Filter a distance from the lens surface

[philthejuggler]

Hi all

I'm an obsessive filter user. I've just picked up a 50mm Makro (Zeiss) with a recessed front element. Will the relatively large distance between the filter and the front element impact on IQ any more or less than it does on other lenses?

Thanks

Phil

 

[Canon Bob]

Internal reflections become worse as the filter to lens surface increases. I guess you have to balance the risk against needing the effect of the filter (unless you're simply talking about protection).

Bob

 

[philthejuggler]

Internal reflections become worse as the filter to lens surface increases. I guess you have to balance the risk against needing the effect of the filter (unless you're simply talking about protection).

Bob

Thanks Bob - to be clear, did you miss the word 'distance' in the above statement?

Thanks

Phil

 

[Canon Bob]

Thanks Bob - to be clear, did you miss the word 'distance' in the above statement?

Thanks

Phil

Yes sorry, multi-tasking here as I'm also the current exhibitor in the "What is it" thread and it's a very sweaty 25 degrees at midnight

Bob

 

[simonblue]

A good lens hood will help

 

[LongLensPhotography]

at longer distances the imperfections of the filter may start to have more effect as it becomes slightly more in focus. If you need a filter for effects, you have to use it. If it is just a UV it is probably not necessary as the element is recessed, and the hood should help further. My [non-zeiss] 50mm doesn't have a UV for this reason.

 

[philthejuggler]

Many thanks guys - I think I'll avoid the filter on this lens! Goes against my instinct but......

 

[Cuchulainn]

Internal reflections become worse as the filter to lens surface increases. I guess you have to balance the risk against needing the effect of the filter (unless you're simply talking about protection).

Bob

Hi bob,

I don't see how this would be the case at all. Could you explain please?

Cheers!

 

[Canon Bob]

Hi bob,

I don't see how this would be the case at all. Could you explain please?

Cheers!

A practical demonstration may be better than a technical explanation.

Grab the tube from a toilet roll or kitchen roll and go to the window. Hold the tube to you eye and look through it with the tube pressed (or very close to the glass). You wont see much of the tube. Now, slowly back away from the window (2-6") and you'll see the refelction...it gets more intense as the distance increases.

Bob

 

[Cuchulainn]

A practical demonstration may be better than a technical explanation.

Grab the tube from a toilet roll or kitchen roll and go to the window. Hold the tube to you eye and look through it with the tube pressed (or very close to the glass). You wont see much of the tube. Now, slowly back away from the window (2-6") and you'll see the refelction...it gets more intense as the distance increases.

Bob

That's not relevant. The tube is acting like a lens hood when pressed against the glass. The demonstration isn't what's happening with most filters since they are either threaded onto the lens and so can't get light coming in from the side or, in the case of slot filters, should be stacked with the filter in the slot closest to the lens first.

Thanks for the reply though - was wondering if it was something I misunderstood.

 

[Canon Bob]

Okey dokey Brian, I'll consult a guru and try to get a definitive answer as to why. Light leakage isn't the issue and it's similar to Canon's early extenders, 100/2.8 macro and the current 50/1.4.

Bob

 

[Richard King]

The reflectivity of each surface of the glass filter wont change one bit, just the origin of the reflection (relative to the sensor) will change. thus it "may" be more or less obserbable

Clearly as you move a filter forwards, if it is not hooded, the chances of light striking the rear of the sensor increase

 

[Canon Bob]

An update from my trusted source.....makes sense too

Lenses where the front element is heavily recessed tend to have a front element with very little curvature (there's no point in having an enhanced convex profile if the element is deeply embedded). It's the flat front profile which gives rise to the reflections and the distance between the filter and the front element means the relection is closer to being in focus.

I guess we've all seen the shot through a broken filter that looked pretty good.....move the broken filter forwards 25-30mm and things would become obvious very quickly.

Bob

 

[Cuchulainn]

Thanks for the update bob. I'm not even sure that it's true that you might see more of a reflection from a flatter element, but reckon it's not the case that you'd see it more due to the distance - it's a bit more involved than that.

That the curvature creates a flare is actually an interesting little problem for ray optics. I might have a play with it over the weekend - it's not something for a weeknight for me!

 

[Canon Bob]

I'm still hoping that an optical engineer will throw his two pennarth into the fray....he's not always around when needed though.

Bob

 

[arentol]

Internal reflections become worse as the filter to lens surface increases. I guess you have to balance the risk against needing the effect of the filter (unless you're simply talking about protection).

Bob

I don't know about internal reflections and such, but I do know that the closer something is to the front of your lens the less impact it has on the image if it is flawed in any way. This is just common photography knowledge (like when you shoot through a fence by pushing your lens right up against it). So of course a filter far from the front lens element is worse than one close to it, the further away it is the more it will be in focus and therefore the more negative its impact on the final image will be. This is even more the case with a lens that will be used at narrow apertures while focused very close to the front element like a 50mm macro that focuses just a few centimeters in front of the end of the lens and is often used at F11 or above.

As to the OP... I have the Zeiss 100mm Makro-Planar. These things are built like tanks. I know you know this already, but you really don't need to worry about the front element of that lens getting damaged, and you REALLY don't need to worry if you always use it with the hood on.

 

[symmetricalOCD]

From owning a Zeiss Planar T* 50/1,4 lens I found that the quality of glass is so high that even 'pro 1 hoya' filters add a lower quality to the image, contrast and colour suffer but it is very minute, the lens hood is of such high quality that I rarely use the lens without it.

 

[Richard King]

Thanks for the update bob. I'm not even sure that it's true that you might see more of a reflection from a flatter element, but reckon it's not the case that you'd see it more due to the distance - it's a bit more involved than that.

That the curvature creates a flare is actually an interesting little problem for ray optics. I might have a play with it over the weekend - it's not something for a weeknight for me!

we are talking about different things here

the reflectivity of a material is dependant of the refractive index of the material, and the medium it is in. As we are talking air, lets look at crown glass n = 1.523

The thinner lenses (higher refractive index) are more shiny. Plastics materials in common use vary from 1.4 - 1.7 and glass n=1.5 - n=1.8 (1.9 is very rare)

As R=((n1-n2)/(n1+n2))^2 we end up with

0.533/2.523^2 = 5.3% reflection from each surface of the lens. The curvature of the lens is irrelevant

Nowadays we have lenses with multiple anti reflective coatings that bring the reflectivity of each surface down to 0.1% and better. They achieve this by phase cancelling the reflection

The thing with filters is that they:

1. have no MAR or just a single layer AR coating, so are more reflective
2. often are not light sealed to the lens, allowing incident light in "round the back)

it is irrelevant where the surface is, if it is light-sealed. if you were in a blacked out studio, with a single window at the other end, you would see no reflection from the INSIDE of the window, as the room is blacked out and there is no light directly striking the inside surface of the window from within the room. This is pretty much the case for most lens elements IN a lens system (camera, telescope, binoculars, microscope etc.)

The only spanner in the works is that you can have internal reflections within a lens, however the power of these are very small, as to see one at the sensor, it has to have been reflected at least twice, and remember, for crown glass, we are talking 5% is reflected, 95 % transmitted at each surface

 

[Cuchulainn]

Heh, thanks for the info Richard, but we're not talking about different things at all.

I'm aware of the reflectivity of dielectrics due to refractive index steps.

I'm also well aware of how AR coatings work.

There's still an interesting little problem about what is observed on the sensor based on the curvature of the front element and the focal distance the lens is set to.

 

[philthejuggler]

Hmm - I don't think I'll use a filter with the lens!

 

[Canon Bob]

......
it is irrelevant where the surface is, if it is light-sealed. if you were in a blacked out studio, with a single window at the other end, you would see no reflection from the INSIDE of the window, as the room is blacked out and there is no light directly striking the inside surface of the window from within the room. This is pretty much the case for most lens elements IN a lens system (camera, telescope, binoculars, microscope etc.)

I'm not for a minute doubting what you've written Richard but it raises a question.
The later Canon Extenders had an "improved" internal coating and blacked-out rims on the elements to reduce ghosting. Is this simply that the older ones were from the film era and film presented far less chance of reflections than the filter in front the the sensor?

Bob

 

[Richard King]

Coatings have improved loads. 20 / 30 years ago, a single layer AR coat on glass was pretty special, expensive and not used on all lenses in a system. Nowadays Multi layer AR coats are the norm on everything.

The lenses are ground to fit the rim (not AR coated) and designers worked out that light hitting the edges bounced about everywhere. By using a matt black finish on the metallwork where the lenses are mounted, it reduces the issue loads. I have seen them paint the edges of the lenses too, to reduce reflections

its just a lot of progressive small steps that improve things a tiny bit, what when added together create something that really performs well

 

[Canon Bob]

Coatings have improved loads. 20 / 30 years ago, a single layer AR coat on glass was pretty special, expensive and not used on all lenses in a system. Nowadays Multi layer AR coats are the norm on everything.

The lenses are ground to fit the rim (not AR coated) and designers worked out that light hitting the edges bounced about everywhere. By using a matt black finish on the metallwork where the lenses are mounted, it reduces the issue loads. I have seen them paint the edges of the lenses too, to reduce reflections

its just a lot of progressive small steps that improve things a tiny bit, what when added together create something that really performs well

Sorry Richard, I wasn't too clear. It's the barrel coating that was improved....ie, a new black paint, not the glass. How does that improve things in a sealed system?

Bob

 

[Richard King]

it stops light bouncing about in the barrel.