Go Back   IceInSpace > Equipment > Astrophotography and Imaging Equipment and Discussions

Reply
 
Thread Tools Rate Thread
  #1  
Old 22-07-2018, 05:14 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Rotator Problem Is Making Me Insane

Apologies, this will be a long post.


Problem: Adding an Optec Gemini to a TEC180 (which has not been done before, or, I was the first to try this.


Solution: Precise Parts adapter based around the design of the TEC Large Focuser that came with the TEC180FL. This is a conical shaped fitting held in place with the original focuser's locking collar.


Result: Imprecise and non-repeatable fitting of Geminir to OTA. Taking a 4 min time exposure of a centered star while making a complete rotation shows about a 5 arc-min off center circle.


Experiment: Removed rotator and ordered another PrecisePart fitting so that a Glatter laser can be fitted directly on rotator. Collimated laser and projected beam 12 feet to wall. Rotation is not perfect but according to Optec is about as good as it gets and seems to be in spec for the Gemini.


Question: Is the rotational error due to eccentricity (i.e., lateral side to side motion, or tilt changes as the rotator turns? I cannot tell and I don't know how to determine this. It could be a bit of both.



Experiment 2: Placed rotator back on TEC180FL this time with a paper mask over the objective and a centered dot. Collimated laser and aimed dot to center on mark. Performed star test/rotation and still get a 5 arc-min error in rotation.


What to do? First I assume the TEC180 is collimated though it has no adjustment on the lens cell. I did check it with a Cheshire and it appears to be collimated however I found it difficult to use and I'm not certain. In any case, collimated or not, I cannot reason out that this would have any effect on the rotation test. Also, I don't feel that my ability to truly mount the rotator without any tilt is good enough using the paper mask method...but, I'm having a hard time thinking up a new method. Appeals to Precise Parts for help have fallen on deaf ears.....I'm basically stuck with this adapter and I don't know where to turn for another in terms of design. Optec initially offered to try to make one but in the end said they didn't have the correct specs on the TEC180 to proceed. Any ideas?


More questions.


1. Am I correct that the rotator test I'm doing is going to be unaffected by collimation? I reason that a star placed on the center of the CCD will be centered and remain centered during rotation regardless of collimation.



2. Tilt is problematic. Once I put a camera on the rotator there are two different sources of tilt. 1. The rotator attachment to the OTA. 2. The assembly behind the rotator. As far as #2, the fittings are a massive dovetail from the Gemini and then all screw fittings to the TEC flattener, FW, Camera. I think I can probably assume (?) that the fittings behind the Gemini are pretty close to concentric and tilt free and that the problem is probably mostly the adapter from the Gemini to the OTA. Is this a reasonable assumption?



Since there are two independent issues at play (the inherent rotator eccentricity or tilt, and the tilt of the adapter) I'm very unsure how to attack this problem. I've considered just trying to shim the adapter until I can reduce the rotational error to min. But, obviously no amount of shimming will reduce the eccentricity error (or inherent tilt error) of the rotator. I don't know how to separate these issues. And if there is tilt behind the rotator I can't see how to chase the adapter tilt issue without possibly being influenced by potential tilt behind the rotator.



Any and all advice will be greatly appreciated!


Peter
Reply With Quote
  #2  
Old 22-07-2018, 08:58 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Every additional piece of equipment makes your system that much less reliable and more susceptible to error.

I take it your TEC is remote?

As I recall the Feathertouch focuser was rotatable but if you want the rotator for framing then I understand.

I don't use a rotator and tend to just image the objects as they fall.
One less headache. If you are imaging with a 16803 the image is large enough to crop if needed.

Greg.
Reply With Quote
  #3  
Old 22-07-2018, 09:57 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hi Peter,


How do you know the camera and laser are centered on the rotator? On off-center chip (wrt the rotator) will make a stars rotate around the rotator axis, not the sensor 'axis'.


I went through this whole exercise on my system and it was pretty involved. In the end I had to shim the focuser and shim the camera separately (in different planes). It's not perfect but being remote I am hesitant to mess with it since I might 'fix it'! The closer you get to perfect the better the seeing has to be to see what you are doing.



So the order to iterate is
-collimate
-shim cam to rotator so that 180 deg rotation does nothing except move star around due reason above
-shim focuser to make aberrations even in corners


Repeat with finer collimation and small changes to shimming until happy or exasperated! And in your case you can skip the collimation step...



I found CCD inspector fairly useless for this task, the values would bounce around on consecutive readings when not doing anything to the alignment. What was really useful was Platesolve2 which analyzes 5 or more frames across focus and calculates the average focus hyperbola in the nine sectors of the chip, then displays the difference of the estimated best focus of the nine sectors relative to the center. Very robust measurement and somewhat immune from seeing variations. From these values I could calculate an approximate tilt plane and shim accordingly.


Remember it doesn't matter if the components are individually crooked, what matter is to get the CCD orthogonal to the rotator axis, then the camera+ rotator unit orthogonal to the optical axis. The camera not being centered on the rotator axis is annoying but not deal breaking.

HTH,
EB
Reply With Quote
  #4  
Old 22-07-2018, 11:44 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Hi Peter,


In addition to what Eric said...


What is the issue with having a 5arc min eccentricity? I have ecentricity when I rotate my focuser rotator due the the camera chip not being centered on axis, its minor though.




Quote:
Originally Posted by PRejto View Post


Question: Is the rotational error due to eccentricity (i.e., lateral side to side motion, or tilt changes as the rotator turns? I cannot tell and I don't know how to determine this. It could be a bit of both.


I can not imagine there will be eccentricity in the rotator alone, you might see it if you mount the rotator slightly off the optical axis, but not in the rotator. so, my guess is tilt, and im guessing the laser is not parallel with the rotators rotating axis or the laser is not perfectly collimated.



I can not see miscollimation (if you have any), having an effect on the rotation test either.



Peter, if you can get your hands on one, get a micrometer dial guage with a needle and magnetic base. mount this on your scope and place the needle on the end of the rotating tube... then rotate the rotator 360 deg and see the runout, if any. this test will tell you if you have tilt right there. This is what i did. However, any tilt between the camera and focuser should be measured with focus runs on different parts of the chip and shimmed accordingly.



To deal with tilt in the adapter between the OTA and rotator, use your laser in the focuser to see where it is pointing with respect to the objective. even if there is tilt at the mounting face of the rotator (or your laser is not collimated) a rotation of the rotator should have the laser scribe a concentric circle with respect to the objective (assuming the laser is not parallel with the rotator) if not, add shims between the OTA and rotator until it does. Now I understand that there is complexity in this step because you say there is lateral movement in how you can mount the rotator to the OTA. my suggestion would be to use a combination of shimming and lateral movement until the laser scribes a concentric circle with the objective.



Quote:
Originally Posted by PRejto View Post
1. Am I correct that the rotator test I'm doing is going to be unaffected by collimation? I reason that a star placed on the center of the CCD will be centered and remain centered during rotation regardless of collimation.



I would agree, as long as you do not change collimation while rotating, which im sure you wont.


Quote:
Originally Posted by PRejto View Post
2. Tilt is problematic. Once I put a camera on the rotator there are two different sources of tilt. 1. The rotator attachment to the OTA. 2. The assembly behind the rotator. As far as #2, the fittings are a massive dovetail from the Gemini and then all screw fittings to the TEC flattener, FW, Camera. I think I can probably assume (?) that the fittings behind the Gemini are pretty close to concentric and tilt free and that the problem is probably mostly the adapter from the Gemini to the OTA. Is this a reasonable assumption?

If it is tilt you have, I would agree, but it may not be, it could be eccentricity and a laser that is placed not parallel to the rotating axis. You might even find tilt in the ccd, that is why you need to address this with the camera on with several focus runs.


Quote:
Originally Posted by PRejto View Post
Since there are two independent issues at play (the inherent rotator eccentricity or tilt, and the tilt of the adapter) I'm very unsure how to attack this problem. I've considered just trying to shim the adapter until I can reduce the rotational error to min. But, obviously no amount of shimming will reduce the eccentricity error (or inherent tilt error) of the rotator.
From our phone conversation, try moving the rotator lateraly on the OTA to reduce the eccentricity. I appreciate it is hard to differentiate between eccentricity or if the laser is just not rotating true. so i would try it with the laser, then move the rotator laterally, and look what happens. if you see a big difference in the size of the circle the laser makes, then you know a portion of the circle is due to eccentricity.





regards
Josh
Reply With Quote
  #5  
Old 23-07-2018, 08:27 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by ericwbenson View Post
Hi Peter,


How do you know the camera and laser are centered on the rotator? On off-center chip (wrt the rotator) will make a stars rotate around the rotator axis, not the sensor 'axis'.


I went through this whole exercise on my system and it was pretty involved. In the end I had to shim the focuser and shim the camera separately (in different planes). It's not perfect but being remote I am hesitant to mess with it since I might 'fix it'! The closer you get to perfect the better the seeing has to be to see what you are doing.



So the order to iterate is
-collimate
-shim cam to rotator so that 180 deg rotation does nothing except move star around due reason above
-shim focuser to make aberrations even in corners


Repeat with finer collimation and small changes to shimming until happy or exasperated! And in your case you can skip the collimation step...



I found CCD inspector fairly useless for this task, the values would bounce around on consecutive readings when not doing anything to the alignment. What was really useful was Platesolve2 which analyzes 5 or more frames across focus and calculates the average focus hyperbola in the nine sectors of the chip, then displays the difference of the estimated best focus of the nine sectors relative to the center. Very robust measurement and somewhat immune from seeing variations. From these values I could calculate an approximate tilt plane and shim accordingly.


Remember it doesn't matter if the components are individually crooked, what matter is to get the CCD orthogonal to the rotator axis, then the camera+ rotator unit orthogonal to the optical axis. The camera not being centered on the rotator axis is annoying but not deal breaking.

HTH,
EB

I don't know! And I have no adjustments possible to fix lateral displacement. I can only adjust tilt. The biggest question in my mind is how to know whether I should adjust for tilt, or more precisely, how much tilt? I pretty much assume most of what I see is tilt due to the quirky nature of fitting the rotator adapter. But if there is a collimation issue and I try to "fix" it by tilting I will make things worse. Though, worse than what I'm seeing is hard to imagine!


I was shown a technique with CCDI where you take 10 short exposures and CCDI averages these together to give an indication of curvature. CCDI calls it collimation but is it? Or, is it more an indicator of tilt? I think this is based on FWHM across the chip. I found results pretty repeatable using this technique. I will check out Platesolve2 and I appreciate this recommendation!



I'm attaching a test jpg I out together with the last time the rotator was on the TEC180. It's pretty clear what is happening on rotation! I'm also attaching a photo that shows what happens to a centered star on a full rotation. These are with two different cameras. The star trails camera has a KAF16200CCD and the CCDI test was done with a KAF8300 chip camera after the camera was shimmed at 270 degrees. Clearly shimming the camera was not the correct order to do things in. I should have shimmed the rotator fitting first. But, still I'm confused as to how I would do that and "know" whether remaining tilt is due to the camera or rotator adapter?


re the star trails test, I subsequently discovered a screw adapter that was moving which explains the non-smooth rotation. The yellow arrow indicates the start position of a centered star.



I still guess that the majority of my tilt is coming from the rotator adapter and that I can shim quite a lot of it away. The 5 arc-sec error means that every slew must involve a plate solve and it obviates my arc-sec pointing model (last night 6.5 arc-sec RMS).



Eric, I appreciate you help quite a lot!
Attached Thumbnails
Click for full-size image (Composite Rotation.jpg)
73.1 KB31 views
Click for full-size image (Rotator Test--.jpg)
177.8 KB31 views
Reply With Quote
  #6  
Old 23-07-2018, 08:47 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Hello Josh,


You have given me much to think about! I will try to find a dial indicator for sure. One misunderstanding may come from something I've said either in my first message or in our phone conversation. I don't think I have any side to side issue in mounting the laser to the rotator. I'm using a PreciseParts adapter that screws directly into the rotator and holds a Baader Click Lock 2" which is holding a Glatter Paralyzer and finally the Glatter laser. What I see when pointing the laser 12 feet away on a wall is a smallish circle when rotating the laser 360 degrees. This could be due to eccentricity or tilt or both (or laser mis-collimation). I did rotate the laser against the paralyzer and noted only the smallest deviation probably due more to my hand shaking the whole mechanism than mis-collimation of the laser. So, I feel my test is pretty valid. Your suggestion of a dial indicator will resolve whether it is tilt or eccentricity.


I appreciate your suggestion of how I might use the laser to initially square up the mounting of the rotator and the OTA. As I mentioned in my response to Eric, I cannot make any adjustments laterally. I also cannot adjust OTA collimation at all. I hope it is good; if it isn't the only way I can address it is to ship the OTA to Yuri at TEC for a complete rebuild which he has offered to do. His latest design has a collimatable lens cell but it only fits a slightly smaller tube meaning all new fittings and rings. This is a last option!


Thanks for your help!


Peter
Reply With Quote
  #7  
Old 23-07-2018, 08:48 AM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hi Peter,

More to chew on...
Assuming your scope has an EFL=1260mm (180mm f/7 correct?)
Your plate scale is = 206225 / efl = 163.7 "/mm

Taking your stated eccentricity of 5' = 300" and dividing by the plate scale and you get ~2mm (assuming you meant diameter?). If the CCD sensor is offset by 1mm from the rotator mechanical axis for whatever reason that would explain the rotating stars around the sensor center.

This can come from a combination of (in order of importance IMO)
a) the camera itself (mounting aperture/holes not centered - direct offset)
b) the rotator bearing itself - this can be checked with a laser.
c) the adapter from rotator to camera sitting crooked on the rotator side.

a) is difficult if not impossible to verify precisely (you are not going to open the CCD chamber for this!!!).

To check for b). First ensure the laser is aligned to it's holder (rotate manually in a v-groove at observe the beam orbit on a screen a few meters away) Then if possible rotate the holder in the rotator manually while observing the distant orbit (a helper here would be good) to get a baseline for measurement uncertainty. Next operate the rotator motor to measure it's wobble.

The angle at which the adapter is sitting is = the contributing offset divided by the adapter thickness. For example if the adapter is 10 mm thick, and c) is solely responsible with offset = 1 mm, the angle is 1/10 radians = 100 mrad ~ 6 degrees. You would definitely pick this adapter as looking crooking.

Later,
EB


Looks like our posts crossed, I'll get back to you later...

Last edited by ericwbenson; 23-07-2018 at 09:00 AM.
Reply With Quote
  #8  
Old 23-07-2018, 02:15 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Quote:
Originally Posted by PRejto View Post
Hello Josh,


You have given me much to think about! I will try to find a dial indicator for sure. One misunderstanding may come from something I've said either in my first message or in our phone conversation. I don't think I have any side to side issue in mounting the laser to the rotator.

Peter

Hi Peter. Yes, i know there was no side to side movement from the laser in the rotator.


In addition... you can then turn the dial indicator 90 deg and put its tip on the side of the rotator tube to measure its eccentricity if there is any, by rotating the rotator.


Josh
Reply With Quote
  #9  
Old 23-07-2018, 06:56 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by Joshua Bunn View Post
Hi Peter. Yes, i know there was no side to side movement from the laser in the rotator.
Josh

I guess i misread this statement:


"try moving the rotator lateraly on the OTA to reduce the eccentricity"


Unfortunately, I cannot do that. There is no lateral motion of any significance.


Regards,


Peter
Reply With Quote
  #10  
Old 23-07-2018, 07:03 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
I thought you could move the rotator laterally on the back of the OTA, not the laser in the rotator (which should be snug)


I think the dial guage on the sude of the rotator while its turning, to look gor rotator eccentricity, will help alot.


Josh
Reply With Quote
  #11  
Old 23-07-2018, 07:06 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by ericwbenson View Post
.

The angle at which the adapter is sitting is = the contributing offset divided by the adapter thickness. For example if the adapter is 10 mm thick, and c) is solely responsible with offset = 1 mm, the angle is 1/10 radians = 100 mrad ~ 6 degrees. You would definitely pick this adapter as looking crooking.

