Power Conditioners fix digital camera problems
Latest news from AEL, the Advance AGT is recommended by phaseone (www.phaseone.dk) to solve lighting problems
The advent of very high quality scanning backs for studio cameras has caused a new interest in mains related image noise problems. Most photographers are not engineers and this page attempts to provide enough information for a studio camera user to try and evaluate whether or not a particular problem may be solved with mains power protection devices.
Diagnosis before cure
In most technical applications where problems occur it is usually easier to effect a cure if the problem is actually understood. By examining images which have been produced by a suspect digital camera under differerent conditions it may be possible to eliminate possible causes of a problem. It is also necessary to have to hand the expected performance for the camera. If the camera was never intended to produce an acceptable image under the conditions of test - then no amount of mains protection is going to improve it! (Some camera manufacturers have been reluctant to give measurable figures for performance and we feel this is unacceptable.)
Popular image problems
We usually ask photographers to examine poor images for the following problems:
variation in lighting conditions during the scan
An example of this is in a studio where ambient daylight is illuminating the subject and the sun goes in or out during the scan. In the simplest example where the sun starts behind a cloud and comes out during the scan there will be an overall brightness increase across the image in the direction of the scan. The change will be directly related to the rate of change of sunlight variation compared to the rate of scan. Clearly elimination is straightforward.
variation in artificial lighting due to mains problems
- these short duration occurrences can produce bright spots in one line of pixels caused by the light actually brightening up momentarily - unusual even in direct off mains tungsten lighting because the spike energy has to be quite large. This effect is normally completely random.
- this is where the mains voltage sags and the light given out by unstabilised tungsten lights literally dims down for the same period as the voltage droops. The result is similar to the sun going behind a cloud during the scan. (This process can work the other way with voltage surges) and may vary considerably during a scan.
variation in artificial lighting due to mains frequency beats
The UK mains supply is modulated at 50 Hz (cycles per second) and direct-off-mains tungsten lights (ordinary bulbs) flicker at this frequency ALL THE TIME! The human eye 'integrates the effect and does not notice it.
This causes a problem which is always difficult for non-physicists to understand but musical people are familiar with tones which 'beat' and the same effect can happen when the effective scan rate is close to 50Hz or a multiple. The result is that some lines of pixels see bright light and some dark.
The picture will have a series of `fringes' parallel to the scanning sensor - width and spacing of the fringes will be set-up and scan-rate dependent. Cameras have anti-fringing software to combat this problem. Usually the result looks like a venetian blind shadow over the image. It can also produce multicoloured effects not just light and dark beacuse only one colour channel may be beating.
light path obstruction
If a piece of dirt is stuck to the scanning sensor it will produce a (usually coloured) line in the direction of the scan. Usually the line is across the entire frame and the problem can only be corrected by using cleaner techniques. If the sensor is clean and the line is always the same colour it suggests that the sensor is faulty.
this is a huge subject and only briefly touched here the image faulty areas should be pushed to the limits of installed software using mainly contrast and brightness adjustments to view the image pixel by pixel to try and assess what is actually happening as the sensor moves along.
If there is quite a lot of colour variation along the line of pixels when scanning an evenly illuminated area it is probably directly related to any interpolation in the process or sensor/amplifier inconsistencies. (Users should appreciate that the analogue scanning data is converted to digital information right after the scanner head. The A to D process is where the electronics value really costs!)
Colour variation in the image under these conditions must be reviewed with the supplier specification to assess if it is typical or worse than usual for the camera back. Tests should be done at different scan rates and exposure combinations (f-stop & light change) to try and discern the relavance of the fault to external influences.
'Fringes' which occur along a distinct line of pixels across the scanning sensor are consistent with a change in camera system amplifier gain between successive pixel rows and really should not happen.
If a step function in the overall `gain' characteristic occurs on a random basis check the fringe to see if it gives the impression that the back slowly drifts in it's gain value and then suddenly regains control going back to a figure from several pixel lines previously in the scan. If so it can be either mechanical sticking of the traverse lack of amplifier gain control or a voltage fluctuation caused by changing loads elsewhere.
Sometimes it helps to build a test setup to eliminate external influences and still demonstrate that the back produces problems. A simple battery driven lamp which has no 50Hz and no spikes! It can be arranged to have no droop through the actual scan process by using a relatively large battery. The best approach would be a car battery and a couple of headlamp bulbs. To be utterly pedantic the bulb voltage could be monitored during the scan.
The subject should be a nice plain background with a small object used temporarily to set up the correct focus to simulate an actual shooting scenario. This test would eliminate a large proportion of the external influences which could otherwise be erroneously blamed.
Finally if after evaluation it is felt that the camera is really suffering from a direct or indirect mains related problem please ask us for help.
We can provide a UPS to provide power when the mains is down a power conditioner to remove spikes or a more sophisticated one which stabilises the voltage suppliying either the camera lights or both. Normally we would loan a unit to allow the photographer a chance to prove what may be an expensive cure actually solves the problem. We have at least one obsessive photographer on site!
If you need any more info please ask!
Latest news from AEL, the AdvanceAGT is recommended by phaseone ( www.phaseone.com) to solve lighting problems