PDA

View Full Version : Tolerance for aperture centring



Barry Kirsten
16-Aug-2013, 06:22
I'm about to start making some Waterhouse stops for an Ajax portrait petzval and am wondering how critical it is to get each aperture centred. Obviously we all strive to be as accurate as possible, but I'm wondering in practical terms what tolerance is allowable. Also where would you see the effect of an off-center stop - all over, or just at the edges? It seems to me that the smaller apertures would be more critical to get properly aligned on the optical axis because of the greater convergence at the smaller stop, but I don't know... I've never thought about it before. Any ideas most welcome. Many thanks.

Barry

BrianShaw
16-Aug-2013, 06:42
Hey Barry... I'm not an optical engineer, nor do I have any amateur opinion/knowledge either... but I've wondered the same thing. I suspect that centered is best since that is how it has been done since the beginning of photography. The soft focus lenses that use "sink strainers", like Fujinon SF, have off-center holes in addition to the regualr center hole. The off center holes are what leads to the aberration that causes softness. It is possible that only having off-center aperture would lead to similar softness in part or all of the image. It might depend upon where the film lies in relaton to the lens' image circle.

jp
16-Aug-2013, 12:48
In using Reinhold's lenses with cardboard waterhouse stops, there have been times where the cardboard has shifted 1/16 to 1/8 inch and I didn't notice a difference on the groundglass. I haven't used the 190 much at smaller than f8. Never done test shots to verify that as it's not hard to get things well centered.

Using basic crude tools I'd think you'd be very able to within 1/16" inch of center.

Mark Sawyer
16-Aug-2013, 16:44
Bearing in mind it would have to be off significantly to have much effect... On a Petzval, it would do funny things to the field curvature and coma, and the coverage would vignette off-center. The farther it is off-center in a soft focus lens, the more spherical aberration it will let in from the outer areas of the lens. Basically, you're moving the "taking" area off-axis and out of the optimal center.

I'd remove the front element, insert the blank Waterhouse slide and draw a circle on it around the inside of the barrel while the stop is well-seated. Then center your hole using that circle.

Vincent Pidone
16-Aug-2013, 17:00
Try making a cardboard stop with the hole way off center and see what it looks like.

If 1/4 inch off looks anything like OK, then your best shot at finding center will be fine.

Steven Tribe
17-Aug-2013, 04:23
In the early days of photography, there was much speculation as to the effects/advantages of non-spherical stops. The general concensus was that it didn't appear to have any real influence, except the obvious outline of the edge of the stop - so well known to us in rudimentary minimum-leaved modern irises.
I think a 5mm offset from the optical centre will be without any marked difference.
I don't consider myself a very careful artisan - but have never had problems in making WHS that are within 1mm of the optical centre.

Reinhold Schable
17-Aug-2013, 11:59
Also: consider the diameter of the lens.
For example, a 60mm diameter lens (having a radius of 30mm)...
3mm off axis amounts to 10% of the radius.
Is that ok for a medieval soft focus lens???
1mm off axis amounts to 0.33% and could be considered "splitting hairs".
In this case, the ground glass tells the story.

What if it were a process lens for a critical application?
Hmm, I'd probaly want a second opinion...

Reinhold

Barry Kirsten
17-Aug-2013, 14:25
Thanks everyone for your very helpful thoughts. I'm reassured that I should be able to do a reasonable job of it. The lens, a 9" Burke and James f/7 Petzval was owned by Scott from this forum who has posted many great images of his boy made with it. I ask myself why I'd want to try smaller stops given the nice 'look' of the lens... I guess just because I can. The slot is there, so why not see what the result is. I'll post some images later, just for interest. Thanks again all,

Barry