Preventing mirror shift on an SCT

[Leveye]

Here's a question for all you SCT imaging owners out there. I have a late 90's Celestron C8 and in some cases when I'm imaging I add a better focuser to the visual back. When I do this can i "lock" the mirror in place by setting the stock focuser either all the way in or all the way out just till there is no more travel? If so which direction should i rotate the stock focuser knob for the best results? I'm able to obtain focus either way with the stock focuser turned out of travel and then using my Feathertouch. 

[JamesF]

I'm not aware that there's any way to lock the mirror in position on anything other than the Edge HD series, though perhaps people have diy solutions.

There does however appear to be an optimal position for the mirror (which is at neither end of its travel).  It has been discussed recently in a planetary imaging thread: http://stargazerslounge.com/topic/237247-saturn-with-asi174mm/

Briefly, the spherical aberration of the SCT is minimised when at its design effective focal length.  That only occurs at one position of the primary mirror.  The only way to measure that appears to be to put a camera at the correct distance behind the end of the baffle (it's different for each SCT model) and adjust the primary until the camera is in focus.  Unfortunately that does seem to leave very little space for kit such as secondary focusers, filter wheels and barlows.

James

[Leveye]

Thank you James. Much appreciated!

[m.tweedy]

or get one of these

http://www.sciencecenter.net/hutech/misc/focuslk/

[Leveye]

Thank you for that link Michael. I did not know of these. James that thread is very informative again thanks. So from what i gather the primary should be as far away from the secondary as possible? Please keep me informed of any future testing you do.

[Leveye]

If i use a focus lock and my Baader 2" quicklock with a 2" barreled 3x barlow and then CCD into that it measures to just about 5" which for a standard C8 is optimal back focus distance from the sensor. Very promising. Now where should the primary be set? All the way back? Middle? Can't wait to compare images.

[RobertI]

or get one of these

http://www.sciencecenter.net/hutech/misc/focuslk/

Never seen that one before looks very interesting.

As I understood it there aretwo types of SCT mirror problem; 'mirror shift' while focussing and 'mirror flop' whilst moving the telescope around. Which problem is the OP trying to resolve?

I am assuming the use of the Borg Helical focusser is designed to eliminate the mirror shift while focussing (and allows imaging at the ideal position), but what does the focus lock do? I am guessing it goes part of the way to stopping the mirror flop but can only remove the flop caused by the poor focusser, plenty of flop will remain after that?

Rob

[Leveye]

 I was referring to shift after getting great focus. With my scope only being an 8 flop is not an issue with mine and in general i take it from what I've read. I'm mainly using my SCT for planetary imaging so it really needs to stay locked down and super secure for about 2 minutes at a time. I of course refocus every time I take a new series of frames. Just want it to be the most secure it can be and stay focused.

[Leveye]

That Borg helical looks like it would add to much to the back focus after adding a barlow and camera. Over 5 inches it looks. Thats why the retrofit focuser with lock intrigues me. I can then use my short Baader quicklock instead of adding all the weight and length of the Feathertouch added to the rear along with everything else. I'd love it if Starlight would make a 2-speed, locking retrofit one.

[JamesF]

So from what i gather the primary should be as far away from the secondary as possible?

No.  The thing to understand about SCTs (and Maks) is that unlike a 'frac or newt, the position of the focal plane (ie where the image is in focus) varies.  In fact that's why they work.  Whereas with a 'frac or newt you put in an eyepiece (or camera) and move it to the focal plane, with an SCT or Mak you put in the eyepiece or camera and move the focal plane (by moving the primary mirror) to it.

The thing about moving the primary mirror, and thereby changing the spacing between the primary and secondary mirrors, is that the effective focal length of the system changes (because the overall focal length of an optical system involving multiple optical compents is partially a function of the distance between the components).  The stated focal length of the C8 is 2032mm.  But it will only be 2032mm at one specific spacing between the primary and secondary.  If the spacing is less than or greater than that then the focal length will be less or more than 2032mm (and the focal plane will be nearer or further from the back of the OTA).

The aberrations (particularly spherical aberration) inherent in the design are supposed to be minimised at the design focal length, so that would appear to be where you ideally want to place your camera.  The question then is how do you know where the mirror has to be to achieve that?  The only way that has occurred to me thus far is to set up the optical train (without any barlows or other optical components) so the camera is the distance stated in the thread I linked to from the back of the baffle tube and then get an image in focus.  That done the primary mirror could be locked off and all fine focusing done with the external focuser from that point on.

Some experimentation and ingenuity may well be required, especially if you want to use a flip mirror or you have a filter wheel, barlows and whatever else to fit into the system.  I'm not at all convinced that with all the additional hardware we use for planetary imaging that it's going to be that easy to actually achieve focus with the primary mirror in the "ideal" position.  I believe that's what Darryl (the OP in the linked thread) is now experimenting with.  It may be that we get to the point of knowing where the optimal position is and then accepting that we just have to get as close as reasonably possible.

James

[JamesF]

or get one of these

http://www.sciencecenter.net/hutech/misc/focuslk/

That looks like it would be a significant help.  The diagram showing how it works also illustrates why these problems with mirror movement occur.  There must be a small amount of play between the mirror carriage and the baffle tube otherwise the mirror just wouldn't move.  Moving the carriage by means of a lever pushed and pulled at some distance from the centre of the tube by a screw thread is going to have backlash in the thread and will tend to "rock" the mirror carriage on the baffle tube when the focuser changes direction.  Also, once focused the carriage may not be perfectly aligned with the baffle tube and the optical axis (because it's being moved by a lever from one side).  That's not a major problem because the mirror is spherical.  But it does mean that as the orientation of the OTA changes on an EQ mount over an extended period of time, the carriage can again "rock" into a new position, shifting the mirror.

We're only talking about tiny amounts of movement here, but when you're working at a focal length of perhaps six metres upwards, "tiny" is amplified quite significantly.

James

[knobby]

It makes you wonder if it has a significant effect on collimation too ?

[JamesF]

It makes you wonder if it has a significant effect on collimation too ?

That's also raised in the other thread, and I'm unable to come to a decision myself based on my limited knowledge of optics.  If the mirror doesn't move during collimation then if you're not in the ideal position you'll presumably see some spherical aberration in the images inside and outside focus.  But I think that if the star image is central then that should affect the image to the same extent in all directions radiating from the star, so it might act to blur or make the rings less distinct, but it shouldn't affect their concentricity.

As I've also posted in the other thread, I think it's also sensible to ask how much this error is and whether it is possible to perceive it either visually or in an image.  My suspicion is that visually you'd not even notice.  In an image it may depend on what focal length you're working at and how large the image is that you're interested in.

James

[Leveye]

Thank you again James. I understand much better now. Looking forward to others results here and hopefully I can contribute with that a bit myself.

[Peter Drew]

The change in focal plane postion in a two mirror system Cassegrain configuration that uses a moving primary mirror for focusing is the difference between the intermirror distance times the amplification factor squared. For example, most SCT's are F2 with an amplification by the secondary of x5 so if the intermirror distance is reduced by 1mm then the focal plane will be increased by 25mm less the 1mm of the primary movement. Very much like a Barlow action. 

[JamesF]

That's useful to know.  Thank you, Peter.  I'll have to look up the figures for the C9.25 as I have a feeling that the primary is f/2.3 so the secondary must be different as well to achieve the same f/10 I assume.

Is the "amplification factor" basically the f-number for the secondary?

James

[russ.will]

I'm a bit new to this SCT lark, but it seems to me that the Meade LX200 mirror lock seems to be ideal? In essence, the mirror carrier is threaded and slotted so that the lock, a collet around the carrier, compresses the carrier down concentrically on to baffle tube. Think self centering EP adapter or drill chuck. Once tightened at the optimal position, it's not going to flop or shift, but will require an external focuser. I can't see how any solution that locks the mirror at any point at other than around the optical axis can really work?

Russell

[Peter Drew]

Certainly not lock well, as although other means could lock the mirror they could still tip the mirror whilst locking causing collimation issues. When I used to make Maksutovs and SCT's commercially I used a completely different primary mirror movement that didn't have these "problems".    

[Peter Drew]

JamesF.  Essentially yes. 

[knobby]

So ... with a completely negative , in a bad mood kind of way !, we've just spent £1000 on a C8 Edge that is inherently badly designed, misadvertised (f10 for imaging ), at risk of easily damaging the optics by forgetting to take the locks off before refocussing and has collimation that changes each time the primary mirror changes direction and cools down slower due to the piddly little vents 

But it does look nice and has great optics !

[JamesF]

The Edge models are slightly different again in that they have a field-flattening lens arrangement in the baffle tube.

James

[Leveye]

I must say that after last night's session viewing Jupiter my stock focuser works just fine. I experienced no shift or flop at all in a 2 hour period all the while never touching the focus knob once after my initial focusing. Focus stayed perfect the entire time. The key (with my older scope i think anyways) is to defocus the object clockwise and then refocus precisely going CCW. This pushes the mirror up and takes up any backlash or slack and uses gravity to eliminate any movement after focusing. At least in my case. I still might give the Hutech a try. Seems a reasonable upgrade for the price. That or a 2 speed feathertouch.

[russ.will]

So ... with a completely negative , in a bad mood kind of way !, we've just spent £1000 on a C8 Edge that is inherently badly designed, misadvertised (f10 for imaging ), at risk of easily damaging the optics by forgetting to take the locks off before refocussing and has collimation that changes each time the primary mirror changes direction and cools down slower due to the piddly little vents

But it does look nice and has great optics !

That's about the gist of it! Doesn't ruin the views though!

In the link to Rod Mollises SCT guide posted in the 8" SCT thread, he talks about a Meade SCT that did away with the mirror shift/flop. Instead of simply sliding on the baffle tube actuated by an off-axis focuser, they did one OTA where there were three focuser mechanisms, each turned by a toothed belt linked to the other. A really simple and obvious idea, that was apparently too expensive for mass production.

I wonder if Peter's solution was similar?

Russell