DIY Flicker Meter and Pulse Measurements! - a how-to guide
So why do you want a flicker or pulse meter for yourself?
- Measure pulse rate from Red Light Therapy or PBM products.
- Test for flicker on your light-bulbs, computer monitors, and LED products.
- Test for PWM (pulsed width modulation) in dimming products.
- Test for flicker or PWM when using a slow-motion camera doesn't find anything.
Well instead of sending every single light I find to my 3rd party lab to test for flicker, and I didn't want to spend $1000+ on a professional handheld meter. I decided to figure out how to make one myself.
This build uses the cheapest oscilloscope that I can find. This is a very common and popular tool used by electricians and electrical engineers to find the oscillations in voltages in a device. Since we know most equipment must run on 60Hz grid, and often devices have their own internal oscillations or clock, then this device is critical for testing, evaluation, and designing electronics.
All we have to do is pair this oscilloscope with a photodiode. A photodiode will receive in light and convert it to a small amount of voltage. Essentially a photodiode is the opposite of an LED, which takes voltage and converts it to light. If the light is pulsing or flickering, then this voltage will dramatically go up and down when connected to the photodiode. Then the oscilloscope will measure those variations in voltage.
That's the entire concept! I have seen people use anything from a cheap garden solar power collector or a fancy photodiode to make these DIY flicker meters. The key is to make sure the photodiode has a fast response to accurately measure the ups and downs, and that the oscilloscope is also fast and accurate to get good measurements.
Luckily, I didn't have to design this myself, just dig deep on the internet to find a forum that was talking about choosing the best photodiode and assembly. Check out these links to see some great information about the circuitry, testing, and different types of diodes.
Oscilloscope: Any oscilloscope will work for this. The higher the quality, the better resolution you will get from the measurements. For this demonstration I found the cheapest possible oscilloscope to prove that this theory works.
This is the one I used (but I ordered directly from China, not from Amazon). I even had to assemble mine and find a power adapter for it! So make sure you find one that says "fully assembled" and with an adapter.
Photodiode: It seems the best diode was BPW34, which is very common and easy to find.
Resistor: A 100k Ohm resistor is also needed for this build. This is a very simple and common electronic part.
Optional - 9 volt battery for bias: This helps "boost" the signal from the measurement. For testing high-powered red light therapy, it hardly seems necessary. For testing flicker on lower intensity items like standard lightbulbs or computer monitors, then it might help. But overall it seems the measurements with or without the battery are the same.
Connector for the battery:
Setup - no battery:
The easiest setup is without the battery. It doesn't matter the direction that the two pieces are held. Ideally you have a way to connect together the photodiode and resistor. Otherwise you can try to just twist them together. Then clamp on the oscilloscope on both sides of the setup.
It is simpler than it looks. Really just both components are connected end-to-end, then the oscilloscope clamps are put on either side. Orientation isn't even important here. Check the pictures in the links above for better schematics and pictures of this setup.
Setup - with battery:
Still relatively simple, this time you need to look closely on the BR34 and find the small dot on one end. This helps you orient it correctly. The end with the dot connects to the resistor, and is where the red (positive) lead from the Oscilloscope connects. The opposite end of the resistor is where the black (ground) lead from the oscilloscope connects as well as the black end of the 9V battery connects. The opposite end of the photodiode is where red end of the 9V battery connects.
This time I started soldering the components together on this little circuit-board, to make it easier to handle the flicker meter later on.
Simply turn on the Oscilloscope and make sure the positive and ground wires are connected properly to the photodiode setup we describe above. Make sure the oscilloscope is in DC mode, not AC mode.
When aiming the photodiode at a light source, you might start to see the numbers and lines on the graph start to change. I usually scroll through the V/DIV and SEC/DIV until I find where the numbers "settle down" and make logical sense. Sometimes moving closer or further away from the light can also help find the ideal testing distance. For example a very bright light such as a Red Light Therapy panel might simply max-out the measurement, so you would have to move further away. A weak light in the ceiling, you might want to move closer to.
It is best to do testing in perfect darkness, as sunlight or other lights in the area might interfere with the measurement.
It is always important to try to verify that your measurement system is reading properly and accurately. Often just because a measurement tool is giving you some numbers, doesn't mean the output is accurate or represents reality. Just look at the mess the entire industry is in with using cheap solar power meters for intensity.
Tested on a 3Hz pulse controller: I was able to verify that the controller was putting out 3Hz pulse on my DIY flicker meter.
In the upper left corner you can see the "Freq: 2.944 Hz" which is telling us the pulse rate. We would expect a small variation and not always measure exactly 3Hz, either the controller isn't perfect, or this flicker-meter isn't perfect.
Tested on a 8Hz flashlight: Check our earlier blog to see how we tested some tactical flashlights for their strobe function. This Hz frequency seems common for tactical flashlights like the ones that I tested.
We can definitely "see" the PWM frequencies from my flashlight on the dimmed setting, and on several other dimmers that I have around the house. This can easily tell me the frequency of the PWM, as the percentage is usually 100%.
It is harder to determine the flicker percentage from this setup. For example it might just be the cheap oscilloscope that I am using, or how I am using it. We can see the clear flicker frequency and the ups and downs.
Ideally the flicker percentage would be simply calculated as:
Flicker% = 100 * (Vmax - Vmin)/(Vmax+Vmin)
However, depending on the settings on the oscilloscope and how close the photodiode is to the light source, I get varying results.
Here we see the image measuring the flicker from an old fluorescent bulb. In this case the flicker percentage is easy, it is 100%. In the upper right corner we see the Vmax of 0.55V and Vmin of -0.02V. So essentially the voltage goes to zero during the cycle. So we quickly know that this will result in 100% flicker if we tried plugging in these numbers into the equation.
We are also verifying the Hz measurement capabilities here. Since we expect most lights to flicker at 120Hz, and we can see that clearly in the upper left corner for both of these flicker examples.
This is a measurement from a red light therapy panel prototype. Now we can do some math with the flicker.
Flicker% = 100 * (5.84 - 1.54) / (5.84 + 1.54) = 58% flicker.
So we can see this prototype red light panel had 58% flicker, which is much higher than the IEEE recommendation of being below 9% flicker at 120Hz. So I cancelled this light and never brought it to production, since I wouldn't want the liability to cause flicker stress in my customers.
GembaRed Lights and low-flicker lights:
I can't get much reading on GembaRed Lights or low-flicker lights. Since my lights are usually only around 1%, then this setup isn't sensitive enough to discern that low of a flicker. So it is still best to send my lights out to the professionals to get the most accurate numbers.
It is clear this DIY flicker meter can help us find the Hz of strobe, pulsing, PWM, or strong flickering lights. This is great, because we can quickly verify the pulsing if we want to use a light strobe for light-therapy. Or I can quickly walk around the house testing lightbulbs and TV screens for flicker.
However, the drawback is so far I cannot measure any low-flicker lights, since they just show up on the screen as narrow squiggles. So in those cases if we really need to know the number, a more accurate method would be needed.