Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Yes, focus issues are very (!!!) different in different batches - big problem!

Status
Not open for further replies.
dmig said:
hi hans,

we had a problem with an actor's belt in the center of frame being sharp whilst his eyes were soft - even though we should have had a depth of field of several feet. back focus had been verified earlier in the day with a properly collimated 18mm master prime - not as wide as one would like, but surely sufficient.
Click to expand...

If a lens is collimated correctly, that doesn't mean there is no astigmatsym (sp?) or decentering. I know we recently sent in some master and ultra primes for adjustment by arri, so even though these are great lenses, these shifts can occur over time. What you are describing could be because of one of these problems.
 
sutherlandprojects said:
If a lens is collimated correctly, that doesn't mean there is no astigmatsym (sp?) or decentering. I know we recently sent in some master and ultra primes for adjustment by arri, so even though these are great lenses, these shifts can occur over time. What you are describing could be because of one of these problems.
Click to expand...

hi,

yes i can see that could be a possibility, but in this case they were a top-notch set of masters that had been on the bench at arri the night before.

it was very strange - i wish i could roll back time and see if the mounts indeed securely in place.

i've only had this problem with those lenses on the two cameras we rented that day. my own camera, which is rented out elsewhere, hasn't had this issue.

cheers all,

david
 
This may sound a little too obvious, but I adjusted the backfocus on both my cameras at the same time. Once we had a very sharp image we would do the final twist so the backfocus was tight enough for regular use.

On one camera this final twist had no problems, but in the other camera this final twist blurred the image a little bit.

So keep monitoring your image until the backfocus is properly tight. Thats all my 2 cents.

After that process we put the cameras through all the lenses in the world and there were absolutely no problems. Katana sharp!
 
I was very much in a rush after the second test. I would like to add a few more observations about the lens test as off today, which might still be interesting, even if the second test proved that the focus readings at 1m and 3m were correct on #1513 after optical collimation with the 14mm. This was pretty much all I could test, simply because there was no more time. I had to take off urgently so I couldn't check #112, but as it showed much better readings in the first place, I figured out my method was simply way to unprecise and I blamed mismatches from yesterday on #112 to my testing configuration.
After accusing Red cam #1513 to be really bad, I felt I should apologise and share the result of this test with the forum right away before I had to urgently run.

This is what happend today:
It was suggested by the technicians at FGV to actually put the camera on the collimator to see where we're are. This was something I had never putten into concideration because I thought it is not possible.
I was quite suprised that you actually use a collimator and look onto the sensor. You see round grey circles. It's all pretty dark, but it works with a good collimator.
Not sure if these round circles are actually representing the equivalent of the film plane, but you can set a pretty good collimation here. At FGV a 18mm Master Prime is used for all their collimation work, which never goes out to rental. Looking at those circles #112 was just perfect like it was.
As expected #1513 was off. From what I just recall it required something like 15mm travel on the barrel to get it to a propper setting on the collimator. When tightening the screws the focal flange distance does go off a bit, so the screws where opened again. Then the barrel was turned about 1mm to compensate for the little jump when the screws are tightend.
With a 4" barrel and 2mm of travel per revolution this would would mean camera was off around 0,09mm. Both cameras looked good and similar.

This collimation was cross checked with a chart on 1.00m on the UP14.
#112 was dead on.
#1513 was off.
To get a propper 1.00m reading with #1513 the barrel on the mount needed to be turned about 4.5mm which means 0,028mm where added to the distance of the flange focal length. This adjustment was done by 3 different persons and the offset was always between 4-5mm. Therefore it was decided to go with 4.5mm.
As I was very unsure what the collimation on the sensor would mean at all, we decided to use the adjustment of the chart at 1m. This let to very good results on 1m and 3m distance.

It is still not clear to me, why #1513 showed a difference between chart and collimator, while #112 had not a difference at all.
Master Prime 75mm was straight on after calibration with the UP14 on 1m and 3m today. I had no time to check if an offset between 1m and inf. could be reproduced between the two cameras. It would be interesting if someone can see any difference in a camera in the lower hunderts and above #1000 on a MP 75 between inf and 1m.
As #112 was proven to be very good adjusted on the test yesterday, is it pure inacuracy that a few lenses where off yesterday on this camera?

Any thoughts?
 
Break those old habits!!!

Break those old habits!!!

Sebastian,
Let me point out few things:
1. You are overcomplicating it a bit- remember the rules:
- Widest lens available, wide open, as close to the chart as possible.
- Take your time verifying your back focus adjustment.
As long as your widest lens is consistant with the rest of the set it will make the rest of your longer lenses hit the marks because they are less critical to back focus adjustment.
2. The sensor is not a film gate!!! It was not meant or manufactured to be a reference surface(unlike the film gate) I would not rely on it as the reference surface for measurements.
3. Now let's look at the numbers- You've initially substracted .09mm based on your collimator reading, then added .028 based on chart test. That brings us to the compounded amount of .062mm correction. Right close to the middle of my estimated .05-.08 range I gave you in my post earlier.(I couldn't resist:tongue: )
So there is no mystery here, just solid tech base.
BTW you are officially a lens/cam tech now. You can add this new hat to your collection.
Congrats:construction:
Jacek
 
Jacek,
very good points indeed. I'm convinced. And yes, I'v noticed your stunning estimation. Chapeau!!:sorcerer:
You're my man!

Cheers,
Sebastian
 
ready,

Is it ok to update the original post so we don't have to read through all this, however interesting this has been.
 
Supernova,
not sure how to edit the post. Look I don't have permission to. I agree this post can be confusing. Maybe a moderator can take it off.

Thanks,
Sebastian
 
Make sure you are signed in first then hit edit button at bottom right corner of post.
 
This thread is done, move along, nothing to see...

OP asked me to delete it, but I'm just going to close it for now.
 
Status
Not open for further replies.
Back
Top