MN190 Collimation: Possibly Worse.

[Shibby]

My MN190 hadn't been collimated for a long time - it's a somewhat daunting task, so I recently bought an OCAL to help out.

The collimation, as far as I can tell, was actually still pretty good but I did end up making a small tweak. However, I may have made it worse - my star shapes aren't as round as they used to be, however I can't 100% rule out my tracking/guiding as the cause.

With the help of some screenshots, can someone tell me if I have done this OK?

In image 1, you can see the initial collimation. You can see that the secondary is not centred under the focuser. My understanding is that is by design on the MN190 and you should not move it up/down the OTA.

In the above, you can see that the reflection of the primary is not quite centred either. This is where I made the (possibly incorrect) decision to tilt the secondary slightly to re-centre it. However, I can't quite get my head around whether it should actually be centred or not based on the fact that the secondary isn't centred??

Below, you can see the final collimation. [Note: The circles appear to expand a little as I change the focus]

 

[Dave scutt]

How do you get the blue, green and red circles overlaying 

[Steve Ward]

The OCAL s/w does it for you ...  

 

Edited July 28, 2022 by Steve Ward

[Shibby]

On 28/07/2022 at 17:55, Dave scutt said:

How do you get the blue, green and red circles overlaying 

Yep, it's in the OCAL software.

You align the outer circle yourself with respect to the focuser, then the inner circles remain concentric to help you collimate those elements.

Can anyone answer the question: Should the primary be centred to the focuser OR the secondary??

[wimvb]

A lot has been written about collimating the Mak-Newt. The consensus is that the secondary offset is factory preset, and it shouldn't be moved up or down the tube. Also, the corrector to primary mirror distance is critical, and should be kept as it is. (The primary mirror has double o-rings on each collimation screw, where normal newtonians have springs. This is because you shouldn't be able to push the primary up the tube, only tilt it.) Unlike an ordinary Newtonian, the optical axis of the corrector (= tube axis) and the optical axis of the primary have to coincide, and the focuser must be at 90 degrees. (In an ordinary Newtonian you can have perfect collimation with everything at an angle to the tube axis.)

https://www.cloudynights.com/topic/701041-before-i-get-crazy-mak-new-collimation/

https://www.iceinspace.com.au/forum/showthread.php?t=140193

I don't have an OCAL, and only use a (barlowed) laser collimator to collimate my 190MN. To get a baseline, I pull the primary all the way back to its support, and start from there. Because the primary is all the way back, I only need to use two screws to adjust it.

The final test is always a star test, with a star only slightly out of focus.

Edited August 1, 2022 by wimvb

[Shibby]

Thanks for you comments @wimvb

I was aware of the dangers of moving the secondary offset, so I have *not* done that! Having read several threads, I still can't quite figure out if my collimation as shown above is even theoretically correct. I'm thinking that, given a secondary is a flat mirror, the offset should not be affect the primary reflection, therefore that primary reflection should indeed be centred under the focuser, same as it is with an ordinary secondary. 

Anyway, I will try a star test the next chance I get.

[wimvb]

Regarding centring, read Vic Menard's (he's the one who literally wrote the book on collimation) reply in the above CN link, msg #19. Essentially, focuser tube, outer edge of secondary, and reflection of primary should be centered. Reflection of secondary is not centered.

[Shibby]

16 hours ago, wimvb said:

Essentially, focuser tube, outer edge of secondary, and reflection of primary should be centered.

Let me clarify my question/situation. Sorry for not being clear enough, or (more likely) still not understanding properly.

So, my scope has its secondary offset towards the primary mirror (as well as away from the focuser). My understanding is that this is a "classical offset" and is common on a Mak-Newt. Therefore, the outer edge of the secondary is NOT centred under the focuser. Please see green circle in my images.

Is that wrong? I would want to be 100% certain before moving the secondary up/down the OTA. If it's not wrong, and I should leave it where it is, my question is: should the primary reflection (red circle) still be centred under focuser.

[wimvb]

3 hours ago, Shibby said:

So, my scope has its secondary offset towards the primary mirror (as well as away from the focuser). My understanding is that this is a "classical offset" and is common on a Mak-Newt. Therefore, the outer edge of the secondary is NOT centred under the focuser. Please see green circle in my images.

The more I think of it, the more confusing it gets.

 

I tried to find a diagram that shows the setup.

https://skyandtelescope.org/astronomy-equipment/offsetting-your-secondary-mirror/

All the discussions (including the links to CN which I posted earlier) sooner or later mix ordinary Newtonian collimation with Mak-Newt collimation. It's there where it gets confusing. Here's my own reasoning (which may very well be completely wrong).

