Electronic Circuits and Projects Forum

Interesting experiment you are working on .. Will be interested to see the results that others get as well

Rolling shutter technology is the curse of my storm filming
All the newer cameras... Last ~ 5 years have CMOS rather than CCD sensors and they all use the rolling shutter
The hi speed of a lightning flash is destroyed by the rolling shutter.
I still have a Fuji S9500 digital camera that uses a CCD image sensor with the standard full chip shutter

Cheers
Dave

No responses yet from anyone that might have an Iphone and the ability to generate the pulses. I'll whip up a circuit and go visit my brother in law within the next few weeks, it's just exploratory right now to see how the rolling shutter actually functions in a detailed way.

I'm still using a Cannon A530 which is a CCD based but it's just a 'point and shoot' with an a manual mode, nothing like the camera you have, as much as photography interest me my other hobbys are far in front of it as far as what I have to spend.

Would you recommend CCD or CMOS? I seem to find CCD faster but lower resolution. I like my camera to be able to shoot a photo of some that well that you can analyse the environment around them so you could zoom in an read a distant title of a book for example. CCD just does not seem to offer that. What are everyone else's view.

For video cameras "Rolling shutter" is a bind, but just as in the early days of Hollywood "we" have to get round the limitation of the tools we use.
My own movie-camera is CMOS and I try to keep large rotating parts out of the picture (eg flywheels, props, etc), but I sometimes wonder how broadcast cameras overcome this.... or do they splice-in some "real-film" footage? I rather doubt that, but then they have much faster data-paths, with correspondingly larger memory demands.

Consumers want their whole holiday on one memory card. (least that seems to be the view taken by consumer camcorder makers).... so this means some economy in pixel rates.
One benefit of CMOS sensors is they work better in low-light - and are cheaper to make ... strong benefits, I guess.

Will be interested to know what use OP will use this experimental Info for - it's npot as though Apple will change their design, - anyone?

One benefit of CMOS sensors is they work better in low-light - and are cheaper to make ... strong benefits, I guess.

This is not true. Charged couple devices are superior in low level light, lower noise. CMOS sensors have had their effective ISO levels sky rocket in recent years, but this comes at a cost of increased noise, CCD's produce a superior quality image.

This has nothing to do with improving the cameras, it's exploiting them for a specific effect. I believe there's a way to do scan line image creation on something like the Iphone's image sensor. I may be able to test it within the next few weeks, mind you I said that a few weeks ago =>

I have the board that will test it on my desk, but I've been short on time seeing as my son was born 6 1/2 weeks early, I was going to finish this a month ago, but life happens! I still don't know if the effect I want to exploit exists, that's the reason for the testing. As far as I know typical CMOS sensors are rolling shutter, but per line. My theory is that the Iphone's camera is a per pixel rolling shutter, the effect will only occur on a per pixel rolling shutter.

looks like you might still be looking for some help. we have an original iPhone, 3gs and 4s in the house at your disposal.

Do you have a micro controller dev board, such as a PIC or AVR and know how to program it? All this will require otherwise is a red LED.

will an arduino suffice?

Sure as long as you can program a software or hardware PWM at a decent frequency.

You need to output a 1mhz square wave with a 25% duty cycle for a starter test, the PWM drives the LED (directly without a current limit resistor) If you can adjust the voltage of your Arduino down a bit that would be ideal, as at 5V you'll probably be over driving the LED a bit not to mention stressing the I/O pin, I don't want to hurt your board.

All you have to do is take a single photograph of the LED while the code is running and the LED is lit, but the LED has to be RIGHT up against the window of the camera so that any light coming from it will be out of focus and saturate the entire sensor.