# Flicker fusion frequency

Discussion in 'The Projects Forum' started by squashjosh, Jul 3, 2015.

1. ### squashjosh Thread Starter New Member

Jul 2, 2015
5
0
Hi guys,
I'm a newbie to circuit building but was hoping for some guidance to build a circuit that I could use to determine at what frequency people start to see flicker vs no flicker (the critical flicker fusion frequency).
I need a LED (preferably red) that has a 50% duty cycle that I can adjust the frequency between 1-100Hz with a button or dial. I also need to know at what frequency the light is flickering (so some form of digital output). I'm guessing from what I've read that I need a 555 timer but beyond that I'm lost.
I couldn't find any old projects doing the same and I apologize if this is a stupid sounding project.
Any suggestions would be greatly appreciated
Cheers.

2. ### ericgibbs Senior Member

Jan 29, 2010
2,418
369
hi,
A 555 timer can be set for a 50% duty cycle.
The variable frequency could be achieved by using a potentiometer, which could have a frequency calibrated dial.
It may be necessary to have selectable ranges over the 1Hz to 100Hz band.

The 555 could drive a transistor/led circuit.

A word of caution, at around a flash rate of 16Hz some people can be adversely affected by the flashes.

3. ### DickCappels Moderator

Aug 21, 2008
2,588
610
There is a lot more to flicker than meets the eye. You might find Joyce Farrell's An analytical method for predicting perceived flicker interesting and maybe useful in preparing your experiments.
http://scien.stanford.edu/jfsite/Papers/ImageRendering/Displays/Farrell_Flicker_1986.pdf

A problem with the NE555 is that it is not a simple matter to vary the frequency and maintain a fixed duty cycle. You could set up the NE555 as a variable frequency oscillator as shown below then follow it with a D, JK, or T flip-flop to divide it by two to get a nice square wave (shown below the NE555 circuit).

Another approach might be to use my fixed duty cycle LED strobe circuit. It is based on an LM393 dual comparator

Edit: Corrected URL per ScottWang's correction.

Last edited: Jul 4, 2015
4. ### ericgibbs Senior Member

Jan 29, 2010
2,418
369
hi Dick,
There is a 555 configuration that will give a 50% duty over a range of frequencies, ref image.
E

File size:
10.1 KB
Views:
17
5. ### squashjosh Thread Starter New Member

Jul 2, 2015
5
0

Thanks Eric. Hopefully I shouldn't have anyone with a photosensitive epilepsy. I'll make sure to exclude people with a history of seizures but if someone did have a seizure, they couldn't be in the better place to have their first. Cheers

6. ### ericgibbs Senior Member

Jan 29, 2010
2,418
369
hi josh,
I would say a starting value for the timing Cap and Variable resistor would be around 3.3uF and 100K.
The CMOS version of a 555 eg: 7555 would be my choice.

What intensity of light flash are you planning.?

DickCappels likes this.
7. ### DickCappels Moderator

Aug 21, 2008
2,588
610
Indeed, don't try that with the bipolar version.

Jul 2, 2015
5
0

9. ### squashjosh Thread Starter New Member

Jul 2, 2015
5
0
Thinking 400-1000mcd but may need to adjust in the future.
Many thanks

10. ### ericgibbs Senior Member

Jan 29, 2010
2,418
369
hi josh,
It would be an interesting exercise to have the option of flashing a Blue, Red or Green LED , all set for the same intensity,

11. ### squashjosh Thread Starter New Member

Jul 2, 2015
5
0
Completely agree and I'm sure there would be a different frequency for different colours and different intensities but I can't test my patients all day so have to go for most bang for my buck. Red is the winner

12. ### DickCappels Moderator

Aug 21, 2008
2,588
610
If you have a way to measure it, the correct quantity to measure would be luminance (Ft. Lmaberts or Nit).

13. ### Kermit2 AAC Fanatic!

Feb 5, 2010
3,671
892
you mayget some strange results via beat freq. if the room has any AC powered lighting. I assume you will be in a darkened room, but if you want to have any light use sunlight or DC powered lights

14. ### DickCappels Moderator

Aug 21, 2008
2,588
610
I heard a well-known expert in human factors tell his audience that flicker from high refresh rate computer monitors (you can tell the time frame in which this took place) can beat with florescent lighting to create visible flicker. The idea was floating around the industry at the time.

Unbelieving, I asked where the nonlinear mixing would occur and he replies "In the eye.". Some days later I took two LEDs, one running at about 120 Hz and another connected to a function generator and I slowly and carefully swept the function generator around 120 Hz and could not detect any flicker in any part of my retina. This was repeated at various luminance levels in a dimly lit room always with no flicker being observed. I would be very excited to learn the details of an experiment that actually produced the effect.

15. ### wayneh Expert

Sep 9, 2010
11,925
2,857
I'd use an op-amp based square wave generator for this one. I think it's simpler to get the ~50% duty cycle over a wide frequency range, at the low end frequencies of this project.

One thing about the flicker is that it is more easily detected in peripheral vision than in the fovea. A moving light in our periphery will appear to flash - even from the movement of the eye itself - whereas a light you are staring at will appear steady at the same frequency.

The few times in my life that I was approaching a faint, I was very aware of the flicker of the fluorescent lights around me. I never noticed them normally.

16. ### AnalogKid Distinguished Member

Aug 1, 2013
4,301
1,178
Why? With the feedback timing resistor circuit, the bipolar 555 duty cycle will not be a perfect 50%, and it will change with load. Is that what you're referring to, or is there some other reason?

ak

17. ### DickCappels Moderator

Aug 21, 2008
2,588
610
Those are the reasons of which
I was thinking.

Aug 21, 2008
2,588
610
19. ### ScottWang Moderator

Aug 23, 2012
4,783
762
Have you try this: