Hasselblad cfv-39 Digital Back giving soft images at infinity

[neilwatson]

I have been using a Hasselblad cfv-39 Digital Back with great success over the last two years. Used the Digital Back mostly on a Silvestri technical camera and on a 503 CXI body.

For the last few weeks I have been using a Phase IQ140 V mount Digital Back on a 503 CXI body. The images came out so much better I figured something must be wrong with the Hasselblad Back.

Using the Hasselblad cfv-39 with a 40mm CFI or 50mm CFI lens I cannot get sharp infinity images until I stop down to F8. Using the Phase IQ140 V mount digital back I get sharp images with the lens wide open.

The issue is consistent across a few 503 Hasselblad bodies and a few wide angle lens.

This is not such a big issue for me as I normally have the lens stopped down to F8 or F11 when using a Hasselblad body. But it probably limits the resale of the digital back. Of course when using a bellows Silvestri Flexi Cam you can focus past where the infinity position stop is, so its not a problem.

Wondering if I should send the Digital Back to Hasselblad to be recalibrated. It was in this state when I bought it but it has taken me a couple years to work out there is a problem so the guarantee will have expired. I think the sensor needs to be slightly further away from the camera ?

Has anyone else using a Hasslelbad CFV digital back had a similar problem with soft images at infinity when lens is wide open ?

Anyone had any experience with getting a Hasselblad V back recalibrated ? any ideas of cost and time it takes.  ( have sent me dealer an email )

Is it even worth considering having the sensor position adjusted or just use the back and always stop the lens down for infinity focus.

Regards

Neil

[John R Smith]

Neil

If there is that much difference at infinity it does sound as if the sensor in the CFV-39 is setup wrongly. I have been setting up screens and lenses over the past week for infinity focus, and the difference between spot-on and rubbish is a very tiny adjustment indeed out at infinity.

I believe that for all sensor work on the CFV backs they must be returned to the factory in Denmark, which makes it a rather lengthy operation. Hasselblad UK quote £312 GBP plus VAT and carriage for the job.

John

[neilwatson]

John,

I think the sensor postion is probably just outside of acceptable tolarance.  It must have slipped through the quality control process or just been right on the limit of acceptable tolarance.  Like you say the adjustment needed is tiny.

At some point I think I will send the digital back to Hasselblad for sensor position calibration.  Waiting for a mail from my dealer to confirm price and procedure to send back.  If the price is in the ballpark you mention then its sort of a reasonable amout to pay to get a perfectly positioned sensor.   I am going to wait until I feel it is an issue.  If it took me two years to notice I must have been working around it by stopping down the lens. 

You said a couple interesting things in your reply.   Can you adjust the infinity stop position on the lens helicol ?  Do you actually adjust your own camera body to the screens you use.   What procedure do you use to adjust the postion of the screen ?  some interactive process ?

Regards

Neil

[Graham Mitchell]

If the sensor is out of alignment, it would affect images at all focal distances, not just infinity.

[neilwatson]

Graham,

Yes it must effect at all focal distances.  But I would not realise as there is no reliable measure of distance for the hasselblad v lens. ( like the Alpa high precision rings) 

 If the sensor was out of position the other way and I could focus past infinity I could have easily worked around this. I would have never realised anything was wrong as would have assumed that's how the helicols are.

Regards

Neil

[John R Smith]

You said a couple interesting things in your reply.   Can you adjust the infinity stop position on the lens helicol ?  Do you actually adjust your own camera body to the screens you use.   What procedure do you use to adjust the postion of the screen ?  some interactive process ?

Neil - You can indeed adjust the infinity stop on the old 'C' lenses, at any rate. But you should never normally need to do this, and it is tricky and involves removing the focus distance ring to access three slotted head screws underneath. The reason I was doing this was because I had built up a lens out of various cannibalised bits (the second time this year!). On the 500 C/Ms I normally use the screen position is fixed. But this last week I was putting new screens into two elderly 500Cs - I must have got hold of the last two brand new screen kits in the universe - and with these you have to adjust four screws to set the screen position. For this I used a 150mm lens of known quality and gauged infinity with everything set up on a tripod. You should really get them properly collimated, but with an 8x loupe on the screen and a black cloth over my head it seemed to work out pretty well.

