Skip to main content
Topic: Monitor backlight flicker measurement (Read 1421 times) previous topic - next topic

Monitor backlight flicker measurement

Hi,

many modern monitors offer a feature called "dynamic contrast" which is achieved by reducing brightness of the backlight if the displayed image is dark. Of course the LCD opacity has to be adjusted as well to achieve consistent results. In theory, that is a great way to handle almost-black image regions.
In practice, hardware vendors screw up the implementation. I've seen TV sets advertising "400 Hz motion rate", and doing PWM of the backlight at less than 50 Hz. Some laptop screens are guilty of that as well. With old CCFL backlights and high PWM frequencies, the afterglow of the phosphor and other effects lasted long enough to smooth out any possibly visible effects of the PWM. With modern LED backlights, there is no afterglow at all, and in a dark environment you can see a distinct stroboscopic effect by waving your hand in front of an affected display.
Even worse are some manufacturers which try to smooth out power usage by having a phase shift for each color of their backlight RGB LEDs, leading to a colored stroboscopic effect when PWM is active.

http://http://www.tftcentral.co.uk/articles/pulse_width_modulation.htm has a nice description of the issue and some suggestions on how to measure flicker. The only catch is that if you want to buy a monitor/TV you have to take all those measurement photos in the store. Most electronics stores won't allow that, so you can either buy a monitor and test it at home or bring a separate testing device (which is apparently totally acceptable in the stores I asked).

Right now I'm using a photo transistor based ad-hoc wiring attached to the microphone input of my laptop, and the results are pretty abysmal because I'm picking up 50 Hz line noise everywhere.

Having a small circuit with a photo transistor or photo diode and directly attached ADC should fix the line noise issue and give me some really usable output.

Suggestions on how to implement this reliably are very welcome. (A modded USB sound card attached directly to the photo receiver circuit is the next step I'm planning for...)


(Just in case you're wondering why I'm doing this: I get a headache from low-frequency flicker, and I'd rather buy monitors which don't exhibit that problem.)
If you have any flashrom problems, please mail flashrom@flashrom.org and include "Bus Pirate" in the subject line. As an alternative, you can send me a private message.
It is recommended to use at least flashrom 0.9.2 or later, and latest flashrom from svn if you want a 3x speedup.

Re: Monitor backlight flicker measurement

Reply #1
electronics/computer stores are known for the very noisy environments they display products. the have no concern in reducing any of the noise, only in selling you expensive merch.

as for the noise unless you plan to bring your own Faraday cage i can't think of anything to help improve the issue. however if you can put test images and video on the monitors and compare them side by side, I would trust your eyes, and stick with manufacturers that have a good reputation.

 

Re: Monitor backlight flicker measurement

Reply #2
Wrap the circuit in aluminum foil and ground the foil. You might also want to try adding shunt resistance.

Pixel Qi actually uses the switching effect to its advantage. The idea is to use a LCD without color filters and a backlight that cycles through all 3 primary colors very quickly, exactly like DLP. (I presume they cycle through all colors at least once in a frame to prevent color breakup.) Since the color filters are only about 33% efficient, doing away with them theoretically makes the display 3 times as efficient. It can also be designed to work like Eink by turning off the backlight.