The corrector or meniscus in a Mak-Newt is fixed in place and must be at a fixed distance from the primary mirror. Also, the center line (optical axis) of both the meniscus and the primary mirror must coincide. This essenially fixes the primary mirror. (Hence the o-rings in stead of springs for adjusting the primary with regards to the cell. IF there were no mechanical tolerances, the primary could be locked in place.)

The secondary mirror sits at 45 degrees angle to the optical axis with what the S&T article calls fully offset collimation. If you move the secondary up or down the tube, you also need to move it closer to or away from the focuser. But this can't be done in a Mak-Newt. In a conventional Newtonian you can collimate and center everything by combining a fully offset collimation with a partially offset collimation. This is a combination of diagrams B and C in the S&T article. But for this to work, you need to slightly tilt the primary, and tilt the secondary an equal amount from its 45 degrees position. Tilting the primary messes up the Maksutov configuration. Hence the number 1 rule for Mak-Newt collimation:

DON'T MOVE THE SECONDARY UP OR DOWN THE TUBE.

In a Maksutov design, all focusing optical surfaces are spherical, so a very small amount of primary tilt with respect to the meniscus may not be visible in collimation, which is done at the optical axis. But it will most likely show up off axis as coma or other aberration.

If I'm correct, the view down the focuser tube of a collimated Mak-Newt should look like that from a conventional newtonian with fully offset collimation. AfaIk, the primary mirror reflection and its center marker should be concentric with the focuser edge.

The description earlier in this thread of how I collimate my Mak-Newt should also hold; pull the primary down to its cell. Adjust the tilt and rotation of the secondary with a well collimated laser and finally adjust the primary to get the laser beam back onto the collimators bullseye.

Edited August 2, 2022 by wimvb

[wimvb]

Btw, do any test on stars with short exposures and near the celestial pole, so you can neglect trailing. 

[Shibby]

@wimvb Just wanted to say thanks for your comments. I have since tweaked my polar alignment and improved the guiding and I'm actually pretty happy with the star shapes. I haven't yet had a chance to a proper star test as I don't want to remove the camera at the moment!

[wimvb]

@Shibby you can do a star test with camera. Just defocus and look at live view / streaming. That is what I usually do. Make sure that the star is on axis, as all other stars radially outwards will look out of collimation. Also, if you defocus the star too much, you won't get the same detail. Try to defocus just enough so that you can see a donut, then zoom in on the image to get a better view. The defocused stars in my off axis guider allways look horrible, like bananas. This is because the oag prism and stem are not aligned perfectly with the optical axis of the telescope, and the stars are off axis. But once focused, the stars look nice and round.

Yesterday, I recollimated my 190MN, more or less from scratch. I got good collimation with a variation of the recipe I described earlier.

1. centering the marker on the secondary with the help of hte crosshairs on a cheshire (using OCAL would also work). The focus tube and cheshire need to be pushed in just so much that you can see both the crosshairs and the marker. Normally, this step shouldn't be necessary, because any adjustments you can make on the secondary, are rotations and tilts around this marker. Unless the secondary is too far inwards or outwards, any secondary mirror rotation or tilt, won't change the position of the secondary marker by much.

2. using a barlowed laser collimator, get the shadow of the marker on the secondary to lign up with the marker on the primary. This is where you adjust the secondary. My barlowed laser collimator is just a cheap but collimated laser collimator inserted in a SW barlow.

3. get the reflected marker of the primary concentric on the target of the laser collimator. This is done by adjusting the primary mirror. Again, OCAL would work in a similar fashion.

A star test with my astro camera showed excellent alignment. During the collimation, I didn't bother with having anything concentric, but once done, all that needed be concentric, was. 

[coenie777]

Your secondary position looks similar to what mine did before I removed the secondary and offset it by 3.5mm on the stalk. Yes I moved the secondary position... I still regret it although I now start to think that it is not such a bad thing.

My factory fitted secondary had its dot applied roughly at 2mm away from center. Since the offset for this scope calls for around 3.7mm, centering the  secondary dot under the chesire/sight tube, will result in the secondary sitting slightly higher up in the tube towards the meniscus (as viewed from the eyepiece. You should also find that a flat frame will show the image spot off to one side away from the center of the camera chip (unless you use a very small chip).