I think Hasselblad may have rushed out the first release of the CFV-39. I bought mine in December 2009 and it had a sensor line fault. Eventually the sensor had to be swapped out under warranty.

John

[neilwatson]

John,

I got my digital back November 2009.  Almost exactly the same time as you.  Still going strong after two years...

At some point thinking to send the digital back to Hasselblad UK for service.  But it's a lot of trouble right now as I am staying in Japn.

I am still wondering if I am expecting too much here given the precision involved.

Plus it's swings and roundabouts as I have to stop down the wide angle lens to around f11 to avoid soft corners at which point infinity is sharp.  ( Interestingly the SWC is ok after stopping down one stop )

So the question is what exactly am I gaining by getting the sensor position calibrated.  It may be good enough for the camera and lens I am using.  Maybe a better idea to wait until I actually feel its a problem that has impacted my prints.

Regards

Neil

[neilwatson]

Managed to sort this problem out without sending the CFV-39 digital back to Hasselblad Denmark.

I got a clue of what to do from reading the Hasselblad Service manual for 503CW Camera.
It said “The Camera body focal length is inspected and adjusted by using tool v-2229. The focal length is 71.40mm plus/minus 0.03mm”.  So I figured that even a tiny fraction of a millimeter would impact the focus at infinity.  This could be in the lens stop, body length or sensor position.  Logical thing was to look at the body length as that was the easiest to check.

I got together as many old Hasselblad V bodies and wide angle lens as I could from different sources.
Tested with 5 bodies and 9 wide angle lens.

What I concluded is that there clearly is tolerance variation in the bodies and the lens.   Two lens had the infinity stop in the wrong place. One lens focused passed infinity and one stopped way before.   Using the other 7 lens I concluded that 3 bodies were in the same ballpark as each other in tolerances. The other 2 bodies were too long and caused the focus to stop before infinity.  Of the 3 bodies one was a better match to my digital back than the other two.   Logical thing to do was just buy that camera body.  ( worked out much cheaper than sending the digital back for service.  You would not believe the premium the agent adds to the service price here in Japan )

The other interesting thing I worked out is that probably the Hasselblad CFV-39  is focusing forward of infinity and the Phase IQ140 back I compared against is focusing a tiny fraction past infinity.  The phase one back is probably within an acceptable tolerance.  My Hasselblad CFV-39 is slightly out of acceptable tolerance but only just.  It is easier for me to compensate by changing the camera body than sending the digital back to Hasselblad Denmark.

All goes to show that we really do need  to match the camera body to our digital backs.  This is equivalent of shimming and makes a big difference to infinity sharpness.

Has anyone else had experiences like this ?    It is worth the effort as the Hasselblad V system is a real pleasure to use.

Regards

Neil

[ced]

For these tolerance issues everyone was always fast on the gun to shoot down the back manufacturer even after they have been supplied with the tolerances down to measurements and adjustments in microns.
Lesson to be learnt look at all the possibilities before jumping to conclusions.

[ErikKaffehr]

Hi,

Not that it helps but this is good reading: http://www.josephholmes.com/news-medformatprecision.html

Best regards
Erik

I have been using a Hasselblad cfv-39 Digital Back with great success over the last two years. Used the Digital Back mostly on a Silvestri technical camera and on a 503 CXI body.

For the last few weeks I have been using a Phase IQ140 V mount Digital Back on a 503 CXI body. The images came out so much better I figured something must be wrong with the Hasselblad Back.

Using the Hasselblad cfv-39 with a 40mm CFI or 50mm CFI lens I cannot get sharp infinity images until I stop down to F8. Using the Phase IQ140 V mount digital back I get sharp images with the lens wide open.

The issue is consistent across a few 503 Hasselblad bodies and a few wide angle lens.

This is not such a big issue for me as I normally have the lens stopped down to F8 or F11 when using a Hasselblad body. But it probably limits the resale of the digital back. Of course when using a bellows Silvestri Flexi Cam you can focus past where the infinity position stop is, so its not a problem.

Wondering if I should send the Digital Back to Hasselblad to be recalibrated. It was in this state when I bought it but it has taken me a couple years to work out there is a problem so the guarantee will have expired. I think the sensor needs to be slightly further away from the camera ?

Has anyone else using a Hasslelbad CFV digital back had a similar problem with soft images at infinity when lens is wide open ?

Anyone had any experience with getting a Hasselblad V back recalibrated ? any ideas of cost and time it takes.  ( have sent me dealer an email )

Is it even worth considering having the sensor position adjusted or just use the back and always stop the lens down for infinity focus.

Regards

Neil

[John R Smith]

Well done, Neil, a very useful piece of practical testing rather than sofa theorising.

I have been coming to much the same conclusions myself, in recent months. Now that I have three 500 bodies and an SWC, with over 16 lenses passing through my hands (of which I have kept seven), I have noticed a great deal of strange variation in results - especially at infinity. I only collect and use the early 'C' series equipment, so all of it is pretty ancient. The truth is that we never saw these problems with film, because grain masked the issue.

John

[John.Williams]

I concur with John Smith, well done Neil...good execution on deducing the major issue of the body tolerance.

I hope your thread has caught the eye of the readers to consider getting a body calibration if one has not been performed in a few years, and the tolerances of the modular equipment strategy; good for cross-compatibility, but weak in cross-compliance - i.e. Can be out of range and require corrective action by the photographer.

Your example is a practical real case of observing the modular tolerance issue common across all digital backs and the solution of dialing in your equipment so it is in tune with all components (lens, body, imaging unit)

These kinds of posts are very helpful for all,

John

[David Grover / Capture One]

Hi,

Not that it helps but this is good reading: http://www.josephholmes.com/news-medformatprecision.html

Best regards
Erik

A good article but now quite old.

One of the reasons why every single H4D has a custom calibration.  Not only mechanical but also the AF system to be in tune with the body as a whole.  Add in that focus correction for the working aperture AND true focus when needed, we have a very accurate system.

David

[John R Smith]

One of the reasons why every single H4D has a custom calibration.  Not only mechanical but also the AF system to be in tune with the body as a whole.  Add in that focus correction for the working aperture AND true focus when needed, we have a very accurate system.
David

Indeed. The snag with this approach is that suddenly the "modular" system gets a lot less modular. And the higher the megapixel count, the worse the problem becomes. With the old V-System kit, there are an awful lot of things which could be wrong -

* The sensor in the digital back could be wrongly set-up (film plane error).

* The bayonet mount to magazine mount distance in the body could be wrong (+ or - 0.03 mm tolerance).

* The focus screen or mirror position could be incorrect (leading to incorrect user judgement of focus).

* The infinity stop on the lens focus ring could be wrongly set.

John

[John R Smith]

Here's another thought -

The body of a Hasselblad 500 is aluminium, which has quite a large coefficient of expansion. If we are really talking about tolerances of 0.03mm, then just leaving the camera in the sun for ten minutes could throw the whole thing out of whack . . .

John

[ErikKaffehr]

Hi,

That is the exact reason that many high end lenses can focus beyond infinity.

Another way to see it is that lenses must be able to focus a bit past infinity so they can be focused at infinity under all conditions. So using infinity stop will result in incorrect focus in most situations.

The solution is live view, as usual.

Best regards
Erik

Here's another thought -

The body of a Hasselblad 500 is aluminium, which has quite a large coefficient of expansion. If we are really talking about tolerances of 0.03mm, then just leaving the camera in the sun for ten minutes could throw the whole thing out of whack . . .

John