Hi Gregory,
That's a nice example of CCD blooming from the metallic surfaces in your first shot.
Blooming is one area of MFD where there should be no mystique, as the datasheets from both manufacturers (Dalsa and Kodak) address anti-blooming performance.
Your sensor is a Dalsa (Shelby: all Leafs from the Valeo on are Dalsa; you have to go back about a dozen years to the Leaf C-MOST, I think, to find a non-Dalsa-sensored Leaf back [the C-MOST used a Fill Factory sensor]; I'm sure Yair can confirm this).
All Kodak MF CCD data sheets state something like "A lateral overflow drain (LOD) suppresses image blooming..." and specify "Blooming Protection" of typically 1000x to 1500X saturation exposure...at 4ms exposure time (1/250 sec). Kodak datasheets include a graph showing how longer exposure times increase this figure and shorter times decrease it. This is because it takes time for the excess electrons to pass into the LOD and be removed: the LOD "sink" electronics have current limits; and current of course is dQ/dt, electrons per second. So if a given flux multiple above saturation isn't reached until a longer time has passed, more of the excess electrons have been drained before the exposure is terminated and the image is read out. Rainfall is a nice analogy: if 3 inches of rain falls on a city over 24 hours, the drains can conduct it away. If the same 3 inches falls in 3 hours, the drains can be overwhelmed and the city floods [this happened in Dublin a month ago:
http://en.wikipedia.org/wiki/2011_European_floods#Ireland].
Interestingly, Kodak MF CCD anti-blooming performance is NOT getting any better with newer sensors. The KAF-40000, in my view the overall best MF sensor to date (Pentax 645D and Hasselblad H4D-40), is specced at 1400x saturation. Whereas, the old KAF-16803 (Kodak Pro Back, Hasselblad CFV-16, PhaseOne H20/P20 etc.) is actually specced better, at 1600x saturation. The KAF-50000 (H4D-50) is only 800x saturation!
As for Dalsa? Their "VNS" serves the role of Kodak's "LOD": "VNS drains superfluous electrons as a result of overexposure. In other words, it only sinks current. During high overexposure, a total current of 5 to 10mA through all VNS connections together may be expected." If I'm reading their datasheets correctly, i.e. if they are using the same basis for their specifications of "Overexposure handling" as Kodak uses, then they are performing significantly worse than Kodak. All their MF sensors (such as the FTF4052C and FTF5066C, used in all the 22 and 33 MP backs, including your Valeo) are specced at only 200x the saturation level; that's
2 to 3 stops less tolerant of overexposure than the Kodaks. Sorry to be the bearer of bad news! OTOH, I calculate that the FTF4052C has an unusually deep full well capacity/saturation level (170,000 e- vs. 100,000 e- for the equivalent Kodak KAF-22000), so that is more like 340x for the Dalsa and 1000x for the Kodak on a normalized comparison basis; a 1.5 stop difference. But taking FWCs into account like this doesn't significantly alter the 2-3 stop disadvantage of the other Dalsa sensors.
One other thing. The anti-blooming performance of MF CCDs appears to be way poorer than that of CMOS DSLRs. If I shoot directly into the sun with my Kodak digital back (exposing for the foreground), the blooming either side of the solar disk covers up to one third the height of the image! I don't have an example to hand, but I'm sure I've done the same with a Canon 5DII and not seen blooming like this.
Ray