16 replies to this topic

[Nicole Sharp]

After an altazimuth SynScan Daytime Alignment on the Sun, I get a drift of about 0.5 degrees per hour in the Solar tracking rate on the Orion StarSeeeker 4 (Sky-Watcher Star Discovery 2).

 

Is this typical for this class of mount? How much drift do you have when attempting Solar tracking with a SynScan Daytime Alignment or a NexStar Solar System Alignment on other Synta mounts such as the SLT, AZ-GTi, EQM-35, Evolution, Advanced VX, or AZ-EQ5?

 

Drift is measured by centering the Sun (with Solar tracking enabled) and then waiting to see how long until the Sun drifts to where the edge of the Solar disc touches the edge of the field of view.  If you know the field of view, the rate of drift is half the field of view minus the Solar angular radius divided by the drift time.  For me, the center of the Sun drifted about 5 arcminutes over a period of about 9 minutes.

 

0.5 degrees per hour means that I have to recenter the Sun about once every 9 minutes at 1000 mm for Canon APS-C. If tracking a Sunspot at 3000 mm, it would have to be recentered about once every 3 minutes.  Recentering the Sun every 9 minutes might become problematic for a partial or annular Solar eclipse.

Edited by Nicole Sharp, 04 August 2022 - 05:38 AM.

[Phil Sherman]

The sun is a large extended target. I'd expect that a solar system align on it would not be accurate enough to keep a long focus lens centered on it. You would have better luck using Venus or Jupiter for your daytime alignment target. Both, especially Venus, are readily visible in a visual light scope in the daytime. Venus isn't usable when it's too close to the sun. If you do your alignment using an eyepiece, it should be a reticle eyepiece which will make it easier to center the target.

 

If you're imaging a solar eclipse, my recommendation is to set up your gear the night before and do your alignments and verification that they're good using visible stars. If you're using a GEM mount, there's a "daytime mount alignment" procedure that, with a single azimuth adjustment to an astronomical target (except the sun) can yield an excellent polar alignment.

[AstroBruce]

If you are only syncing on one object, you must make sure the mount is perfectly level.

 

Bruce

[Nicole Sharp]

The sun is a large extended target. I'd expect that a solar system align on it would not be accurate enough to keep a long focus lens centered on it. You would have better luck using Venus or Jupiter for your daytime alignment target. Both, especially Venus, are readily visible in a visual light scope in the daytime. Venus isn't usable when it's too close to the sun. If you do your alignment using an eyepiece, it should be a reticle eyepiece which will make it easier to center the target.

If you're imaging a solar eclipse, my recommendation is to set up your gear the night before and do your alignments and verification that they're good using visible stars. If you're using a GEM mount, there's a "daytime mount alignment" procedure that, with a single azimuth adjustment to an astronomical target (except the sun) can yield an excellent polar alignment.

Setting up the telescope at night is not an option, especially for a Solar eclipse, unless perhaps I could find a campground without any obstructions.

A multi-object StarSense Solar System Alignment (for example, a two-object alignment using both Sol and Venus) should increase the daytime tracking accuracy but the StarSense for Sky-Watcher is sold out until October and I would like the mount to work without needing StarSense if possible.

According to Orion, it is not possible to realign a SynScan mount on Venus (which would be much more accurate since Venus is much smaller than Sol in apparent size). A realignment will require returning the mount to the home position first, which means that the Sun is the only object that can be aligned on in the daytime (unless the Moon is up). However, the StarSense for Sky-Watcher allows a NexStar Solar System Alignment which does not require returning to a home position before realignment, so you can realign on Venus with StarSense or NexStar. But again I would like the mount to work without StarSense if possible.

I remeasured the Solar drift this morning. Yesterday afternoon it took 9 minutes for the Sun to drift 5 arcminutes (a Solar drift rate of 0.5 degrees per hour) and this morning it took 8 minutes for the Sun to drift 5 arcminutes (a Solar drift rate of 0.6 degrees per hour). So this seems to be a consistent value and I think might be normal for the StarSeeker. I was able to find Venus at 1000/27 (with Venus in the telescope field of view) and then track Venus at 2000/27 with a 2X Barlow. So the alignment appears to be good but there is a similar drift on Venus as well with the Daytime Alignment on the Sun.

My main question is what is a typical Solar drift rate on other mounts? If about 0.6 degrees per hour is typical for a StarSeeker, how does that compare to other GOTO mounts with daytime alignments on the Sun?

The tripod base was leveled to within 1 degree. The optical tube was leveled to within 1 degree. The optical tube was oriented to True North within about 4 degrees. The OTA was balanced separately on a dovetail bar to find the center of mass to within about one quarter of an inch. Latitude, longitude, elevation, and time were set manually from a smartphone GPS. The tripod was also lowered to half-height without the legs extended. That is about as good as you can get.

Edited by Nicole Sharp, 04 August 2022 - 04:12 PM.

[Nicole Sharp]

Perhaps this thread should be moved to the Solar Observing Forum?

[Nicole Sharp]

Orion tech support says that 0.6 degrees per hour of drift in Solar tracking might be normal for the StarSeeker with a SynScan Daytime Alignment on the Sun.  I would concur since I experienced a consistent Solar tracking error (to within plus or minus 0.05 degrees per hour) over two separate days.

 

Has anyone ever bothered to measure the Solar tracking error with any other GOTO mounts?  How often do you have to recenter the Sun when Solar tracking?  Do any GOTO mounts exist that can track the Sun all day without needing to be aligned at night or use a Solar autoguider?

 

Unless I am okay with manually recentering the Sun every 8 minutes during a Solar eclipse, it looks like my only other alternative would be a Solar-guided mount like an Orion Solar StarSeeker or Sky-Watcher SolarQuest.  Or using a focal length shorter than 1000 mm for APS-C.

 

Not sure if 0.6 arcseconds per second of tracking error will be acceptable for altazimuth APS-C exposures of the Solar corona during totality at 250 mm of focal length.  Otherwise having to babysit the mount and recenter it every 8 minutes during the partial or annular phases of an eclipse is primarily just an inconvenience.  The main difficulty would be realigning the images if I wanted to make a timelapse video or attempt frame stacking.

Edited by Nicole Sharp, 08 August 2022 - 02:44 PM.

[Nicole Sharp]

For 250 mm of focal length using 6000*4000 pixels over 22.3*14.9 mm^2 of sensor, a Solar tracking error of 0.6 arcseconds per second will limit exposure time to less than 2.1 seconds, down from the maximum of 3.7 seconds with perfect altazimuth tracking.  NASA's "Mr. Eclipse" Fred Espenak recommends only about 1/2 of a second on the corona for 250 mm f/5.6 (about 4 Solar radii) so that seems to be okay.

 

https://eclipse.gsfc...lp/SEphoto.html

 

At 100 mm of focal length (about 8 Solar radii) with f/5.6, Espenak recommends about 2 seconds of exposure time.  With 0.6 arcseconds per second altazimuth Solar tracking error, this limits the exposure time to 2.9 seconds with 100 mm of focal length for Canon APS-C (down from 3.8 seconds with perfect altazimuth tracking).  So this should still be acceptable with the tracking error being sub-pixel for exposures of 2 seconds or less.

 

I should probably retest the Solar tracking error using the 250/45 telephoto lens instead of the 1000/90 telescope.  But the smaller apparent size of the Sun and the smaller moment of inertia of the optical tube assembly might actually reduce the tracking error.

Edited by Nicole Sharp, 08 August 2022 - 03:16 PM.

[rimcrazy]

Get a Hinode Solar Guider and you won't have to worry about it.  This tracks dead nuts on all the time.  Love it.

[sctchun]

Get a Hinode Solar Guider and you won't have to worry about it.  This tracks dead nuts on all the time.  Love it.

I second this, not totally sure if it would work in Alt-Az mode.

[Nicole Sharp]

Get a Hinode Solar Guider and you won't have to worry about it.  This tracks dead nuts on all the time.  Love it.

 

A Sky-Watcher SolarQuest will cost less.  So could using just any off-the-shelf cheap astrocam with a Solar-filtered guidescope.

 

However, the majority of Solar autoguiders require an equatorial alignment.  The only two exceptions I know of are a custom altazimuth autoguiding application written specifically for the Celestron NexStar 4SE/5SE mount and the Sky-Watcher SolarQuest (Orion Solar StarSeeker) mounts which have built-in altazimuth Solar autoguiding.

 

My question though is how does the Solar tracking on the Orion StarSeeker compare to other GOTO mounts without using Solar autoguiding, aligning the mount at night first, or using StarSense?  I speculate that the majority of the error is just from the daytime alignment routine, especially since the SynScan Daytime Alignment can only be aligned on the Sun (and not realigned on Venus).

 

My hypothesis is that the tracking error should be similar on any other GOTO mount with a single-object daytime alignment on the Sun, since a mechanically superior mount (a NexStar Evolution or an AZ-EQ6 for example) would still have the same initial alignment error that could only be fixed with Solar autoguiding or StarSense.

Edited by Nicole Sharp, 08 August 2022 - 04:21 PM.

[Hesiod]

Even cheaper than the solarquest (and in my opinion far more reliable) would be a simple clock-drive eq mount.

With such kind of device you have basically to worry only about the drift due to azimuthal error of your PA because if level the tripod and align the mount by night through its polarscope once do not have to worry anymore about altitude error unless mess with the polar wedge (which is not something you should do anyway).

Then aim the Sun and will know that only have to nudge the azimuth knobs if notice it drifts (which is something you can do without prejudice even while observing once the Sun reaches the S quadrant).

It has been several years since the last partial eclipse here in Italy but remember I used a TAL 100/1000 refractor with my clock-driven mount and it kept the Sun in the fov of the eyepiece (I used a 12mm/60°) for the whole event.

 

Anyway, to your question, which I assume is limited to alt/az gotos (for eq gotos applies the same above), I believe that a big role is played by the model employed by the mount.

6-7 years ago the Synscan was very coarse (almost a beta!) if compared to Celestron's Nexstar (which is not the state of the art either, but probably good enough for the majority of us, especially with smallish telescopes): being not a software engineer I can not tell why, but since one of the most striking issue was how quickly the goto accuracy deteriorated I believe it was most likely a consequence of a very basic model.

I do not know if or how much it has been improved, nor if using a gps unit would do any good (if the model uses very rough approximations I assume there is little need for great accuracy when entering time and place), but in any case I do not think the starsense would help (again, as far as I remember this thing needed a pretty dark sky to work as advertised).

Personally have a roughly ten years old SLT, and in the past have used it for low level solar system imaging (3.5" mct and 4" refractor, with focals around 2200mm) but do not remember much, except that I was amused to see how rough and uneven looked its tracking if observed on the laptop screen; If temperature get a bit cooler (now we have 38-40° C, and neither I nor the SLT are willing to test our mettle) I will gladly take it out for a run

[Nicole Sharp]

Even cheaper than the solarquest (and in my opinion far more reliable) would be a simple clock-drive eq mount.

With such kind of device you have basically to worry only about the drift due to azimuthal error of your PA because if level the tripod and align the mount by night through its polarscope once do not have to worry anymore about altitude error unless mess with the polar wedge (which is not something you should do anyway).

Then aim the Sun and will know that only have to nudge the azimuth knobs if notice it drifts (which is something you can do without prejudice even while observing once the Sun reaches the S quadrant).

It has been several years since the last partial eclipse here in Italy but remember I used a TAL 100/1000 refractor with my clock-driven mount and it kept the Sun in the fov of the eyepiece (I used a 12mm/60°) for the whole event.

 

Anyway, to your question, which I assume is limited to alt/az gotos (for eq gotos applies the same above), I believe that a big role is played by the model employed by the mount.

6-7 years ago the Synscan was very coarse (almost a beta!) if compared to Celestron's Nexstar (which is not the state of the art either, but probably good enough for the majority of us, especially with smallish telescopes): being not a software engineer I can not tell why, but since one of the most striking issue was how quickly the goto accuracy deteriorated I believe it was most likely a consequence of a very basic model.

I do not know if or how much it has been improved, nor if using a gps unit would do any good (if the model uses very rough approximations I assume there is little need for great accuracy when entering time and place), but in any case I do not think the starsense would help (again, as far as I remember this thing needed a pretty dark sky to work as advertised).

Personally have a roughly ten years old SLT, and in the past have used it for low level solar system imaging (3.5" mct and 4" refractor, with focals around 2200mm) but do not remember much, except that I was amused to see how rough and uneven looked its tracking if observed on the laptop screen; If temperature get a bit cooler (now we have 38-40° C, and neither I nor the SLT are willing to test our mettle) I will gladly take it out for a run

I have noticed some deterioration in GOTO accuracy with extended use.  I assumed that this was due to the mechanical limitations of the entry-level GOTO mount, and not due to any software problems.  Even with StarSense though, it is recommended to periodically realign the mount to maintain accuracy but you probably do not want to do this in the middle of an eclipse.

 

I don't like SynScan, but the Celestron StarSense for Sky-Watcher is sold out until at least October 2022.  Adding StarSense should improve the alignment accuracy for both daytime and nighttime use.

 

Regarding SynScan Daytime Alignment, I think this was a new feature that was added in the excitement leading up to the 2017 Solar eclipse in the USA so that SynScan mounts could be aligned in the daytime for Solar eclipses.  However as far as I know, SynScan Daytime Alignment is still exclusively available only on SynScan altazimuth mounts, and not available at all on SynScan equatorial mounts.  The fact that Sky-Watcher hasn't been able to figure out how to align a SynScan equatorial mount on the Sun is troubling since NexStar equatorial mounts have been able to do this for a while, and both NexStar and SynScan are built by the same company (Synta).  The only way to align a SynScan equatorial mount in the daytime is using the Celestron StarSense Solar System Alignment for Sky-Watcher, but a SynScan altazimuth mount can use either the SynScan or NexStar alignment routines in the daytime.

 

Since the StarSense is currently unavailable, I have to fully test out the StarSeeker within the 30-day return period without using StarSense.  I thought about trying to hold the order for the StarSeeker until the StarSense was also available, so that I could order them together, but I decided that it might be better to make sure that I have a fully functional mount without needing any additional add-ons such as StarSense.  As far as I can tell, the mount is working correctly.  I can find and track Venus in the daytime, and I can track the Sun for at least a few minutes at a time.  That might be good enough, especially since my calculations indicate the Solar tracking error is small enough to not affect long exposures of the Solar corona during totality.  So the mount seems to be functional with just the SynScan Daytime Alignment, but the accuracy should be improved by adding StarSense at a later date.

 

The point of this thread was to try to figure out if there were any mounts that I could exchange the StarSeeker for to get better tracking of the Sun without needing a Solar autoguider, setting up the mount at night, or using StarSense.  Unless there are software differences in the daytime alignment between SynScan and NexStar like you suggest, without any data on the Solar tracking from any other mounts, my gut says that all GOTO mounts will likely have about the same amount of daytime tracking error since all of the mounts will be aligned on the same object (the Sun) with about the same amount of initial alignment error, regardless of the cost or superior mechanical design of the mount.

 

The only big difference is that NexStar mounts can be realigned without needing to return to a home position first.  This allows for example to align on the Sun, then GOTO Venus, and then realign on Venus for better accuracy.  I can't do this with SynScan unless I add the StarSense to use the NexStar alignment instead.  SynScan can only align on the Sun.

 

A GPS may help slightly.  I am not sure what portion of the 0.6 arcseconds per second tracking error is due to the manual time entry (which should be accurate to within a second or two).

 

Celestron StarSense consists of the StarSense camera and the StarSense handcontroller.  You do not use the StarSense camera in the daytime.  You use only the StarSense handcontroller without the camera attached.  The StarSense Solar System Alignment allows you to align on multiple objects in the daytime.  For example, you can have a three-object alignment using Sol, Luna, and Venus for much better accuracy than you can get with an alignment on just one single daytime object.  You align on the Sun, then GOTO Venus.  On NexStar, you would have to replace the initial alignment with a new alignment on Venus.  On StarSense, you can instead add Venus as a second alignment point within the current alignment.  I think you can add up to 10 different alignment points (objects) in the daytime, which is probably more than you can see.  I was able to find Vega in the daytime last year but I don't think you can add stars (other than Sol) to a Solar System Alignment.

 

Sky-Watcher tests all of their mounts to make sure that they work from 0 degrees Fahrenheit (-17 degrees Celsius) to 120 degrees Fahrenheit (48 degrees Celsius).  Sky-Watcher and Celestron are the same company (Synta) so Celestron mounts should be able to tolerate a similar temperature range.  I was operating the StarSeeker at 100 degrees Fahrenheit (38 degrees Celsius) without any problems but did not test it in any temperatures below freezing (32 deg F or 0 deg C) last year.  On a hot day, you should set up the mount on grass and not on pavement though.

 

I am using a 1000/90 MCT so have a wider field of view than you might have with a 1250/90 MCT.  I had thought about upgrading to a 1300/102 MCT for the upcoming annular Solar eclipse, but this seems like a bad idea unless I have a Solar autoguider to keep the Sun centered in the field of view.  I added the 2X Barlow and was able to track Venus at 2000/90.  The drift is primarily an inconvenience but it means that frames will have to be aligned before they can stacked or time-lapsed.  The other problem is that if you walk away from the mount for too long, you lose the alignment, which is also an annoyance.

 

NexStar would be easier to use out of the box, but I can still use NexStar alignments on the StarSeeker if I add the StarSense adapter, so that might not be a good reason to exchange a SynScan mount for a NexStar mount if the SynScan mount is mechanically superior.  The biggest problem with Celestron is that none of the NexStar altazimuth mounts can go to zenith with a DSLR camera (not even the 90SLT).  So switching from SynScan to NexStar might mean getting an AVX instead, but that would be a lot heavier and less portable than a StarSeeker.

 

I bought the Celestron NexStar 90SLT last year but didn't measure the Solar tracking error before I returned it due to the clearance issue (unable to go to zenith).  Let me know if you get anything similar to the Orion StarSeeker (Sky-Watcher Star Discovery).

Edited by Nicole Sharp, 08 August 2022 - 07:03 PM.

[Nicole Sharp]

Something else important to remember is that neither SynScan nor NexStar can compute a King tracking rate.  So the most accurate alignment should be when the Solar System object is as high in altitude angle as possible.  Attempting a Daytime Alignment or Solar System Alignment on the Sun early in the morning shortly after Sunrise is thus a bad idea and will introduce a lot of error.  If you align on the Sun at 9 AM for example, you might want to realign on the Sun around Noon.  However, if you align on the Sun around Noon, you probably don't want to realign on the Sun again when the Sun is lower in altitude angle.  Even with a perfect alignment, without a King rate, the Solar tracking will deviate when the Sun is close to the horizon.  Applying the tracking from a third-party application like Stellarium can use a King rate though.

Edited by Nicole Sharp, 08 August 2022 - 06:49 PM.

[Nicole Sharp]

Even cheaper than the solarquest (and in my opinion far more reliable) would be a simple clock-drive eq mount.

With such kind of device you have basically to worry only about the drift due to azimuthal error of your PA because if level the tripod and align the mount by night through its polarscope once do not have to worry anymore about altitude error unless mess with the polar wedge (which is not something you should do anyway).

Then aim the Sun and will know that only have to nudge the azimuth knobs if notice it drifts (which is something you can do without prejudice even while observing once the Sun reaches the S quadrant).

It has been several years since the last partial eclipse here in Italy but remember I used a TAL 100/1000 refractor with my clock-driven mount and it kept the Sun in the fov of the eyepiece (I used a 12mm/60°) for the whole event.

 

Anyway, to your question, which I assume is limited to alt/az gotos (for eq gotos applies the same above), I believe that a big role is played by the model employed by the mount.

6-7 years ago the Synscan was very coarse (almost a beta!) if compared to Celestron's Nexstar (which is not the state of the art either, but probably good enough for the majority of us, especially with smallish telescopes): being not a software engineer I can not tell why, but since one of the most striking issue was how quickly the goto accuracy deteriorated I believe it was most likely a consequence of a very basic model.

I do not know if or how much it has been improved, nor if using a gps unit would do any good (if the model uses very rough approximations I assume there is little need for great accuracy when entering time and place), but in any case I do not think the starsense would help (again, as far as I remember this thing needed a pretty dark sky to work as advertised).

Personally have a roughly ten years old SLT, and in the past have used it for low level solar system imaging (3.5" mct and 4" refractor, with focals around 2200mm) but do not remember much, except that I was amused to see how rough and uneven looked its tracking if observed on the laptop screen; If temperature get a bit cooler (now we have 38-40° C, and neither I nor the SLT are willing to test our mettle) I will gladly take it out for a run

 

A simple clockdrive is an interesting idea if it has a Solar tracking rate.  There is no GOTO alignment, so if the Sun is drifting, you can just shut off the clockdrive and nudge your polar alignment.  However, getting the polar alignment accurate in the daytime is challenging and I think the altazimuth GOTO alignment will have less error, especially for long exposures of the Solar corona during totality.  Celestron tech support also told me that an altazimuth GOTO mount will have less alignment error in the daytime than an equatorial GOTO mount, since the equatorial mount will suffer from the combined error of both the Solar alignment and the polar alignment.  This was one of my motivations to get an altazimuth mount instead of an equatorial mount for daytime use.  The best way to reduce the daytime error is to add additional alignment points with StarSense but I think I already have the best possible alignment without using StarSense or GPS.

[Nicole Sharp]

Something else to consider is that SynScan requires a home position whereas NexStar does not.  Technically, adding a home position to the alignment routine should make the alignment more accurate.  For SynScan, I had to level twice, so that the mount head is level and the optical tube is also level, but this is not necessary with NexStar (only the mount needs to be level).  However, you don't need a home position when using the StarSense for Sky-Watcher since it uses the NexStar alignment routine.

[Hesiod]

The answer you got from Celestron is not entirely accurate as it likely refers just to how they mounts work.
It is possible for eq mounts to take care of the drift due to PA error and several units have such feature (the original Vixen Sphinx was especially "in"famous for this).
Also a very big role is played by the "model tech" and how the alignment points are handled.
As an example, a eq mount which "knows" how is angled with respect to the sky either because uses "absolute encoders", or a given "starting position", etc...theoretically does not even need alignment points, just time and place; alignment points are just something it uses to refine its model and to compensate for potential errors, as an example a poor PA.

Speaking of which, it is not hard to get a very decent accuracy by daylight.
As said in my previous post, you basically have to take care only of the azimuthal bearing because if level your tripod do not have any reason to touch the altitude knob (you may use the latitude, or just align once the mount to Polaris by night).
To find the right azimuth may use a compass (the noaa website has a very nice tool you can run on your phone), the shades, and always can quickly "drift align" the mount on the Sun when it is in the southern quadrant (that's something I do while observing).
As far as I know sideral/sun/moon is standard on most drive kits; mind that some goto mounts will not allow you to select manually the speed but automatically adjust it depending on the target you chose (of course have to use the goto to aim the mount)

[Hesiod]

Intrigued by your results have dug out the slt and made a test with a 66/400 refractor brought to 1600mm ca via the Quark.
I followed my usual routine (level tripod through its bubble, got the time from the phone for better accuracy over my clock) and used "solar system align" on the Sun itself.
After one hour the drift was roughly 4/5 the diameter of the Sun, therefore "solar system align" does not give any meaningful improvement over your findings.
I think to remember that the slt could be set to work as an eq fork, even if to do so I would need a custom wedge*:if this is possible with your mount as well maybe this could give you a more significant improvement


*I think I could use the wedge from one of my trackers: if it is the case will run another test