118 replies to this topic

[dhaval]

I use an Edge 11, and you have to lock down the mirror, and use an external focuser. This way the model doesn't have an issue when solving on either side of the meridian due to the mirror flop.

Thanks Andrew. I might reach out to you about unguided imaging with the EdgeHD on the 10Micron mount.

 

CS!

[GShaffer]

This topic is temporarily locked and will be reopened after the mods have a chance to clean some issues up..... 

[Arie]

Not locked anymore.

Please continue our conversation.

[SimonIRE]

Yes. I’m using the same settings I had for my 2000 a few weeks ago with the same scope so it should be ok.

I do need to change the DST; that’s obviously wrong and hopefully is the only issue. It makes sense too; polemaster and PHD2, two applications independent of the mount computer suggest arc minute PAE which should be amenable to plate solving (albeit slow); unless the mount computer doesn’t know where it’s at due to an incorrect internal setting (ie DST on when it should be off) and is pointing way off.

 

This morning I woke super early (3am) and noticed it was clear so I opened up the observatory and started model building. 

 

This time I correctly turned off DST and plate solving worked like a charm - model building resumed. 

 

Note to those with systematic errors plate solving while model building; make sure your time/zone/DST settings are correct. If the error you are getting is repeatable despite PA being decent, then it's likely to be a "wrong" setting in the mount computer. 

 

Admittedly a rookie error on my part. 

[Tonk]

Following on from Simon's miss set DST option and thus not having the correct 10Micron controller box settings for model building - does anyone have direct experience of building a model in the *southern hemisphere*.

I'm going to find out about this in a few months time and I've armed myself with a few controller settings I *think* I need to change but it would be useful to hear from anyone who as actually done this already and have ironed out the necessary setting changes.

TIA

 

If no-one has then I'll be back with what I experienced - stay tuned.

[SimonIRE]

Hi Folks,

I have been building models with a new 10 Micron 4000 HPS II. This mount is set up in an observatory with 4 scopes on it, each for a different application. I experienced a number of quirks (compared to my GM2000) that I would be keen to get opinions on here.

First some background. The configuration of the scopes is as follows





The largest scope is an Edge 1100 with Bisque rings (Planetary/lunar) which allow mounting of a 73mm WO Zenithstar on the side. The other two scopes are an APM LZOS 152mm (deep sky AP) and a Takahashi TSA 120mm (Solar). The system is as rigid as one can reasonably make it (short of doing chemistry experiments with ring felt ). Its mounted on a Baader/Planewave 8 inch plate/clamp system and it is well balanced. The total load is about 78kg so well under the mounts capacity.

I planned to polar align with the WO and then create models in the WO, APM and TSA. Keep in mind that I wanted to get a very good polar alignment (sub 1 arcminute with a low RMS). With my GM2000 and an FSQ106 it would be pretty standard to get a PAE of <30 arcseconds with an RMS of <5 arcseconds with 50 points. This was what I was aiming for. You may disagree with me being fastidious about PAE but that is not the issue I wish to discuss.

This is what happened.

I started with the WO; I ran a 25 point model and then the mounts polar alignment routine. Regulus was relatively close to the meridian south, so this was the star I used - I polar aligned. I then ran another model of 25 points. Usually with my GM2000 this single step would get me within 1-1.5 arc minutes. With the 4000, I was out by about 3-4 arcminutes. I repeated the polar alignment and re-ran a 25 point model. I got to 1.7 arcminutes with an RMS of about 20 arcseconds. I repeated again, this time with 50 points, ran the polar alignment routine and was out by 2 arcminutes with an RMS of about 30 arcseconds. I noticed the orthogonality error was around 20 arcminutes; this is something I have never really paid much attention to before with my 2000. Despite repeated attempts, I could not get a sub arcminute PAE and the RMS was consistently much higher than I would have liked.

I wondered if the position of the scope on the side of Edge was impacting the model building (because of flexure) so I tried model building in the TSA. The camera I had set up on this scope was an ASI 174mm (so not ideal, small sensor etc). On the first model (30 points) the PAE was 5 arcminutes (despite being reported as circa 1 arcminutes on the last model with the WO). For this reason, I assumed there was some flexure between the WO and the TSA. However, again, I couldn't get the PAE less than about 2 arcminutes; the best RMS I got was 17.6 arcseconds with 90 points (none removed and no clear outliers).

Thanks for reading so far.

So my questions are:

1. Is there any reason to think that you may not achieve the same sort of PAE/RMS numbers with the SAME scope on a multi-scope set up on a GM4000 vs as a single scope on a GM2000?

2. Would you have chosen a different scope to do the polar alignment i.e. the WO is mounted on the side of the Edge and I guess its possible that this scope experiences the greatest flexure and has the greatest orthogonality error? Despite the longer FL, would the LZOS be the best scope to run the polar alignment based on its position in the set i.e. closer to the declination plate of the mount?

3. Does sensor size make any difference when building models? Is a larger sensor preferable?


(Incidentally fine adjustment of the mount, despite its size and weight is child's play with the correct size ratchets; easier than with the GM2000. The shear inertia of the whole system means that there is zero wobble or "settling" which I have noticed occasionally with the 2000. Also, the set up seems to be really solid - with my 2000, if I hit the pier, I would see some vibration of a star on the monitor. With the 4000 set up [heavy pier full of sand], there wasn't a flicker).

Simon

Edited by SimonIRE, 04 February 2021 - 12:21 PM.

[Tonk]

Simon - what you need to do is strip down and centrally mount just *one* of the refractors -  and see what results you get. This will tell you more than just trying different scopes in that 4 scope config

[SimonIRE]

Simon - what you need to do is strip down and centrally mount just *one* of the refractors - and see what results you get. This will tell you more than just trying different scopes in that 4 scope config

I know you’re right.

The idea of pulling it all apart gives me a sinking feeling though. With so few clear nights it’s a pain. But I know that what you’re suggesting is the most definitive way to understand what’s going on.

[EFT]

I'm not too worried about the WO since it weighs next to nothing.  However the TSA/APM stack concerns me.  You have to be very careful of flexure with piggybacked systems.  The EdgeHD will model completely different from the other scopes.  I would expect the TSA to model different than the APM and the WO will likely model differently.  I would expect the APM to be the most stable of the group and I would pair things down to that for the polar alignment.  Then add the EdgeHD followed by the TSA and WO.  You have to go one component at time to narrow things down.  What size SBS bar do you have.  The bigger the better.  The focusers on all of the scopes need to be high quality.  I have a system with a Stellarvue 70 in it and we put a Feather Touch focuser on there to stabilizing it.  WO focusers are not particularly stiff.  The EdgeHD needs a rear-cell focuser and the primary locked down tight.  The other two refractor focuser also need to be firm.  For the polar alignment, use the APM and lock the focuser in place.  Do what you can to place the APM in the center with the TSA not above it even if you have to use some dead weight to get the RA balanced.  My only concern is how little instrument weight you can get away with as a minimum on the RA.  The PA still doesn't have to be.  Get the polar alignment and do some test imaging with the APM before you do anything else.  

[SimonIRE]

I'm not too worried about the WO since it weighs next to nothing. However the TSA/APM stack concerns me. You have to be very careful of flexure with piggybacked systems. The EdgeHD will model completely different from the other scopes. I would expect the TSA to model different than the APM and the WO will likely model differently. I would expect the APM to be the most stable of the group and I would pair things down to that for the polar alignment. Then add the EdgeHD followed by the TSA and WO. You have to go one component at time to narrow things down. What size SBS bar do you have. The bigger the better. The focusers on all of the scopes need to be high quality. I have a system with a Stellarvue 70 in it and we put a Feather Touch focuser on there to stabilizing it. WO focusers are not particularly stiff. The EdgeHD needs a rear-cell focuser and the primary locked down tight. The other two refractor focuser also need to be firm. For the polar alignment, use the APM and lock the focuser in place. Do what you can to place the APM in the center with the TSA not above it even if you have to use some dead weight to get the RA balanced. My only concern is how little instrument weight you can get away with as a minimum on the RA. The PA still doesn't have to be. Get the polar alignment and do some test imaging with the APM before you do anything else.

Thanks Ed.

The LZOS and TSA both have FT focusers. The Edge has a Litecrawler. The mounting hardware is the Baader 8 inch clamp with the 8 inch plate and 2 19 inch Baader Pan clamps so the system is pretty stiff. I suspect the WO focuser is a bit slack.

I’ll start with the LZOS alone and work from there

[mccomiskey]

Posting an update to a line of comments that I initiated on this thread's ancestor, as I had promised I would:

 

https://www.cloudyni...5#entry10729009

 

I was having some trouble getting the RMS down when building my models.  Per recommendations, I made a series of changes:

 

1. Replaced the existing dovetail plate with a thicker, shorter plate

2. Replaced the existing scope rings with Tak scope rings that were wider

3. Repositioned the whole OTA assembly on the saddle so the scope rings are centered over the dovetail saddle on the mount

4. Added weight to the front of the dovetail plate to offset the rear-weight balance created by item 3.

5. Moved the one wire coming off the OTA from a counterweight bar attachment to running off the rear of the OTA.

6. Increased the height of the tripod to create a wider base.

 

Of course, I implemented all of these at once, so not exactly the scientific method.

 

I have a portable setup, and need to break it down and reset it up for every imaging session.

 

The results have been impressive.  From high single digits, my pointing RMS figures are now regularly between 2.8"-4".  That is based on running a 25 point model, running the polar alignment routine in the mount, then running a 60 point model (dual axis tracking turned off during the modelling, 3 seconds settling, model creator software, no points deleted).  Relatively effortless.  Given my image sampling is 3.5" (and even if it was lower!), this is near enough to perfect for my purposes.  Polar alignment much better now, also.  I stop working on refining it whenever I am under 6 minutes (field rotation not an issue at that level for my purposes (15 mins subs), but last night got to 5.3".

 

Thanks to everyone for the suggestions.

[psandelle]

Whooo-hooooo!

 

And now you never have to ever think about your mount again.

 

Paul

[Tonk]

The only thing I would cautions is

6. Increased the height of the tripod to create a wider base.

Increasing the leg length increases twisting flex at the mount head. If you want the stiffest tripod you use the shortest legs. If you want a wider base then just increase the leg angle - not the leg length

[mccomiskey]

 

The only thing I would cautions is
 

 

Increasing the leg length increases twisting flex at the mount head. If you want the stiffest tripod you use the shortest legs. If you want a wider base then just increase the leg angle - not the leg length

 

That makes sense to me.  The next time I set up, I will give it a try, if only because lower is easier.  Will report back.

[yzhzhang]

 

1. Replaced the existing dovetail plate with a thicker, shorter plate

2. Replaced the existing scope rings with Tak scope rings that were wider

3. Repositioned the whole OTA assembly on the saddle so the scope rings are centered over the dovetail saddle on the mount

4. Added weight to the front of the dovetail plate to offset the rear-weight balance created by item 3.

5. Moved the one wire coming off the OTA from a counterweight bar attachment to running off the rear of the OTA.

6. Increased the height of the tripod to create a wider base.

 

Congrats! Now it's time to revert them one at a time and tell us which one is the golden ticket!

[SimonIRE]

Following on from the comment regarding focusing and model RMS;

 

Is there any reason to believe that binning may impact RMS in the model? I always use 2x2 binning when model building and if I am honest, if I am very slightly out of focus I don't concern myself too much. 

 

If this is impacting my model building, it's a great pearl to pick up. 

 

Any thoughts?

[RazvanUnderStars]

I don't think it should have any negative impact, on the contrary, SNR will be better. Modern cameras (like yours) have a very high resolution and plate solvers downsample the images anyway to reduce computation time. So if you don't do the binning, they will.

 

For instance, ASTAP's documentation reads: "Downsampling:  For large image with a height above 3000 pixels select downsampling factor 2 or to speed up the solving and increase the signal noise ration of the stars. Also colours are combined to monochromatic so this option is beneficial for DSLR images" and "Image height in pixels after ASTAP subsampling should be somewhere between 1000 and 3000. If it is higher set subsample at 2. If you specify 0 for subsampling, the program will select a subsampling factor automatically"

[dan_1984]

Hey guys, quick question. After you model the mount what kind of guiding RMS you get? Or you go unguided all night? Do you recenter and plate solve after a while, if unguided?
I have a Mesu 200 mount, but seriously thinking about a 10 micron 3000 for my refractor

[SimonIRE]

Hey guys, quick question. After you model the mount what kind of guiding RMS you get? Or you go unguided all night? Do you recenter and plate solve after a while, if unguided?
I have a Mesu 200 mount, but seriously thinking about a 10 micron 3000 for my refractor

 

Unguided - up to about 20 minutes at 530mm on my GM2000. With pseudo-guiding - i.e. dual axis tracking on and exposures every 15 seconds, under 0.4 is typical and on the best nights, under 0.3. But it is the consistency night after night after night that you are paying for. I never re-centre and plate solve, ever.

 

Be careful here...this might be considered off topic by some and we'll get shut down. 

[EFT]

Per Frejvall was considered an expert on 10Micron mounts and his opinion was that anything under 10 RMS was good.  A lot of people here have been shooting for much lower than that although it might be an unnecessary challenge.  When it comes to the models for 10Micron mounts, there are some systems, particularly compound reflectors, that never model down below 10 RMS but still do well depending on a variety of factors.  The number of model points run also impacts the results depending on the system being modeled.  When it comes down to it, general conclusions regarding very low RMS values (e.g., 5 or below) are difficult to make for different imaging systems since they can all model differently.  Perhaps you can conclude that two identical optical/imaging systems will behave the same and provide the same modeling results, but that is about as far as you can go.

[psandelle]

Do you recenter and plate solve after a while, if unguided?

You will never ever need to re-center and plate solve after doing the modelling (unless a moose comes and knocks into your rig and tips it over). In fact, it could screw up your model if you do it.

 

Quick story: one of the first times using my first 10Micron, I was trying a new scope out (a Borg 90). I couldn't get it to come to focus (I think I had it on Vega, as it was so off I always like to start with a recognizable star). I had to take the camera off, add a spacer, put the camera back on...still didn't come to focus. Did it again: took the camera off, changed the spacer, put the camera back on. Vega was STILL (each time) dead center in the FOV. I then did my model and went to M51...couldn't see it. Where was it? I realized I had the crosshairs on (in the software). I clicked them off...M51 was dead center...the crosshairs were covering it! That's how accurate that puppy is (and the absolute encoders compensate for shakes and rumbles...but no moose).

 

Paul

[Raginar]

Dan,

 

   I don't have any issues unguided at 1200mm for 20 minutes.  Eccentricity < .6.  I don't recenter or use platesolving anymore other than for modeling.

[dan_1984]

Still a little unclear, if you can't go longer than 20 minutes, that means there is some drift after some time, and the target won't be centered anymore...am I not getting something?

[yzhzhang]

Still a little unclear, if you can't go longer than 20 minutes, that means there is some drift after some time, and the target won't be centered anymore...am I not getting something?

You may have slight drift up and drift downs perhaps, that makes the single frame not as nice as you want if you go over say 45min; but overall you won't be drifted away for too much overall. I am guessing you won't be drifting over say 5-10", so it's hopefully not much on your FOV over the whole night.

 

On the other hand re-center is easy (depending on what software you use). why not just do that every 1 hour or so?

Edited by yzhzhang, 10 February 2021 - 01:33 PM.

[Raginar]

You may have slight drift up and drift downs perhaps, that makes the single frame not as nice as you want if you go over say 45min; but overall you won't be drifted away for too much overall. I am guessing you won't be drifting over say 5-10", so it's hopefully not much on your FOV over the whole night.

 

On the other hand re-center is easy (depending on what software you use). why not just do that every 1 hour or so?

I don't because of time... I find the image doesn't drift; it's the individual frames where something happens that causes them to exceed eccentricity if I go with long(er) exposures.

 

Stick to 15 minutes.. no issues.