Eyepiece undercut.

[Don Pensack]

On 19/01/2020 at 22:13, Rusted said:

I routinely drill and tap for more thumbscrews until there are three. It still isn't enough for the extenders on the long filter stack of a PST mod.
The Baader Twist-lock mechanism is very secure for holding heavy loads like the Lacerta 2" solar prism + binoviewers.
It also has the advantage that you don't need to reach tiny thumbscrews tight up against its large, flat surface.

For the worst, ever socket design and construction I offer the PST eyepiece receptacle.  Or just the entire PST itself, if you like.
Until it is turned into something much more useful.   No, not a desktop pen and pencil holder!

Two thumbscrews are better than one because two screws prevent wiggle in the eyepiece and automatically register the eyepiece barrel against the other side of the focuser.

Three screws are a bad idea, though, because the 3rd screw presses the eyepiece away from contact with the focuser and can cause registration issues thereby.

[Alien 13]

I like the undercut on my FR/FF when I have a heavy DSLR hanging off the focusser....

Alan

[Rusted]

9 hours ago, Don Pensack said:

Three screws are a bad idea, though, because the 3rd screw presses the eyepiece away from contact with the focuser and can cause registration issues thereby.

That would only be true if all three screws were tightened a little at a time. Which "might" produce a concentric "float" on the tips of the screws.
And then only if the difference in diameters between the "plug and socket" were large.

If one screw, then two are well tightened, in sequence, then the third simply provides extra resistance under heavy loads.
Try it yourself with a long PST etalon/filter stack. Now add a binoviewer on the "silly" end! Good luck with that!

I gave up on T2 because even three thumbscrews were never secure enough.
Only 2" fittings can cope and larger still would be better. I can still see half an inch of sag even with a dinky little ZWO on the end.

A "compression band" only matches its given name with three screws. Becoming a crude collet.
One screw pushes the eyepiece to one side, two is better for mechanical retention but does exactly the same.
Three screws can get a firmer grip but follows the same pattern of behaviour. 

All systems push the eyepiece against one wall of the receptacle. It cannot possibly be otherwise.
The EP is always eccentric to the optical axis with thumbscrews or compression bands.
Partial concentricity is only achieved with tight tolerances. Which would be difficult to fit and remove.

My own preference is for the Baader Click-Lock system. It grips like a mad thing!
Perfectly self centering? Who cares? How can it possibly be worse than [eccentric] thumbscrews?

My next/present project is cantilevered support for the far end of the PST stack on a new 6" f/10 H-alpha OTA. 
Not an easy task, at all, given the need for easy hand access for tuning and focusing adjustment midway.
Plus the regular changes in terminal components like camera, binoviewer or diagonal.

Now add in complete system rotation about the optical axis on an equatorial mount.
Somehow, I don't think three garden canes and rubber bands is going to do it!

 

[John]

For all our discussion of the minutiae of eyepiece design and performance, there are still compromises in the basic mechanics of the way they a held into the scope. A bit crazy that

 

[Rusted]

2 hours ago, John said:

For all our discussion of the minutiae of eyepiece design and performance,
there are still compromises in the basic mechanics of the way they a held into the scope.
A bit crazy that

 

Agreed. Undercuts are probably the manufacturer's response to the sale of very expensive eyepieces for very expensive APOs. 
APOs, being refractors, tilt upwards at the front and downwards at the back. So the expensive EPs can easily fall out onto the concrete paving.
This ignores the sale of expensive star diagonals to go with all of the above but not to expensive binoviewers to go with the above. 

A.Cynic

[Don Pensack]

Repeatable and identical registration in a focuser for collimation tools, eyepieces, or star diagonals demands one of two binding systems:

--a centering collet which squeezes down uniformly all the way around and centers the inserted accessory

--a two screw binding.  While Rusted is probably correct that a 3-screw binding would work so long as the first two screws are very tight first, I would bet that, in the field, they might not always be tightened in exactly the same sequence, which is what is necessary to always guarantee identical registration each time.  And since the linear contact on one side of the inserted accessory will add as much or more friction than a 3rd thumbscrew, I see no reason to use a 3rd screw.

Though no one has yet made such a focuser, one with a long collet from opening to, say, 1.5" further in, would work with all eyepieces, undercut or not.  You can find similar 1.25" adapters, but not, so far, focusers containing that idea.

The Baader Click-Lock system is getting close, but it's still too short internally to work well with all undercuts (there is a wide range of widths and shapes to accommodate, there).

[Alan White]

I raised the issue of undercuts with David Nagler at Astrofest last week, asking why produce kit with undercuts?

He smiled and then explained why they are so good for stopping eyepieces falling out, I said no, they make things catch and fall or get dropped, he insisted on showing all TV kit working well, which it did, no catching, but it’s all matched kit.

The problem is non standardisation, we mix and match kit, problems arise, he smiled again and said, simple solution buy all TV kit, we parted smiling after I said, I still dislike undercuts.

The initial look on his face showed I was not the first to raise the subject.

 

[John]

51 minutes ago, Don Pensack said:

Repeatable and identical registration in a focuser for collimation tools, eyepieces, or star diagonals demands one of two binding systems:

--a centering collet which squeezes down uniformly all the way around and centers the inserted accessory

--a two screw binding.  While Rusted is probably correct that a 3-screw binding would work so long as the first two screws are very tight first, I would bet that, in the field, they might not always be tightened in exactly the same sequence, which is what is necessary to always guarantee identical registration each time.  And since the linear contact on one side of the inserted accessory will add as much or more friction than a 3rd thumbscrew, I see no reason to use a 3rd screw.

Though no one has yet made such a focuser, one with a long collet from opening to, say, 1.5" further in, would work with all eyepieces, undercut or not.  You can find similar 1.25" adapters, but not, so far, focusers containing that idea.

The Baader Click-Lock system is getting close, but it's still too short internally to work well with all undercuts (there is a wide range of widths and shapes to accommodate, there).

Interesting stuff Don.

I have both the 2 inch Baader Click-Lock and the 1.25 inch one (to a T2 thread). The 2 inch one uses a collet as you say so it still wont deal with all undercut designs equally well. The 1.25 inch one uses a different grip which comprises of 3 equally spaced plastic bars about 25mm in length which squeeze in from all sides equally along the length of the eyepiece barrel as the clamping action is operated. The 1.25 inch adapter is this type:

If this design could be adopted for a 2 inch adapter as well, then maybe that would be the answer ?

 

[johninderby]

The self centering twist lock clamps on the StellaMira 80mm f/10 focuser are pretty good.

The clamping arrangement inside the drawtube seemed a little odd at first with the two black compression bands and the central gold band instead of one wide band? I realised that by using two bands that are compressed by the taper inside the gold band one of the black bands can be squeezed down to a smaller diameter than the other thus allowing for the taper or undercut of a nosepiece. You can see in the two photos of the bands that one ring can tighten more than the other on a tapered nosepiece.

The 1.25” eyepiece adaptor also has a 20mm wide clamping band which will clamp pretty well any eyepiece barrel undercuts. 

Bands on a tapered nosepiece with no pressure.

Bands when pressure applied.

1.25” adapter.

Edited February 7, 2020 by johninderby

[Rusted]

I agree with Don that the present clamping system is far too short.
It ought to grip, via a full depth collet, over the entire length of the spigot.

The present system relies on a close fit at the tip of the tube being clamped.
A far from ideal situation given the range of dimensions to be found in different receptacles.

My use and recommendation of three thumbscrews is a personal choice which I find is safe under heavy loads.
I like my 2" Baader Grip-lock too on plain 2" fittings for its grip and release again under heavy loads.

In an ideal world there would be slip-on bands to fill the undercuts when the user so desires.
A bronze sleeve could be sprung into the undercuts and simply left there.
There is no reason why some manufacturers couldn't ask for $199 each for these [Chinese manufactured] "patented" bands.
Fortunately Darwin has dictated that I am never required to struggle with expensive undercuts.
 

[Don Pensack]

Bear in mind that all eyepieces with an undercut, whether a cylindrical groove or a conical taper, have a section of smooth, full diameter, barrel above the undercut, and a

longer section of smooth, full-diameter, barrel below the undercut.

Even though a collet generally won't snag, it is essential that it extend all the way up to the opening of the adapter so it is guaranteed to grab on that small section of full diameter above the undercut.

And, it should be long enough to grab some smooth barrel below the undercut.

Some collet adapters are well-designed (Olivon, Antares) in that regard.

The red topped adapter pictured above looks to be OK in that regard, though the collet could be a little longer for safety.