If you round the secondary under the focuser (the secondary's outline), you are effectively aiming at a spot on the mirror roughly 3.7mm away from center. That happens automatically as you rounding the secondary mirror under the focuser is offsetting it by the desired number towards the primary. So doing that would not get you to point at the factory applied secondary mirror dot. And as noted, if you aim at the factory dot you end up with the secondary not rounded under the focuser.

I personally think that this is not a problem since you have an oversized secondary mirror to work with. As you can see from your images, your primary reflection do end up concentric to the sight tube barrel/eyepiece barrel. Having the secondary sitting high up in the view only affects your image spot position. You can get perfect collimation with an oversized secondary without the secondary outline also concentric in the eyepiece in my view.

Having moved my secondary further on the stalk now results in me getting a rounded secondary when I am aiming at my new secondary mark applied at 3.5mm. I also get perfectly centered illumination.

I am not sure what camera you image with but using a DSLR with APS-C sized chip should not be able to get pin point stars in the corners. You should get round stars up to about a diameter of 80% of the field of view. Then the stars become elongated. This is especially noticeable in the corners. If you use a chip with a size below or up to 20mm, you should have round stars all the way to the corners.

But I digress; your images taken with your scope clearly show that you have mastered the scope itself. Despite my views above, I would not recommend anyone move their secondary mirror if it has not been done before. I still need some answers from optical savvy people before I will start suggesting that.

 

[wimvb]

Maksutov telescopes need to have a fixed distance between the meniscus and the primary mirror. Also, meniscus and primary must have the same optical axis. In the original Maksutov design, meniscus and mirror are the only optical image forming elements. The image plane /focal plane is on axis. A Maksutov Newtonian has a secondary mirror to fold the light path off to one side in order to accomodate for a focuser. Where that mirror sits in the optical path is not critical, but the position of the focuser determines the position of the secondary. Moving the secondary along its reflecting surface is always possible without affecting alignment.

Edited August 11, 2022 by wimvb

[coenie777]

1 hour ago, wimvb said:

Maksutov telescopes need to have a fixed distance between the meniscus and the primary mirror. Also, meniscus and primary must have the same optical axis. In the original Maksutov design, meniscus and mirror are the only optical image forming elements. The image plane /focal plane is on axis. A Maksutov Newtonian has a secondary mirror to fold the light path off to one side in order to accomodate for a focuser. Where that mirror sits in the optical path is not critical, but the position of the focuser determines the position of the secondary. Moving the secondary along its reflecting surface is always possible without affecting alignment.

Hi Wim,

I follow your work on Astrobin as well. Nice work with the MN190!

With the MN190's focuser being able to shift up and down the tube it is clearly there to ensure focuser placement in the correct place without moving the secondary. I was told by many other users before that the distance the secondary is dropped from the meniscus is very critical. I could not find anyone yet that could explain to me how you determine this distance though. The telescope tube's construction is such that you will always have the primary mirror and meniscus at the same distance from each other, unless you raise the primary mirror with springs or larger rubber o-rings. So getting them at the right distance to each other should almost always be as set by the factory. The problem is the secondary. But maybe it is not that difficult.

The secondary needs to bring the light cone out from the tube to the eyepiece/camera. To do this in the most efficient way, it needs to find itself at 45 degrees inside the tube while capturing the whole light cone. If the whole light cone is caught with a center 5mm away from the secondary mirror's center, and if there is sufficient secondary mirror area available, and the secondary mirror remains at 45 degrees, you will meet the required condition. If alternatively you capture that light cone right in the optical center of the secondary, your collimation or details delivered from the primary/meniscus light cone will be identical to that from the 5mm mark. I am really struggling to understand how the secondary mirror's distance from the meniscus can influence the image if you meet the condition that you reflect the full light cone at exactly 45 degrees to the eyepiece. By rounding a fully offset secondary mirror under the focuser you would take care of the offset towards the primary mirror. Where the secondary sits on the center screw away from the meniscus should not influence the image as you have the ability to shift the focuser to compensate for the center shaft movement.

As stated before, the only critical component in terms of the meniscus is that it should be on the same optical axis as the primary. Once that is met, you need to just introduce the secondary as described.

Or am I completely wrong in my simplified view of it? I have been struggling to answer that the past four years with this scope, without much luck.

Your views?

[wimvb]

10 hours ago, coenie777 said:

I follow your work on Astrobin as well. Nice work with the MN190!

Thanks!

10 hours ago, coenie777 said:

I am really struggling to understand how the secondary mirror's distance from the meniscus can influence the image if you meet the condition that you reflect the full light cone at exactly 45 degrees to the eyepiece.

You are correct of course. Where on the tube you put the focuser is irrelevant, because the only task of the secondary is to fold the light cone 90 degrees. It also needs to bring the focal plane at a sensible distance from the tube’s surface. But without any adjustments you have zero tolerance in the fabrication process. So, screws to the rescue. I am writing this on a bus on my iPhone without a proper keyboard, and can’t elaborate much more atm, but during the weekend I’ll try to write in more detail. The short version is that to get even illumination you need to slide the secondary along its reflecting surface once it is at 45 degrees opposite the focuser. But you can’t do this with a fixed distance between the secondary center and the cone’s axis (lateral offset).

[wimvb]

I don't have a proper drawing app on my tablet, therefore low tech rather than high tech.

If you lower the secondary mirror (green), you also move the reflected light cone down, and it won't be centered under the focuser anymore (bottom green). You could of course lower the focuser as well, but there's little point in doing that. You would also need a larger secondary to cover the wider light cone. The position of the focuser determines the position of the secondary.

If you have a large enough secondary mirror, you can slide it along the plane of its reflecting surface without upsetting collimation or the Maksutov criterion.

 

The collimation view will look asymmetric.

If you slide the secondary along the plane of its reflective surface, you move it both down (or up) and away from (or towards) the focuser. But, in a telescope, the physical centre of the secondary is at a fixed distance from the optical axis. So, unless you remove it from its stalk and glue it in another position, you can't move the secondary up or down without also moving the light cone up or down. In an optimised configuration, the secondary mirror is only slightly larger than the reflection of the primary mirror. If it would be much larger, it would obstruct the incoming light too much and you would lose contrast. If it would be smaller, you wouldn't use the scope to its full potential. Essentially the aperture would decrease, making the scope slower. In an imaging system, the aperture is determined by the element that restricts the light cone most, not necessarily the front opening of the scope. In a Newtonian telescope, this is always the primary mirror. It should never be the secondary. In an optimised aystem, there is a very narrow dark ring between the reflection of the primary mirror in the secondary, and the edge of the secondary.

Btw, if you move the secondary mirror down and tilt it to "recenter", you will also need to tilt the focuser to achieve something which resembles proper collimation. If you don't, it will look like you have unintentional tilt in the focuser.

Edited September 3, 2022 by wimvb

[coenie777]

Great information Wim, thanks so much for taking the time. Nothing wrong with your "tech" here.

You mention:

If you have a large enough secondary mirror, you can slide it along the plane of its reflecting surface without upsetting collimation or the Maksutov criterion.

This point is what I tried to convey as the possible reason for Shibby's high position of the secondary. I am suspecting that there is a margin of movement available in the 64mm secondary option for the MN190, to move the secondary past the normal Newtonian collimation's optical center position. In so doing, because of the extra space available,  you can collimate effectively and possibly only sacrifice the position of the fully illuminated field.

I had a factory set MN190 that displayed exactly what Shibby is seeing. I also had what I assume would be the same factory applied secondary dot which is not the offset distance for these scopes but a number close to 2mm. I suspect that Skywatcher apply these dots to the secondaries after they have set the scope up on a bench. They get the primary and meniscus optical axis aligned by some method and then adjust the secondary and focuser position to intercept this. When it is done, where the secondary finds itself is marked with that dot. It does not follow a normal collimation procedure but it delivers the correct position for the rest of the scope's setup. From what I could see with my factory secondary, it was not offset on the stalk. The offset was worked in by shifting the illuminated field across the face of the secondary as it had sufficient room left to do this. That is why the dot is where it is. It is not the traditional offset dot you find on a Newtonian.

Lastly, there is a reason why the focuser can slide. There is a sweet spot (I still need to figure it out), where the secondary position is optimal but not necessarily under the focuser. You then should not tilt the secondary to make up for this but rather slide the focuser over the right intersection. 

I need to read through your explanation again in the light of day and will add some additional comments tomorrow. This is a really great discussion in solving the enigma of the MN190. I hope Shibby does not feel like we completely hijacked his post.

 

[wimvb]

@coenie777, where did you get the 3.7 mm offset for this scope? If I use the formula given by Suyter, I get close to 2mm. And I also have an uneven illumination, but this is easily corrected with flats.

[coenie777]

10 hours ago, wimvb said:

@coenie777, where did you get the 3.7 mm offset for this scope? If I use the formula given by Suyter, I get close to 2mm. And I also have an uneven illumination, but this is easily corrected with flats.

