Noctilux Focussing Problems

[Phil_Hawkes]

I think my Noctilux is focussing incorrectly when open to 1.2 and 1.0. It seems to focus just behind what I am focussing on. I haven't done a ruler test, but I do have (frustrating) experiences from the field. If I take my time and the object is static, then I have more success. The strange thing is that the focus seems fine at smaller apertures. I am sure that it is not just a depth of field issue.

I seem to remember someone commenting that the focus can change with aperture. Does anyone know if this applies to the Noctilux?

Has anyone else had this experience or is it just me?

I keep meaning to get my R-D1 calibrated to makes sure the camera isn't the problem. The lens may also need looking at. One of these days I'll get around to it.

Cheers,
Phil

 

[Sonnar Brian]

Have you tried the lens on a different camera, say an M3?

The 1x finder of the RD-1 and its short RF base makes focussing an F1.0 short tele (Equivalent to a 75mm lens for full-frame) pretty tricky!

I had to calibrate the RF on my Canon 7 to work with the F0.95. It has a much larger base than the RD-1 and is full frame.

 

[Rocamadour]

Phil, I have heard about this before, but I haven´t heard any explanation.

I know that Mark Davison wrote in Digital Users Report that his borrowed r-d1 was back-focusing with his noctilux... same in this link, and I think Sean Reid has mentioned something about focusing-difficulties in a thread here at the rangefinderforum.

So it is a fact that (at least some) noctilux lenses don´t focus properly with the r-d1... 🙁

 

[Rocamadour]

And if it is consistantly back-focusing the same length, I don´t understand how it could be due to short RF base...?

 

[LCT]

According to the formula b' = (e * f^2) / (k * z) where b' is the effective base length, e the visual acuity (0.0003 at approx. 1 arcmin), f the focal length, k the aperture and z the circle of confusion (0.020mm), the baselength of the R-D1's RF (38.2mm) is long enough in theory as a 50/1 lens needs 37.50mm to be focused accurately.
Now if the actual focal length of the lens is 52mm for instance, the minimum baselength ot the rangefinder shoud be 40.56mm for the same purpose.
This problem as well as the QC issues of the R-D1 can explain some focusing problems with the Noctilux but also the Summilux 50 at f/1.4 sometimes.
Best,
LCT

 

[Doug]

One other source of focus errors is the user moving slightly forward or back when pressing the shutter release. This can be an unconscious habit. With such narrow range of reasonable focus sharpness, any slight movement of either the camera or subject can cause a focus error. In particular, leaning into the picture as the shutter is released is not uncommon. Fixing the camera on a tripod for focus tests eliminates this possible factor.

 

[Phil_Hawkes]

One other source of focus errors is the user moving slightly forward or back when pressing the shutter release. This can be an unconscious habit. With such narrow range of reasonable focus sharpness, any slight movement of either the camera or subject can cause a focus error. In particular, leaning into the picture as the shutter is released is not uncommon. .

I'm not sure if I'm leaning in when pressing the shutter, but it is very possible that I do this. However, I still think the problem is more than my movement. The plane of focus often seems out by more than a foot (at distances ~4m). For example, the wall behind a person will be in focus, but not the person.

This seems to agree with the observation in the article mentioned by Rocamadour

Using an R-D1 by Henning Wulff

....the Noctilux. It was downright weird. At 1m it was off by about 3cm; the same as the other lenses. At 2.2m, it was off by 15 to 20cm, or 2 times as much as the other lenses. I haven't any idea at this time what could be the reason for this. I've tried it a number of times, and the results are the same.

I need to double check those exposures but that is how it "feels" to me. I don't think I lunge forward a foot every time I take a photo... but maybe I do!!

Fixing the camera on a tripod for focus tests eliminates this possible factor.

V. true. I hope to try this sometime soon. I will let you know how I go. Now, where has that ruler got to... 😉

Phil

 

[Phil_Hawkes]

Initial Test results

Initial Test results

A quick update following some tests with a variety of lenses on a tripod last night.

I did tests at 4m, 2m, and 1m (approximately) but not with every lens.

I took home some rulers from work... these were useful for testing focus at around 1m for 50mm and 2m for 90mm, but not so useful otherwise because the detail in the rulers was too fine. If anyone else wants to try these tests, then I would recommend also having some thing with less detail to compare against (such as a music keyboard). I tested the cameas at all apertures. I didn't notice the backfocus error changing with aperture.

Estimates for the backfocus error (with my camera) are shown below. All distances shown in mm. An asterisk "*" indicates that the detail was so fine that it was difficult to determine the error distance. K= Konica, L = Leica, V = Voigtlander.

Obj Distance     Lens        Approx Backfocus Error
1000             L50/1.0         40
1000             K50/2.0         40

2000             V28/1.9        100*
2000             L35/1.4        100*
2000             K50/2.0         80
2000             K90/2.8        100      

4000             L50/1.0        400*
4000             K50/2.0        400*

From the results thus far, it looks to me like every lens is focussing behind where I think it is focussing. In other words, my rangefinder needs adjusting...
which I knew deep down but was in denial :angel: . Ah well. I knew I should have had the rengefinder adjusted before that wedding :bang:

At the moment, there isn't much point doing more tests until I get the camera adjusted (booked in for next week). I'll do some more tests then.

If some people are interested in seeing the test shots thus far, then I can try putting together a few examples... but only if people are really interested.

Ciao,
Phil

 

[peterbilitch]

Phil,

That's interestring information, and yes it would help to clarify your testing if the images are available for viewing.

Since the distance of error is significant, it would certainly be interesting to see the comparisons of the same test with a correctly adjusted rangfinder. Assuming that you have the time and the inclination :angel:

Peter

 

[Phil_Hawkes]

Noctilux Update

Noctilux Update

An update on my Noctilux. I took my R-D1 to the local Voigtlander importer and had the rangefinder checked (they pass it on to someone who does the repairs for them). When they returned it, the tehcnician said that the focussing mechanism wasn't out by much... I am not sure if he actually corrected it (I think he did because it seems to focus better now).

This means my focussing problem is one of two things:
-lens problems, or
-problem with the distance in camera from the mounting to the sensor.

Now, I am pretty sure that the camera focuses accurately for most of my lenses... except my two 50mm lenses: Noctilux and Konica Hexanon f2. They still seem to have a back focus problem.

First regardinging the Nocti. The nocti consists of two parts:
-a mounting/focusing part
-a lens/aperture assembly.
The lens/aperture assembly screws into the mounting/focusing part. The mounting/focusing part moves the entire lens/aperture assembly in and out from the camera as you focus.

I'd noticed that the lens/aperture assembly of the noctilux was becoming loose from the mounting/focusing part. It seems that when it is screwwed in until it is tight, then it screws in past where it stopped screwing before. This means that the lens assembly is closer to the sensor that it was meant to be. I have found that I can correct for this by unscrewing the lens assembly to a certain point. When it is att eh correct point, the the lens focuses beautifully. However, when it is unscrewed to this point, then the lens assembly turns loosely and it easily moves from the optimal position. Also, since the aperture ring is part of the lens assembly, it becomes difficult to adjust the aperture. This is not so much of a problem, because I tend to shoot at a constant aperture ina given situation.

I will need to have the lens looked at. But at least I have an interum solution. I took some great photos at a wedding reception in low light, using f1.0 and 1600 iso. When I had the lens unscrewwed to the optimal positoin, then the shots were perfect... but the lens occasionally unsrewed and I ended up with a set of fuzzy photos.

My nocti-lust is back on fire! I love that lens.

Regarding the Konica Hexanon.. I have no idea why it is not focussing properly. It may have been dropped in the past. The first shots I got with this lens (on the Hexar RF) were unbelievably crisp, but I think the quality of later shots were not so good.

I hope that proves useful information for people.


One last comment... has anyone had problems with the sensor to mount distance being wrong with the R-D1. Most problems I heard of were regarding the focussing mechanism rather than the sensor to mount distance.

Phil

 

[J. Borger]

Phil
glad to hear your Noctilux is focussing fine now ..... i have always been tempted by that lens. But got it of of my wish list because of likely focussing issues.
Does the problem you describe come with age/use/wear of the lens?

I would love to see some samples with the R-D1 /Noctilux como BTW do you have some pictures up somewhere?

Han

 

[willemvelthoven]

my noct backfocuses too

my noct backfocuses too

here are some more examples taken at f 2.0 http://www.flickr.com/photos/willemvelthoven/tags/goos/

i carefully focused on the eyes... i dont have the problem with my CV 40/1.4 and my CV 75...

the unscrewing of my noct seems a very unattractive option. does any of you have experience with a more durable fix?

 

[peterbilitch]

I suspect that the problem is less related to faulty mechanics and more inherent in the mechanics of focusing at wide apertures.

At f/2.0 and wider I need to consider the fine depth of field that entails at these apertures. In other words, if the subject's face is even slightly askew within the frame (i.e. the person is not directly facing the camera lens) then the correct focus on one of the eyes will result in the other eye appearing to be out of focus.

At f/1.0 the depth of field becomes an extremely critical factor in assessing the desired focal point, such that it is all too easy to find that one is consistently failing to achieve accurate focus.

Peter

 

[Phil_Hawkes]

Han: I don't have photos up yet. I am currently cleaning up some photos on photoshop. I'll try to remember to upload them when I am done. My currrent opinion is that reliable focussing is possible with the Noctilux, provided that everything else is working properly. Over the weekendI photographed a range of subjects: from 1m to landscapes, and including people dancing. I got a majority of photos in focus at f1, once I had the lens unscrewed the correct amount.

Willem: I plan to take the camera to some people with Leica knowledge to see what can be done. I am guessing that this problem is unusual, but I imagine that there is an easy fix if you know what you are doing. If Leica can't help me, then maybe I'll get a thin washer ground to the right width to keep the lens mechanism at the right distance. I am concerned that what I am currently doing will damage the thread if I keep adjusting the lens.

Phil

 

[Phil_Hawkes]

I suspect that the problem is less related to faulty mechanics and more inherent in the mechanics of focusing at wide apertures.

At f/2.0 and wider I need to consider the fine depth of field that entails at these apertures. In other words, if the subject's face is even slightly askew within the frame (i.e. the person is not directly facing the camera lens) then the correct focus on one of the eyes will result in the other eye appearing to be out of focus.

At f/1.0 the depth of field becomes an extremely critical factor in assessing the desired focal point, such that it is all too easy to find that one is consistently failing to achieve accurate focus.

Peter

Hi Peter,

The focussing certainly becomes more critical at f/1.0!!!!. I have photos like yours with the eyes only just out of focus.

In my case there was something else going wrong in addition to the low depth of field. My lens was focussing so bad that the lens would have been focussing on the back of the ladie's head in the photo you attached. When I unscrew the lens assembly, then I can focus accurately enough to have the focussed-on subject within the in-focus zone (even if it is not at the centre of that zone).

Most of the photos I have taken recently have been in the range of 2m-8m and within that range I have been happy enough with the results. I would be happier if I didn't have to unscrew the lens🙂

Phil

 

[Ray]

My rangefinder was slightly off both horizontally and vertically so I adjusted it using my 35 Cron on the camera. After the adjustment the focus was sharp for all my lenses except my Noctilux and my 75 Summilux.

So then I decided to do an experiment. I readjusted my rangefinder using the Noctilux and the Summilux. Funny but now all my lenses focus fine with no issues. Is it possible that by using the Noctilux and Summilux to adjust range finder accuracy that there be such a difference then using my 35 Summicron? 😕

 

[Ed Schwartzreic]

Yes indeed. I used a Nikon eyepiece magnifier to make the final RF adjustments after DAG had completely adjusted all the settings, because the Nocti was still off a little bit. The magnifier did the trick, and now every lens I have tried focuses correctly.

Ed

 

[willemvelthoven]

where's the step by stap guide to ragefinder adjustment?

where's the step by stap guide to ragefinder adjustment?

Phil
i think adjusting the ragefinder is a better idea than the lens. I just recieved an other r-d1 body and my front focus problem with the noct is almost gone. it's not the lens, it's the rangefinder.

Ed and Ray:
can you point me to a page describing the process of adjusting the rf my self? you guys seem to have experience there...

maybe we shoud have a dedicated sticky forum about this...

 

[Ed Schwartzreic]

It's in the archives, except for the rangefinder base adjustment, which entails removal of the top plate, not something to be taken lightly due to the electronic connections involved. For that adjustment, a skilled camera repair expert like DAG is needed. You can however view the R-D1 innards at my wife's site: imagere.com. Click on "Ed's Works", then on "R-D1".

 

[Phil_Hawkes]

I took my Nocti in for repairs, and I was showing my R-D1 to the guys there. They noticed that the rangefinder was now not meeting at infinity for some lenses. It was particularly off for the Konica Hexanon 50mm. With some lenses the images seemed to meet. As a result, the R-D1 is visiting another technician to get the alightnment checked again.

Phil