Later,
EB

Eric,


Thanks for the additional math and thoughts on the issue! I would be really surprised if there is much contribution due to A-C as it is a pretty robust and simple screwed together camera assembly. But, certainly I will be looking at it closely!


One of the difficulties of this adapter is that most of it is covered by the locking collar and it is quite difficult to see if it really is square. I must say, however, that I agree with you that 6 degrees of rotator tilt should be fairly obvious.


I think there will be a dial indicator in my future. I just had a thought about how I might square up the rotator adapter using a dial indicator. It is possible to mount the rotator backwards on the OTA since both dovetails are the same. If I mount the dial indicator on the OTA I could measure tilt as the rotator turns if it is put on backwards. Once it is square I can then reverse the rotator without disturbing the fitting to the OTA. I imagine this would be much more accurate than pointing a laser at a center spot. Do you imagine this as a reasonable procedure?


Regards,
Peter
Reply With Quote
  #12  
Old 23-07-2018, 09:12 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Quote:
Originally Posted by PRejto View Post

I think there will be a dial indicator in my future. I just had a thought about how I might square up the rotator adapter using a dial indicator. It is possible to mount the rotator backwards on the OTA since both dovetails are the same. If I mount the dial indicator on the OTA I could measure tilt as the rotator turns if it is put on backwards. Once it is square I can then reverse the rotator without disturbing the fitting to the OTA. I imagine this would be much more accurate than pointing a laser at a center spot. Do you imagine this as a reasonable procedure?


Regards,
Peter

Peter, if I'm understanding you correctly, I dont think this will work since the point where the dial indicator needle is, will or should be at the same height. In other words, tilt in the adapter has just added a constant offset. If you could put the dial indicator on one side then the other and noting the difference, it would work... but you'd have to move the dial indicator the exact same amount each time.


Another alternative is to mount the dial indicator on the cross slidd of a lathe, then put your adapter and rotator if you like, in the chuck. Now you can move the dial indicator around with the cross slide, knowing exactly how much you are moving it by looking at the measurements on the cross slide wheels. I did this with my previous focuser.


Josh

Last edited by Joshua Bunn; 23-07-2018 at 09:32 PM.
Reply With Quote
  #13  
Old 23-07-2018, 10:15 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Right you are, Josh. I wasn't thinking carefully about that. It would be a constant offset on rotation and quite useless.



Peter
Reply With Quote
  #14  
Old 24-07-2018, 12:14 AM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hello Peter,
Ok maybe too much going on here. Can you put the camera on the telescope without the rotator? The goal would be to get an acceptable CCDI map (i.e. symmetric), then use this as your baseline or target configuration (maybe sorta like the 270deg CCDI image you showed?). Since you cannot collimate just shim the camera until you are content. I think you will be less frustrated with this difficult process if you have a known target to aim at.

Side note -----
the TEC180 apparently has a field curvature of 450mm (from the Baader website). This implies a built-in defocus that varies with the distance from the optical axis (center of your CCD is close enough). I worked out the formula:
defocus = fcr * ( 1 - cos(distance/fcr) )

where fcr = 450 mm in this case.
Unfortunately using CCDI doesnt tell you the defocus amount, but by taking two pics with a focus change of (defocus) between them, see if the star in the corners snap-in versus the stars in the center.
----

Can you do the measurement I outlined in b) in the previous post? (radius of orbit divided by distance from laser to screen) I think you actually have already done it but did not mention the numerical result in your first post. This angle would be a minimum amount in the shimming exercise of the camera on the rotator. The rotator to camera adapter will potentially add to this. This will maybe put you at ease knowing the rotator is at least X good.

Next put the rotator back in and don't worry about is being mounted crooked on the scope (within reason) just shim the camera wrt the rotator so you get the SAME pattern at 0 and 180, then 90 and 270, NOT the perfect symmetric pattern, the same ugly pattern at opposite angles, now the sensor is orthogonal to the rotator axis, maybe not centered but you can't fix that here so don't sweat it. The problem I see here however is the slight wobble in the image can confuse the CCDI fitting parameters and lead you astray, not sure what to do about this except don't use CCDI or be very carefully interpreting its output?

Lastly shim the rotator to make the CCDI pattern nice and symmetric (within the wobble envelope so to speak).

Yes it is annoying to have a few arcmin of pointing error just from flipping the mount and derotating when the model is good to arc-sec...(in my case 1-2' of rotate offset for 13" RMS pointing). Basically the mount models are missing a term to account for rotator motion, AFAIK there is nothing we can do about it. But since I am using ACP it just resyncs or offset slews and continues on, no problem.

Regards,
EB
Reply With Quote
  #15  
Old 24-07-2018, 07:07 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Hello Eric,


This is a very clever way of separating the two sources of error!!


The problem as I see it is that the TEC focuser (which would allow me to mount the camera system independently of the rotator) uses the same type of adapter and isn't accurate either. This would matter because I would be shimming to fix the focuser error, no?


As for "b" in the previous post, I will supply those numbers soon.


Just to clarify, are you suggesting that I do the first step without the flattener (i.e., mounting camera on TEC180 without rotator) so that the defocus to edges is more apparent and provides an easier target to find the optical center? Currently I don't have adapters that will allow me to put the camera on without the flattener. But, even so, the stock focuser is as big a problem as the Gemini!


As far as roughing out the optical center, might I use a flat frame to get there? I would look for symmetrical darkening in the corners. And the beauty of this is I could do in during the day and in relative warmth not to mention light. Doing any of this in darkness at -2 is daunting and maddeningly hard.


Peter




----------------------
Would it be a mistake to consider having an adapter made that directly uses the threads on the TEC180 to which the locking collar fits? Yes, there would be no auto-centering by fitting 2 conical parts together, and the accuracy would be entirely based on how accurate the machining was done manufacturing the adapter. However, it would be repeatable and that would make shimming much easier.



Many thanks,


Peter
Reply With Quote
  #16  
Old 24-07-2018, 09:24 AM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hello Peter,

I missed the bit about the flattener. Ok then, aligning without the Gemini is probably more difficult than it's worth since I presume you only have adapters and spacers to fit the Gemini+camera to the flattener. This also means my side note bit about field curvature is not of any use, but it does bring up the question, are the star shapes and sizes in the corners of the sensor for your best aligned config so far (the 270deg one?) satisfactory? What is making the stars bigger off-axis, coma/astig or still round just bigger?

A flat frame would be misleading IMO for collimation and tilt adjustments. For example the peak of my flats are off-center by a few millimeters, I think because of a non-perfectly orthogonal fixed primary mirror, but the system still collimates perfectly and the star images to the corners of the 16803 are very similar. The causes of vignetting and optical correction are just different.

I understand your trepidation of being out all night playing with little fiddly bits, that's why a reliable measurement system that you can trust is golden here.

Since you are comfortable using CCDI, and before you try another software, validate CCDI for your situation. That means: introduce known amounts of tilt with shims (forget the rotation bit for now just use one position angle) and see what it reports. Is the output consistent with your known input?

Use two values of tilt and look at the difference, that way you can ignore the unknown started value. Ideally you would be increasing the tilt to go against the starting value to not make it super extreme. This could take a little bit of guess work at first to go in the right direction.

If CCDIs numbers do not match up with your input then proceed with great caution and maybe you will have to use something like Platesolve2, which I can help you with but that's another learning curve.

I wouldn't make more adapters yet, unless you are certain that what you have mechanically just won't do, but it might not be the limiting factor here

Best,
EB
Reply With Quote
  #17  
Old 24-07-2018, 12:19 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Hello Eric,


Thanks for your further thoughts. Just playing the devil's advocate here, might my refractor be somewhat less sensitive to things such as non-orthogonal secondaries. etc when considering if a flat might help with the initial alignment of the rotator? I am of the opinion that I must do my very best to at least place the rotator as close to square as I possibly can. At this point the only "solutions" seem to be using a laser to aim at a mask over the objective, or "maybe" look at some flats to see if vignetting is even. Perhaps even a combination of both ideas would prove of some benefit. I don't know. (Josh suggested using a lathe with a dial indicator. Nice idea but I think impractical. How would I hold the OTA? And I don't have access to a lathe!).


I know that the earlier idea of putting a camera directly on the OTA is no longer under consideration, however, you did ask about numbers from my rotator test: The rotator was 20 feet (ca 6,000mm) from the wall and the beam moved a max of ~10 mm from center in a complete rotation. There are also 2 times where the displacement is quite rapid, almost like something is slipping in the rotator. I'd appreciate it if you could take a look at this video and give an opinion. Normal? Each square is 5x5mm.



https://drive.google.com/file/d/14PY...ew?usp=sharing


Thanks,
Peter
Reply With Quote
  #18  
Old 24-07-2018, 02:16 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Quote:
Originally Posted by PRejto View Post
Hello Eric,


Thanks for your further thoughts. Just playing the devil's advocate here, might my refractor be somewhat less sensitive to things such as non-orthogonal secondaries. etc when considering if a flat might help with the initial alignment of the rotator? I am of the opinion that I must do my very best to at least place the rotator as close to square as I possibly can. At this point the only "solutions" seem to be using a laser to aim at a mask over the objective, or "maybe" look at some flats to see if vignetting is even. Perhaps even a combination of both ideas would prove of some benefit. I don't know. (Josh suggested using a lathe with a dial indicator. Nice idea but I think impractical. How would I hold the OTA? And I don't have access to a lathe!).


I know that the earlier idea of putting a camera directly on the OTA is no longer under consideration, however, you did ask about numbers from my rotator test: The rotator was 20 feet (ca 6,000mm) from the wall and the beam moved a max of ~10 mm from center in a complete rotation. There are also 2 times where the displacement is quite rapid, almost like something is slipping in the rotator. I'd appreciate it if you could take a look at this video and give an opinion. Normal? Each square is 5x5mm.



https://drive.google.com/file/d/14PY...ew?usp=sharing


Thanks,
Peter

Hi Peter.

In my opinion a flat frame isn't going to tell you anything about tilt, nor will tilt effect the look of a flat frame. Would you agree Eric?
With my idea of a lathe, i am asuming the mounting surface of the TEC is orthogonal to the optical axis... that leaves the rotator and its adapter to test... which you can put alone in the lathe, no ned for the scope aswell.


Was tht 10mm from center on side and 10mm on the other or 10mm total runout?


A dial indicator on the side of the rotating rotator will give you the runnout... i know i keep going on aboit it



Regards, Josh
Reply With Quote
  #19  
Old 24-07-2018, 07:37 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hi Peter,

Agree with Josh about the flat frame. The vignetting pattern in a flat is due to the baffles and the lens cell aperture relative to the sensor location in the tube. The focal plane tilt is related to the orthogonality of the various adapters. Two different types of surfaces doing completely uncorrelated things, which leads to uncorrelated measurements. They only happen to look correlated when everything is setup perfectly!

Your rotation test was good. The wandering you see is likely the gear play that must be present in any moving system. I see about 5mm of radial runout in your movie (call it 6 mm to make the math easier), so that gives 6 mm / 6 m = 1 mrad of runout, which is 206". That should be OK at f/7 with an APS chip.

What is the distance from the rotator to the sensor? That number in millimeters times 1 mrad runout will equal the radius of the orbit on the CCD in microns. I'll guess for now ~40 mm. So the orbit on the CCD could be 40 um or about 7 pixels (KAF16200?). Much smaller than your picture previously IIRC. So not a worry at the moment.

So you can try to mount the rotator with the laser in it to point at the center of the primary lens with paper cover as you described previously. That sorta helps, I'm just not sure how precise it will be. When shimming you will be at the few mrads level in resolution, that corresponds to less than 2mm offset of your laser dot on your objective.

BTW where is the flattener in all this, before or after the rotator? You could post a pic of the optical train if you like, that might help in the conversation.

Regards,
EB
Reply With Quote
  #20  
Old 24-07-2018, 07:53 PM
Joshua Bunn's Avatar
Joshua Bunn (Joshua)
Registered User

Joshua Bunn is offline
 
Join Date: Oct 2011
Location: Albany, Western Australia
Posts: 1,459
Nice reading Eric!
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 04:28 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement