ($$, time, effort, frustration, etc.).

Clear skies,


Post Extras: Print Post   Remind Me!   Notify Moderator  
Ray Gralak
Vendor (PEMPro)


Reged: 04/19/08

Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886663 - 10/29/11 09:52 AM

Quote:

Quote:

Hi Jerry,

The reference is plate-solved images precessed to JNow coordinates. The rate corrections are made by offseting the tracking rate in RA and Dec slightly from sidereal... enough to compensate for refraction, polar alignment errors, flexure, etc.


Now I realize that is just one image so if you have better images then my conclusion might be different.

-Ray




Hi Ray,

So your system DOES use a guide camera to acquire images for correcting PE. The only difference between using PHD and your software guide correction is that yours uses multiple stars to calculate a plate center as opposed to using an individual star to adjust the drive rate.

Well, isn't this cheating a little on our discussion here? The premise I presented was that the EQ6 coupled with the TDM performed as well as an expensive mount WITHOUT GUIDING on STARS. That was the basis for comparison. Further my contention is that with an excellent polar alignment and the TDM the results are comparable to mounts in the $5000-$10000 range for the type of imaging I am doing, UNGUIDED. I don't believe we can have an honest conversation until we see the results under the same requirements.

Let me ask an interesting question, Don't take this wrong, I am honestly curious about why this is:

If Guiding is absolutely necessary, why don't the manufacturers make large mounts (load capacities in the 20-50Kg range) that have mediocre (30-40 arcsec p-p) PE. It seems they could make mounts cheaper if they didn't have to worry about trying to achieve unguided PE <5 arcsec p-p. I understand that everything needs to be beefier but I am sure a lot of time and effort is put in to make sure the PE is low on these mounts resulting in the higher cost and price for the buyer. If external guiding with a guidescope is necessary anyway, why not create a mount that has an equipment capacity of 50Kg and offer it for say $2500?

I think I know the answer, When customers pay >$5000 for a mount the expectation that not only the capacity increases, but that all the performance parameters improve including PE. That's reasonable.

But to expect the performance improvements when paying a large sum of money, and then come back and then say it doesn't matter that PE has improved on the expensive mounts, we have to use guide cameras anyway, is a bit disengenuous.

Jerry




Um... I never said there was a guide camera. In fact, let me state this now... there was NO guide camera. NO autoguiding was done!!!

What was done was pre-measurement of pointing errors across the sky. Have you heard of TPoint or MaxPoint? They collect the positional errors of the mount by doing plate solves. The process can take anywhere from 10 minutes for a small model to several hours for a 500+ point model. Once the modeling is done the positional and tracking rate errors can be calculated from the data. The key to making this work is the quality of the data model and the precision of the mount. The mount has to be able to repeatably slew accurately and have the ability to precisely change the rate slightly from sidereal to match the required (modeled) RA and Dec tracking rates. The technique is not new. It's been used in production Paramount ME's for perhaps 10 years (it's called ProTrack if you want to Google it). Also, major observatories have used this method for even longer.

So, as I clearly said before, using an encoder is not good enough to do long exposure unguided images at longer focal lengths. You would also need software tracking correction and a precision mount.

-Ray


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jerry Hubbell
member
*****

Reged: 02/16/09

Loc: Locust Grove, VA, USA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Gord]
      #4886672 - 10/29/11 10:01 AM Attachment (26 downloads)

Quote:

Hey Jerry,

You install it (bolt it on), plug it in, and turn it on. Instant result. No software, no configuration, no playing around, no extra steps. It just works...




Yes Gord, I have stated that more than once in this discussion. That is a big part of it, again for me.

For those who have not read the paper I posted, here is my initial chart of my measured Tracking Error (TE) as measured by the TDM encoder, and then the results when I turn on the TDM, first at 1 Hz/ 1 arcsec control band, and then at 5 Hz / 0.5 arcsec control band. You be the judge. What I have measured and reported in my paper is that with my TDM mounted on my EQ6 Pro, I get a TE value < +/-0.7 arcseconds 95% of the time. That is the starting point, or my foundation for my imaging system performance.

You be the judge, is it worth $2200 to get this type of improvement for your imaging program with practically zero setup time or ongoing maintenance during your imaging session? That's up to you.

Jerry Hubbell
Lake of the Woods Observatory
Locust Grove, VA

Edited by Jerry Hubbell (10/29/11 01:57 PM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jerry Hubbell
member
*****

Reged: 02/16/09

Loc: Locust Grove, VA, USA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Ray Gralak]
      #4886694 - 10/29/11 10:16 AM

Quote:


Um... I never said there was a guide camera. In fact, let me state this now... there was NO guide camera. NO autoguiding was done!!!

What was done was pre-measurement of pointing errors across the sky. Have you heard of TPoint or MaxPoint? They collect the positional errors of the mount by doing plate solves. The process can take anywhere from 10 minutes for a small model to several hours for a 500+ point model. Once the modeling is done the positional and tracking rate errors can be calculated from the data. The key to making this work is the quality of the data model and the precision of the mount. The mount has to be able to repeatably slew accurately and have the ability to precisely change the rate slightly from sidereal to match the required (modeled) RA and Dec tracking rates. The technique is not new. It's been used in production Paramount ME's for perhaps 10 years (it's called ProTrack if you want to Google it). Also, major observatories have used this method for even longer.

So, as I clearly said before, using an encoder is not good enough to do long exposure unguided images at longer focal lengths. You would also need software tracking correction and a precision mount.

-Ray




Okay, I give you that, but if you recall, I did originally state that either you were doing some kind of static modeling using a camera OR using a guide camera realtime. I am familiar with T-point, but did not realize it had the resolution and capacity to correct for small errors or non-periodic errors in the mount, only for correcting pointing errors. I understood it was only for the correction of possibly refraction, and flexure pointing errors. I knew you were having to put some work (a lot of work) into it at somepoint, LOL

This method is even more onerous for the less expensive mounts because of the inconsistent performance of the mount. This is very similar to creating a PEC curve for your mount. I have found that for my EQ6 mount. The PEC curve is bounded by the range of positions you were pointing at when acquiring the data, and only corrected 60-70%, at the best, of the PE. So for a typical EQ6 class mount you may be able to bring the PE down from 40 arcsec p-p to 10-15 arcsec p-p. I suspect if I used T-Point on my EQ6 I would manage >5 arcsec p-p but not much better.

I guess the only other question is how much is that software, and what is the cost of upkeep?

Thanks for the insights into what it takes to get the performance you are getting.

Jerry Hubbell
Lake of the Woods Observatory (MPC I24)
Locust Grove, VA

Edited by Jerry Hubbell (10/29/11 10:41 AM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jerry Hubbell
member
*****

Reged: 02/16/09

Loc: Locust Grove, VA, USA
Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886739 - 10/29/11 10:51 AM Attachment (15 downloads)

Just as another piece of data to examine, here is an image from the TDM developer site:

http://www.mda-telescoop.com/index.php?option=com_content&task=view&id=55&Itemid=85

This image shows the tracking performance using a star image not the output from the encoder when the TDM is on and controlling the mount (an EQ6). It looks to me that over the 300 seconds of data shown, the star image only exhibits scintillation on the order of +/- 2-3 arcseconds; there are no other defects in this image using a correctly tuned, polar aligned mount. I am sure over a 30 minute period you would not see much difference than what you are seeing here. There are other images on the linked page above that show the various defects that may be present in your mount that would need to be corrected to expect the nominal kind of performance that Ray and I are talking about.

Jerry Hubbell
Lake of the Woods Observatory (MPC I24)
Locust Grove, VA

Edited by Jerry Hubbell (10/29/11 10:59 AM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
Ray Gralak
Vendor (PEMPro)


Reged: 04/19/08

Re: Telescope Drive Master - The Truth of the Matter new [Re: Jerry Hubbell]
      #4886779 - 10/29/11 11:20 AM

Hi Jerry,

Quote:

Quote:


Um... I never said there was a guide camera. In fact, let me state this now... there was NO guide camera. NO autoguiding was done!!!

What was done was pre-measurement of pointing errors across the sky. Have you heard of TPoint or MaxPoint? They collect the positional errors of the mount by doing plate solves. The process can take anywhere from 10 minutes for a small model to several hours for a 500+ point model. Once the modeling is done the positional and tracking rate errors can be calculated from the data. The key to making this work is the quality of the data model and the precision of the mount. The mount has to be able to repeatably slew accurately and have the ability to precisely change the rate slightly from sidereal to match the required (modeled) RA and Dec tracking rates. The technique is not new. It's been used in production Paramount ME's for perhaps 10 years (it's called ProTrack if you want to Google it). Also, major observatories have used this method for even longer.

So, as I clearly said before, using an encoder is not good enough to do long exposure unguided images at longer focal lengths. You would also need software tracking correction and a precision mount.

-Ray




Okay, I give you that, but if you recall, I did state that either you were doing some kind of static modeling using a camera OR using a guide camera realtime. I am familiar with T-point, but did not realize it had the resolution and capacity to correct for small errors or non-periodic errors in the mount. I understood it was only for the correction of possibly refraction, and flexure. I knew you were having to put some work (a lot of work) into it at somepoint, LOL





Actually, TPoint cannot correct for periodic error. That is still done with PEC. But some higher end mounts, like Astro-Physics mounts, ship with PEC already recorded so, without installing anything extra you get a mount with typically less than 1 arc-sec PE. In fact the other night I programmed my Ap1200 and the resulting PE was +/-0.2 arc-secs. So right there this mount probably tracks better than any cheaper mount plus encoder.

BTW, there are other advantages to having low inherent periodic error. As I alluded to (and I think you acknowledged) pointing errors will be lower with lower PE mounts. That's because when you slew to the same RA/Dec at different times it will be at different positions in the worm so positioning accuracy is affected by the amount of periodic error. When doing plate solves have a precision mount (low PE, great pointing accuracy) can make a big difference in calculating an accurate model from which precise tracking rates can be calculated. That's not saying it can't be done with a less accurate mount. It just won't be as precise.

And, once you go non-sidereal, for instance if you are tracking a fast-moving earth satellite, high PE might clearly show as a wobbling motion because I don't think the encoder solutions available work at rates significantly faster than sidereal, correct?

Quote:


This method is even more onerous for the less expensive mounts because of the inconsistent performance of the mount. This is very similar to creating a PEC curve for your mount. I have found that for my EQ6 mount. The PEC curve is bounded by the range of positions you were pointing at when acquiring the data, and only corrected 60-70%, at the best, of the PE. So for a typical EQ6 class mount you may be able to bring the PE down from 40 arcsec p-p to 10-15 arcsec p-p. I suspect if I used T-Point on my EQ6 I would manage >5 arcsec p-p but not much better.





Oh, I agree that software tracking rate correction probably will not work as well for less accurate mounts. But that was the big debate this whole discussion, wasn't it? To show if there is a difference in potential performance between a less precise mount and encoder versus what you can accomplish with a higher precision mount. I do acknowledge the value of the setup you are using but I think that I have shown that there is improvement that could be had in unguided tracking with a precision mount and tracking rate correction software.

Quote:


I guess the only other question is how much is that software, and what is the cost of upkeep?




Interesting that you should mention cost! In general I think higher end gear holds its value more and generally looks and works better and longer than cheaper gear. For instance my AP1200 still looks great after 5-6 years of use while a couple cheap mounts I own have corroding bolts already. The value of the cheap mounts probably has gone down more than my expensive mounts. In general I think most higher-end mounts are designed to last longer while cheaper mounts have had "corners cut" to make them as cheap as possible and probably not last as long. I think you know what I'm talking about, I hope?

If one is really interested in this hobby they will likely progress from less expensive mounts up to more expensive mounts. If they just save and purchase a good expensive mount first then in the long run I think they may save money. It might cost a lot more initially but over the long term they won't lose money every time they upgrade. If they started with a higher-end mount it would retain value better, last longer, and they will probably be a lot happier with the imaging results they get.

-Ray


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
23 registered and 30 anonymous users are browsing this forum.

Moderator:  Dave M, richard7, bilgebay 

Print Thread

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


Thread views: 18203

Jump to

CN Forums Home


Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics