10 micron performance?

[SimonIRE]

9 hours ago, DaveS said:

If you want 0.25" or better then find yourself a S/H ASA DDM. But finding hens' teeth might be easier.

Unless you can afford a DDM100 at in excess of £25k

I have a GM4000 and recently have been testing my DDM100 in Spain. The GM4000 guides beautifully with a hybrid guide-model approach. 10-15 sec pulses get it to guide at 0.15-0.20". But the DDM100, with its predictive tracking, truly doesn't need guiding. I have attached a FITs file of M45, 10 minutes unguided at 0.33"/px (QHY600) using an iDK14.5 (AG Optical). The image is nothing much, other than a demonstration of star quality at this FL with small pixels. I was really happy with this; it's an unforgiving imaging train. 

 

M45-Ha-0001.fits

Edited February 15, 2023 by SimonIRE

[DaveS]

Fabulous kit, not envious at all. Yeah, right 🤣. Actually the first time I've seen an actual real photo of the DDM100, as distinct from renders.

I'm still holding on to my two DDMs by the way!

[ONIKKINEN]

21 minutes ago, SimonIRE said:

I have a GM4000 and recently have been testing my DDM100 in Spain. The GM4000 guides beautifully with a hybrid guide-model approach. 10-15 sec pulses get it to guide at 0.15-0.20". But the DDM100, with its predictive tracking, truly doesn't need guiding. I have attached a FITs file of M45, 10 minutes unguided at 0.33"/px (QHY600) using an iDK14.5 (AG Optical). The image is nothing much, other than a demonstration of star quality at this FL with small pixels. I was really happy with this; it's an unforgiving imaging train. 

 

M45-Ha-0001.fits 116.68 MB · 3 downloads

Have no say in the mount matter, couldn't afford any of the mounts discussed in this thread 😬 but obviously want one one day.

But on the fits file, are you sure its a good idea to not guide with the mount? Looks like you have somewhere in the range of 3-4'' (or even more, depending on software used to measure) FWHM stars, which is unexpected with a high spec instrument such as this and wouldn't believe for a seconds its optical quality related. Honestly this is what i get with an 8'' newtonian on an AZ-EQ6 on an average-a bit worse than average seeing night guided at 0.6-0.8'' RMS, so either you really do want to guide or the seeing was horrendous this night. My 2 cents anyway.

Stars do look round but just too large to believe its doing a great job without guiding.

[SimonIRE]

57 minutes ago, ONIKKINEN said:

Have no say in the mount matter, couldn't afford any of the mounts discussed in this thread 😬 but obviously want one one day.

But on the fits file, are you sure its a good idea to not guide with the mount? Looks like you have somewhere in the range of 3-4'' (or even more, depending on software used to measure) FWHM stars, which is unexpected with a high spec instrument such as this and wouldn't believe for a seconds its optical quality related. Honestly this is what i get with an 8'' newtonian on an AZ-EQ6 on an average-a bit worse than average seeing night guided at 0.6-0.8'' RMS, so either you really do want to guide or the seeing was horrendous this night. My 2 cents anyway.

Stars do look round but just too large to believe its doing a great job without guiding.

The image is over exposed and I agree the stars are large. The guiding on this mount is the best I’ve seen among the mounts I’ve used. I’m not sure how you attribute the FWHM to guiding issues when other parameters indicate arguably as close to visually perfect star roundness as you could ask for. I can post some other images. 
 

The main point I was trying to make was that the 10 Micron mounts, in my experience, guide really well but do need to be guided. 

[SimonIRE]

1 hour ago, DaveS said:

Fabulous kit, not envious at all. Yeah, right 🤣. Actually the first time I've seen an actual real photo of the DDM100, as distinct from renders.

I'm still holding on to my two DDMs by the way!

It was some your posts that helped me make a move from 10 Micron to ASA. That said, the 10 Microns are easier to use (I think…I’m an ASA newbie)

[ONIKKINEN]

15 minutes ago, SimonIRE said:

The image is over exposed and I agree the stars are large. The guiding on this mount is the best I’ve seen among the mounts I’ve used. I’m not sure how you attribute the FWHM to guiding issues when other parameters indicate arguably as close to visually perfect star roundness as you could ask for. I can post some other images. 
 

The main point I was trying to make was that the 10 Micron mounts, in my experience, guide really well but do need to be guided. 

Saturated stars are not useful for FWHM measurements, and of course in did not use them so the exposure is perfectly valid. Round stars are easy to get and thats not the point of guiding at all, the point is a round and small star. Equal errors in both RA and DEC at the same time result in a round star, yours are so big there must have been a sizeable error or terrible seeing, possibly both.

[SimonIRE]

8 hours ago, ONIKKINEN said:

Saturated stars are not useful for FWHM measurements, and of course in did not use them so the exposure is perfectly valid. Round stars are easy to get and thats not the point of guiding at all, the point is a round and small star. Equal errors in both RA and DEC at the same time result in a round star, yours are so big there must have been a sizeable error or terrible seeing, possibly both.

Again, it feels like looking for things to say. To say that round stars isn’t the point of guiding isn’t right. To say that round stars are easy to achieve (in this case, without active guiding) isn’t true (and feels tone deaf). Look at all the posts here and on CNs by people struggling with trailing stars and having difficulties with guiding. 
 

And to have equal guiding errors in both RA and dec such that the stars are basically perfectly round, also seems overblown. The unguided performance of this mount….which I am now using, is phenomenal. My point, relates to this mount and 10 Micron mounts. And, I was merely trying to provide some data. 

Edited February 16, 2023 by SimonIRE

[Justin richens]

Now that mount is gorgeous!

[inFINNity Deck]

On 13/02/2023 at 21:14, dan_adi said:

Hello,

Recently I stumbled on this video of a beautiful 10 micron 3000 HPS mount.

 

But when I reached the 9:55 minute mark, I was kind of disappointed with the numbers reported on 'guiding'. Since there were no corrections made, everything was running on a model. Considering he was using a C11 Hyperstar config, at a forgiven focal length ... 0.7 " RMS is quite disappointing. 

It could be the SCT quality. These scopes should not model well.

Bottom line, do you think it's the scope at fault? or is this a regular performance of this kind of mount?

Hi Dan,

not sure what your actual thoughts behind the question is. Are you considering getting one yourself or are you just wondering about these figures?

Anyway, I do have a GM3000HPS in my observatory with some 60kg of equipment on it, consisting of five scopes, the main imaging scope being a SkyWatcher Esprit150ED with a focal length of 1050mm. To be honest I do not look at the tracking figures, all I care about is more or less round stars, so when it looks right, I am happy (I do not even bother to measure them). Yes, there are shortcomings even on a mount like this and I had a fair share of trouble with it over the past five years and only after one swap and a hardware upgrade after that it started performing to my liking. Having that said we need to keep in mind that there is no perfect mount around, see this review:

https://www.macobservatory.com/blog/2022/12/30/a-comparison-between-the-astrophysics-mach2gto-and-10micron-gm1000hps-telescope-mounts

Last year I considered switching to ASA myself (DDM100), but found their response to my question about the option to interface a weather station rather unsettling as to their opinion that this is not required at all. With my GM3000HPS the modelling RMS went from 9 arc-seconds to 5 arc-seconds thanks to the use of a weather station (measured over 74 plate-solves), so weather stations do improve performance on these mounts.

When using long focal lengths it is best to do hybrid guiding, a method where guide commands are issued only once per minute to ensure the mount stays on track. This seems to work best from what I read on the 10Micron owners forum (which sadly is only open to owners). So far I have stayed away from guiding, only used it a few times to measure drift when I had issues. Using an SCT or RASA for unguided imaging is asking for poor performance as a significant number of these scopes have serious issues with mirror stability.

I think in general people buy these mounts as they are so very easy to use, do high speed slews (up to 12 degrees per second), can do unguided tracking of any object whose ephemeris is known (programming them is a breeze), and can handle even more payload that officially is stated. Besides, it is just an awesome looking mount (but that is my biased opinion).

If you want to see unguided images taken at 1050mm focal length with the GM3000HPS:

http://www.dehilster.info/astronomy/nebula.php

http://www.dehilster.info/astronomy/galaxies.php

Currently I use 7 minutes subs for narrow-band to be around the optimum S/N ratio for my equipment and location, and I have successfully done 20m unguided exposures, but at those long exposure times success is not guaranteed.

Nicolàs

[vlaiv]

1 hour ago, SimonIRE said:

Again, it feels like looking for things to say. To say that round stars isn’t the point of guiding isn’t right. To say that round stars are easy to achieve (in this case, without active guiding) isn’t true (and feels tone deaf). Look at all the posts here and on CNs by people struggling with trailing stars and having difficulties with guiding. 

Round stars are in fact consequence of equal random error in both DEC and RA.

On a mount like that - it is sign of good tracking performance - as equal DEC and RA error must be down to seeing if mount is stable and rigid. DEC axis should not move at all and therefore DEC error due to tracking should be 0. If both DEC and RA errors are equal - then we can conclude that RA tracking part is also very close to zero. What is causing bloated stars is optics + seeing.

On low quality mount - it does not need to be indicative of good guiding performance. With low quality mount strong corrections in RA can cause vibration in DEC as well if frame is not rigid enough - which can result in equal random error in both axis - which will result in round but bloated stars.

In general - star FWHM is influenced by seeing, optics and tracking / guiding performance (or error in both axis which does not need to be equal nor truly random - as you mention trailing stars)

[DaveS]

37 minutes ago, inFINNity Deck said:

Last year I considered switching to ASA myself (DDM100), but found their response to my question about the option to interface a weather station rather unsettling as to their opinion that this is not required at all. With my GM3000HPS the modelling RMS went from 9 arc-seconds to 5 arc-seconds thanks to the use of a weather station (measured over 74 plate-solves), so weather stations do improve performance on these mounts.

When using long focal lengths it is best to do hybrid guiding, a method where guide commands are issued only once per minute to ensure the mount stays on track. This seems to work best from what I read on the 10Micron owners forum (which sadly is only open to owners). So far I have stayed away from guiding, only used it a few times to measure drift when I had issues. Using an SCT or RASA for unguided imaging is asking for poor performance as a significant number of these scopes have serious issues with mirror stability.

ASA support are right, their mounts don't need, nor would benefit from, a weather station.

The software (Sequence and Autoslew) makes two sky models. The first is a whole sky model which is for pointing , taking care of polar alignment, collimation, wedge, etc, while the second, called MLPT is done for your exposure sequence. When you start a sequence Sequence wil slew to a focus star, carry out a synch, then it will step the telescope through the imaging path making exposures at each point, the step size of which is normally 15 mins but you can alter this. For each exposure it will plate solve and calculate the errors. Finally Sequence will calculate an overall correction for the whole path and begin the sequence. If a meridian flip is needed Sequence will know this and carry out a MLPT for the first part of the run, do the flip, tesynch and carry out a second MLPT for the rest of the run.

This takes care of the parameters that a weather station is needed for to calculate eg atmospheric refraction.

[andrew s]

An interesting topic. As a sometime control engineer may I add some comments. 

Firstly,  feed forward control uses a model of the system to calculate corrections. This requires an excellent system (not just the mount) without hysteresis.  It can have "issues" as long as they are smooth and repeatable. Very few if any systems don't have some misalignment and or shift due to the changing position with respect to gravity.

They need good measurements of all the relative quantities to keep the model updated.  

Due to varying refraction, optic axis shift etc. both Dec an RA corrections will be needed. They can't correct for unmodeled factors like seeing

Feedback systems rely on measuring the control variable (star position) and correcting it based on the error. They can be very good but are limited both by the measurement, inertia of the rig and the speed of the loop.

The highest quality pro systems use both feed forward model based system and adaptive optics where they measure the wavefront errors and correct them in a fast closed loop.

Quality costs and there is no one perfect solution. 

Regards Andrew 

[SamAndrew]

The proof is in the pudding, the stars aren't round, although there are other optical issues going on here too:

http://www.theinvisibleuniverse.com/Gallery_DeepSky/index.php

Lovely images, as any normal person would conclude, but alas we have the astrophotographers curse of only looking at the corners of an image at 200% zoom 😄

The drift in RA should be due to the differential flexure between the guide cam and mirror flop in the SCT, the model could be perfectly compensating for the mirror flop for all we know, the proof would be fitting an OAG.

If someone can point me to evidence that these £10K-25k+ mounts perform as well without guiding as a guided mount, I'd be keen to see it.

[ollypenrice]

1 hour ago, SamAndrew said:

 

Lovely images, as any normal person would conclude, but alas we have the astrophotographers curse of only looking at the corners of an image at 200% zoom 😄

 

I don't!  I may be odd (as has been suggested frequently over the last seventy years) but I do not pixel peep the corners and recommend this omission whole heartedly.

lly

[inFINNity Deck]

4 hours ago, DaveS said:

ASA support are right, their mounts don't need, nor would benefit from, a weather station.

The software (Sequence and Autoslew) makes two sky models. The first is a whole sky model which is for pointing , taking care of polar alignment, collimation, wedge, etc, while the second, called MLPT is done for your exposure sequence. When you start a sequence Sequence wil slew to a focus star, carry out a synch, then it will step the telescope through the imaging path making exposures at each point, the step size of which is normally 15 mins but you can alter this. For each exposure it will plate solve and calculate the errors. Finally Sequence will calculate an overall correction for the whole path and begin the sequence. If a meridian flip is needed Sequence will know this and carry out a MLPT for the first part of the run, do the flip, tesynch and carry out a second MLPT for the rest of the run.

This takes care of the parameters that a weather station is needed for to calculate eg atmospheric refraction.

Forgive my ignorance, but can you perhaps explain how this MLPT takes into account changes in refraction during a whole night of unguided imaging (i.e. changes after this MLPT is done)?

Thanks,

Nicolàs

[DaveS]

Well, it wouldn't but I've never had any problems even when imaging until dawn. Maybe I don't have very steep temperature falls.

A note: If you are imaging several objects through the night then Sequence will, of course, make as many MLPTs.

And now I'm out.

[dan_adi]

On 16/02/2023 at 12:48, SamAndrew said:

 

If someone can point me to evidence that these £10K-25k+ mounts perform as well without guiding as a guided mount, I'd be keen to see it.

Don't think you will find evidence with 10 micron. Maybe ASA or Planewave ... 

Anywho, .. I see the benefit in ease of use, by adding some absolute encoders to a mount. But I would also guide on-top of a model. Get the best of both worlds. What the model can't correct, an off axis guider will. I do have an intent of ordering a Mesu with 26 bit Renishaws AE. They are not as expensive as I thought, around 2100 EUR.

[dan_adi]

On 16/02/2023 at 00:42, SimonIRE said:

I have a GM4000 and recently have been testing my DDM100 in Spain. The GM4000 guides beautifully with a hybrid guide-model approach. 10-15 sec pulses get it to guide at 0.15-0.20". But the DDM100, with its predictive tracking, truly doesn't need guiding. I have attached a FITs file of M45, 10 minutes unguided at 0.33"/px (QHY600) using an iDK14.5 (AG Optical). The image is nothing much, other than a demonstration of star quality at this FL with small pixels. I was really happy with this; it's an unforgiving imaging train. 

 

M45-Ha-0001.fits 116.68 MB · 7 downloads

My thoughts also .... 10 micron benefits from a little bit of guiding, while ASA and probably Planewave, Nova120 etc. will do fine on their own.

Don't know from where the difference spawns. Maybe because 10 micron don't use Renishaws? or simply there is a limit on performance with  worm drive, no matter what encoder you choose.

[DaveS]

On 16/02/2023 at 10:48, SamAndrew said:

The proof is in the pudding, the stars aren't round, although there are other optical issues going on here too:

http://www.theinvisibleuniverse.com/Gallery_DeepSky/index.php

Lovely images, as any normal person would conclude, but alas we have the astrophotographers curse of only looking at the corners of an image at 200% zoom 😄

The drift in RA should be due to the differential flexure between the guide cam and mirror flop in the SCT, the model could be perfectly compensating for the mirror flop for all we know, the proof would be fitting an OAG.

If someone can point me to evidence that these £10K-25k+ mounts perform as well without guiding as a guided mount, I'd be keen to see it.

I wasn't going to reply to this thread again but here is an image of the Coma Cluster. Red, Green, and Luminance are 10 min subs, Blue is 17.5 min. Luminance is 0.61"/px but seeing limited rather than guiding limited.

North is to the left.

[DaveS]

13 hours ago, dan_adi said:

My thoughts also .... 10 micron benefits from a little bit of guiding, while ASA and probably Planewave, Nova120 etc. will do fine on their own.

Don't know from where the difference spawns. Maybe because 10 micron don't use Renishaws? or simply there is a limit on performance with  worm drive, no matter what encoder you choose.

What sets ASA and, I think, Planewave apart is the combination of very high resolution shaft encoders, 0.02" per tick on the DDM 85, finer on the DDM 100, and Direct Drive with no backlash allowing a very stiff feedback loop. The encoders are read at 100Hz so any wind gust that might shift the OTA is immediately read and corrected.

[dan_adi]

On 19/02/2023 at 14:00, DaveS said:

What sets ASA and, I think, Planewave apart is the combination of very high resolution shaft encoders, 0.02" per tick on the DDM 85, finer on the DDM 100, and Direct Drive with no backlash allowing a very stiff feedback loop. The encoders are read at 100Hz so any wind gust that might shift the OTA is immediately read and corrected.

Indeed, 0.02" per tick means Renishaw 26 bit encoder. If DDM 100 uses finer encoders, that could be the 32 bit Renishaw.