My 3.7mm is the actual distance on the face of the secondary. I mainly use this formula: A * (D-A )/ 4 * (f-H)

A:   Secondary minor axis

D   Primary aperture

f:   Focal length

H:  Intercept distance secondary to focal plane

This gives the offset as 2.7mm which is the distance it would be when viewed from the eyepiece, along the diagonal face. I think the hypotenuse would then be 2.7mm with the adjacent side being 3.7mm (I see it is actually 3.8mm rounded up).

One thing about this that I could not yet get clarity on is whether to use the 190mm aperture of the meniscus or the 200mm of the primary mirror in the calculation. The difference in this is 2mm.

[wimvb]

31 minutes ago, coenie777 said:

One thing about this that I could not yet get clarity on is whether to use the 190mm aperture of the meniscus or the 200mm of the primary mirror in the calculation. The difference in this is 2mm.

I guess that your formula is derived for a Newtonian, where the mirror is the aperture stop. In a MN I believe the meniscus is the aperture stop. Simply put, if you can see all three mirror clips in the Cheshire eyepiece, use the mirror. Otherwise, use the baffles or meniscus.

I have my formula from Suyter’s book Star Testing, which uses the simplified formula

Offset = H / 4F^2 = H/112 for the 190MN

A guesstimate of H is 230mm, giving an offset of about 2 mm, give or take a few 0.1mm

[wimvb]

10 hours ago, coenie777 said:

My 3.7mm is the actual distance on the face of the secondary. I mainly use this formula: A * (D-A )/ 4 * (f-H)

A:   Secondary minor axis

D   Primary aperture

f:   Focal length

H:  Intercept distance secondary to focal plane

This gives the offset as 2.7mm which is the distance it would be when viewed from the eyepiece, along the diagonal face. I think the hypotenuse would then be 2.7mm with the adjacent side being 3.7mm (I see it is actually 3.8mm rounded up).

One thing about this that I could not yet get clarity on is whether to use the 190mm aperture of the meniscus or the 200mm of the primary mirror in the calculation. The difference in this is 2mm.

I plugged in 190MN numbers in Suyters exact formulas, and got the same value as you: 2.7 mm

Here's the spreadsheet I used (LibreOffice, converted to Xcel format)

Yellow fields are input

Red fields are calculations

Offset_calculation_190MN.xls

[coenie777]

Thanks for sharing Wim. That 2.7mm is not the distance your dot needs to be from the center of the mirror but the perceived distance when viewed at 45 degrees, The actual distance on the mirror that it has to shift is roughly 3.8mm. It was my experience that my factory fitted secondary was not offset on the stalk.

[wimvb]

2 hours ago, coenie777 said:

It was my experience that my factory fitted secondary was not offset on the stalk.

Does that mean that the marking on the secondary mirror is at its physical center? I always thought it was at the optical center, hence trying to get that little circle to line up with the cheshire crosshairs.

Edited August 14, 2022 by wimvb

[coenie777]

18 hours ago, wimvb said:

Does that mean that the marking on the secondary mirror is at its physical center? I always thought it was at the optical center, hence trying to get that little circle to line up with the cheshire crosshairs.

My money is on neither. In my case it was not on the geometric center and at 2.5mm odd from geometric center, it was also not the expected offset. That said, my secondary was also definitely not offset by 3.8mm on the stalk originally. Maybe it was offset by the 2.5mm distance. I unfortunately did not take decent measures before I removed it. Since this dot placement can be found on a number of posts around the web, there had to be some reason why Skywatcher applied it as such. If you have the factory offset on the secondary, that dot should corresponds with it.  I can recall when I still had it factory installed, when that dot was under the cross-hair it always resulted in a high secondary as the original poster here shown. If you rounded the secondary under the sigh tube, you would be around 1.5mm away from the factory dot.

Your dot may still be where it should be relative to the amount of offset the secondary was given. At 2.5mm you will still show the slightly off-center illumination that you said you show. It would be highly unlikely in my view that you will find your secondary being offset by 3.7 or 3.8mm and the dot being where it is. For some reason SW only applied 2mm odd offset and marked it accordingly. It does make me wonder if someone could have gotten it wrong in the factory and used the raw offset number and not factoring in that you need to multiply it by 1.4142 to get the actual physical distance you need to move on the stalk. Just wondering... The flip side is also possible in that we are missing something about this optical design and how to offset its secondary.