One for the MacGyver curious types, FX6 + B4.

cyvideo

Veteran
A colleague of mine borrowed my Sony SEL20TC. For those not familiar, it's Sony's 2 x teleconverter which can be used with Sony's A7 series cameras and a number of Sony's longer lenses, starting with their 70-200 mm.

On returning it, he asked me, "Have you tried it on the FX6?" "No." I replied. Which got me to thinking. Be it madness or not. But being an old 2/3" shooter, I've always the hankering for the flexibility and range in one lens that a B4 servo zoom can offer. What does a B4 lens offer? It can offer a cheap solution to fill the need for a long range parfocal constant aperture lens. I've played around with adapting a B4 to an FS7, FX9 and now an FX6 before. So the two of us decided to chase this Unicorn. This is an experiment down that path.

NOTE! To be able to fit the Sony SEL20TC to the back of MTF's S16 B4 to E-Mount adapter meant, after careful measurements and calculations, I had to shim the rear flange of the MTF unit with a 0.7 mm shim. This prevents the protruding front element of the SEL20TC from meeting the rear element of the MTF S16 adapter. This still leaves sufficient back focus adjustment on the B4 lens. Though, the adjuster ends up about 50% through its adjustment in an anti-clockwise direction.

On the way back from a shoot, I diverted for a coffee at a local scenic observation point and shot the following clips using a variety of configurations that this MacGyver rig can be used in.

A B4 lens creates am 11 mm image circle without it's 2 x doubler engaged.

Add an MTF B4 S16 adapter, and we end up with an image circle expanded to 14.55 mm. The Abakus '132' S16 PL Adaptor will do he same. With either of these adapters, we generate a 14.55 mm image circle for use in the S16 crop mode offered in the FS7 and X9. The name of the Abakus '132' S16 adapter derives its name from this 1.32 expansion factor.

Let's now add a Sony 2 x E-Mount teleconverter lens, the SEL20TC behind this S16 adapter.

The image circle result of combining a B4 lens with a S16 adapter and a Sony SEL20TC 2 x teleconverte will be the creation of 29.1 mm image circle.

All Sony S35 sensors require a 27.1 mm lens image circle to cover the sensor. This 27.1 mm image sensor circle when taken from the 29.1 mm image created by the adapters above gives us a very clean OPTICALLY CORRECTED image circle with very clean crisp edges, minimal distortion and minimal CA and Coma artefacts. These two S16 adapters listed above have been optically corrected to converge and focus a B4 lens' light paths onto a single sensor plane. B4 lenses are designed to split and focus the light through a prism to three individual R, G and B sensors.

Think one step further. Now introduce the B4 lens 2 x converter into the equation. The S16 adapter is now receiving a 22 mm image circle. The Sony 2 x teleconverter is now doubling this to 44 mm. A full frame sensor needs an image circle of 43 mm to cover it. Though in reality, there is a slight amount of vignetting if using both the B4 Lens 2 x and the SEL20TC 2 x together in front of and behind the MTF S16 adapter.

What we have here is a 100% optically corrected 44 mm image circle that can work with any Sony FF sensor.

If we take an FX6 for example which has CIZ (Clear Image Zoom) the above B4 lens combo can be used in a variety of ways.

1) Optical + CIZ solution
.
The B4 lens without its 2 x doubler engaged plus a S16 adapter plus a CIZ of 1.5 will cover the FX6 full frame with a 2.7 stop light loss. You end up with a 0.7 of a stop loss from the S16 adapter and a 2 stop loss from the 2 x teleconverter. This works in FF mode in either HD or UHD / 4K. In FF HD mode, you actually have access to a 2 x range of CIZ, further extending the reach of the lens by a further 50%.

2) Optical only solution.

The B4 with its 2 x doubler engaged plus the S16 adapter and the Sony 2 x teleconverter gives you a 100 optical coverage of the FF sensor. On top of this you can add in 4K CIZ of 1.5 x. Or in HD a CIZ of 2 x.

Question? How does this actually look in reality?

Chris Young

 
Great write up and demonstration—thanks for sharing! At least on YouTube the quality for such an extreme telephoto looks pretty damn reasonable. And am I correct in assuming that's a Standard Def B4 lens?

Do you happen to know the focal length at the extremes of the wide and zoom?
 
drboffa.

Trying to keep it all in pretty simple terms it works a bit like this:

The Sony FX6 has a photosite size of 8.42(µm) (micrometers) and an effective resolution of 10.2 megapixels.

The Airy Disk Diameter [µm] of a green light with a wavelength of 520nm at F/5.6 is going to be 7.11µm. Therefore, we get no spillover onto adjacent photosites as they are in the case of the FX6 8.42µm across

Talking HD here. Every manufacturer rates their three-chip cameras by the size and pixel matrix of the individual sensors, and not the combined result. Therefore, a 2.1MP HD three-chip camera has three 2.1 MP sensors. If the effective pixel matrix is full HD at 1920 x 1080, then the resulting pixel size is approximately 2.5 μm. This requires 200 LP/mm to resolve. This is why good B4 HD lenses are rated up to 200 LP/mm. This is way in excess of the LP/mm resolution required by our FX6 sensor, which will perform well at 60 LP/mm.

Back to the FF sensor on the FX6. For an 8.47µm photosite we need 59 LP/mm (closest example I can get on the calculator to the FX6's 8.43µm.) Do the calcs yourself here:


The Sony SEL20TC used has been optimised to work with the 60 megapixel Sony A7V in both FF and APSC crop modes. I haven't been able to fault it on the FX6.

The optically corrected MTF S16 adapter delivers an image circle of 14.55 mm and has a 0.7 of a stop light loss. Divide this by the 11.0 mm diagonal of a B4 sensor, and you have a crop factor of 1.32. In other words a 7.6 mm wide end B4 lens now becomes a 10.03 mm wide end. The same 1.32 factor adds to the long end of the lens used. All of the above is purely optical. If you now add in the B4 optical 2 x on top of the Sony SEL20TC 2 x teleconverter you then double the length of the lens again. But obviously you now lose two more stops of light. Example. The 18 x used in this experiment becomes 36 x. Now calculate in the 1.32 crop factor, and it is effectively a 47.52 x lens. Crazy I know.

An HD lens with up to 200 LP/mm performance and an MTF of 100 is well and truly going to perform very well with the FX6. It is way in excess what is required by an FX6.

Larry Thorpe from Canon broadcast explains some of this tech in his discourse on some of the considerations when talking about MTF, LP/mm 2/3" and 4K.


I went into all this a few years ago when testing lenses for the Australian Navy video unit. A bit of a detailed explanation here in relation to B4s and S35 sensors:


Smaller formats need better lenses.

QUOTE

"An electronic imager has a fixed number of sensors. A lens does not have a fixed resolution. It simply has an MTF curve (or more than one, for different colors, different focal lengths, different apertures, and different focus settings). It provides greatest contrast at minimum resolution and reduced contrast as resolution increases.

So-called HD lenses may be better than their SD counterparts, but that doesn’t mean the SD lenses can’t pass HD resolutions. The HD lenses just offer more contrast at HD resolutions and, perhaps, reduced aberrations."

Much more related to the above comment can be found here:


To sum up. The very last of the best SD B4 broadcast lenses were designed to work with cameras like the highest resolution Sony BVP-90 broadcast cameras. They seem to deliver decent lens MTF curves and resolving power to pair with the FX6's 10.2 active megapixel sensor. Or basically any 4K sensor of a similar spec.

My MacGyver lash-up is just a derivation of the principles outlined by a one time regular visitor to these pages, Mitch Gross, explains the theory behind B4 lens adapter applications.


More reading if required.

MTF Curves and Lens Performance

The Airy Disk and Diffraction Limit

Blame it on Nyquist I say!


Chris Young.
 
Thank you for such a detailed explanation, Chris! I'd never even considered the variable of the image circle and sensor size when taking into account the resolving power of those old B4 lenses. I really appreciate that.

I don't have any old B4's or adapters lying around, but I'm tempted to try this out somehow as well ...
 
Back
Top