For people who might be first time on this forum, and that were brought here by eyestrain, headaches, migraine, nausea, disorientation and other unpleasant symptoms caused by working with computer displays, this is a guide how to build a device that can measure whether the backlight of those displays flickers, how and when. The problem is called Pulse Width Modulation (PWM) and you can read more about it here - http://www.tftcentral.co.uk/articles/pulse_width_modulation.htm There may be more reasons for the issues people who come to this forum have, but this is the one so far we can measure. It is important to note that you cannot trust any reviews that claim the device youre about to buy does not use PWM, as displays on phones, tablets, laptops and other tech are sourced from more manufacturers for each brand. All you can do is test for yourself and trust yourself. We are building an amateur oscilloscope with fraction of the money the real professional ones cost. If you have a friend who can lend you a real one, it is a better and more accurate choice. First you need to download Christian Zeitnitz soundcard oscope software here - https://www.zeitnitz.eu/scope_en We will be transferring light signal to sound, so you need a probe that can pick up light. A simple photodiode will do, for my device I used this one (Europe shop) - https://www.conrad.de/de/fotodiode-dil-1100-nm-60-osram-bpw-34-153005.html A better one recommended to me was this one, I will use it in my next probe - https://www.conrad.de/de/fotodiode-to-39-820-nm-55-osram-bpw-21-152977.html You have to solder the diode to a sound cable that you can connect to the soundcard or your computers microphone jack. A simple 3,5mm stereo jack will do. That one, when taken apart, has three pins - Left ear, Right ear and Ground. You have to connect the photodiode pins to either left or right, and then to Ground. As long as one pin is connected to the Ground, if the other is L or R does not seem to matter. This kind of soldering is fairly easy, just be careful not to use too powerful solder tool so you do not melt the photodiode with the heat alone. There is a certain voltage going through when the diode is picking up light, as it serves similarly as when you have a solar powered calculator. There is a report to be careful not to fry the sound-card, however I encountered no problems so far and the voltage seems to be very small to be able to cause any damage. Just in case though, its good to - Use external sound card. You can buy one cheaply second hand off ebay, or a new one. The reading should be better when a quality sound-card is applied, instead of basic notebook sound card. Also if you would happen to fry it, you just have to replace the card. My external card has a range of 44,1 Khz, which means it can probably read PWM up to 44Khz. Any higher and I would need more professional grade equipment. Newest macbooks are reported to have PWM at 112Khz for example. Now, connect the card to the computer, open up the oscilloscope software, select the input card in the settings and you are set. Example photo is my external display with lowered brightness to 80%. PWM free only at 100% brightness. A more portable option, although maybe not so accurate, is to use your smartphone and connect the diode to its headphone jack. You can then get an oscope application from either android store or apple store. More info here: http://www.retropcmania.com/2017/02/measurements-flicker-of-light-with.html Feel free to ask any questions.

Many thanks. With your help, I finally succeeded. I don't have any soldering equipment yet, but I have a male-male audio cable, a BPW34 photodiode, and adhesive tape. Works great so far. Edit: In this picture the photodiode is attached in the wrong direction. It works, but the amplitudes are reversed, as seen in the following posts. To get correct amplitude directions, attach the diode to the right (not the left like in the picture).

KM Could you post some results? I am interested to see. Amazing idea btw, I wouldnt think of taping it🙂

martin OnePlus 3 at brightness 65/255: It looks like terrible PWM, but (on custom ROMs) it's actually the first brightness step of the PWM-free range. What looks like high differences in amplitudes in reality is pretty flat. The sound card oscilloscope may not be able to show the full voltage range from zero to spike. OnePlus 3 at brightness 0: This time it IS terrible PWM that's visibly flickering, but doesn't look so bad on the oscilloscope. But the frequency gets picked up and you can see that something's going on which is not flat. I also tested my BenQ EW2740L (2 devices) and my iPad 4: perfect flat line on all brightness levels, everywhere on the display surface. This doesn't explain why I have to turn down the brightness on all those devices (especially on the iPad 4), and why I feel a difference in eye comfort between the two monitors. They certainly have a different backlight color tone. Their status LED has a flat line, too. Buy my Android device's status LEDs flicker when pulsating. Hard to capture. It seems their brightness is PWM-controlled, and only 100% brightness is flat. It might be a good idea to probe all lights in one's house, even the small ones, to be sure there's no PWM - once and for all. I measured the voltage of the BPW34 photodiode (hanging on the cable) with a cheap multimeter. I could never get the voltage past 0.5 V, even on 100% monitor brightness, holding the photodiode at the window, or using an LED flash light (which has a flat oscilloscope line, too).

KM Thank you for the post, that is pretty amazing. I see on the first image the freq is almost 60Khz, what soundcard are you using? I must say for myself, I tried measuring iphone 6s, that gives me one of the worst eyestrains and could not detect any PWM. What I detected was weird flickering lines when hitting the lowest brightness levels. I detected the same ones on my iphone 4s though, that is the most comfy device I can use for hours on end. I blame my non-professional gear and wanna test again with some proper gear, as I am absolutely convinced even the newest iphones use PWM - my symptoms are the same with them. EDIT: Try playing with the soundcard dials if it has some. I noticed sometimes the amplitude was very low, and then realized I have a very small gain on the card.

martin It's just 60 Hz. The soundcard is an integrated Intel HDA chip.

martin The BPW 34 seems to be pretty fast. 100 nanoseconds rise/fall time for the Vishnay Model, 200 ns for OSRAM's. I found this link where someone further explains those values: https://electronics.stackexchange.com/questions/118141/high-frequency-blinking-leds-and-sensor-for-that The OSRAM BPW 21 you mention seems to be much slower - 1.5 microseconds, which are 1500 nanoseconds: https://www.osram.com/os/ecat/TO39%20Ambient%20Light%20Sensor%20BPW%2021/com/en/class_pim_web_catalog_103489/global/prd_pim_device_2219533/ (There's a linked datasheet PDF with specs) OSRAM BPW 34: https://www.osram.com/os/ecat/DIL%20BPW%2034/com/en/class_pim_web_catalog_103489/global/prd_pim_device_2219534/ The next step might be those more expensive photodiodes. Somewhere I saw a price of $40. But maybe that's overkill. We better upgrade our oscilloscope first. The cheap Hantek 6022BE USB oscilloscope has sample rates of up to 48 MSa/s. Compared to 44100 Hz of our sound card, that'd be a big upgrade. www.amazon.com/Hantek-HT6022BE20Mhz-Digital-Oscilloscope-Bandwidth/dp/B009H4AYII It even has an open source project with Linux and Mac support: http://openhantek.org

KM Ask @TechSensitive about oscilloscopes. He might know.

KM That Hantek one looks good, I might buy it. However, do you know how to attach the photodiode to it? Seems like a completely different cable.

martin Not sure yet. It looks like it has two connectors. Maybe holding/taping them at our audio cable would work. But I found hints at German forums that an additional amplifying circuit may be needed: https://www.mikrocontroller.net/topic/417617 English translation https://www.prad.de/board/thread/50411 English translation (bottom of page 1)

KM Might be. @TechSensitive , could you shed some light on this? Just a photo of how your setup maybe looks like with a small explanation how the photodiode is attached, would help a lot.

Homemade oscilloscope to detect PWM DIY guide

martin

KM Could you please share the name of the rezistor?

KM

martin 100K Ohm

KM

I just assembled a second BPW34 probe, this time without battery, just using a 100K Ohm resistor. And even then it is responsive enough, as seen in the following screenshots. Blue = battery + resistor, red = just the resistor.
The red line is waving a little, which is due to some neighbored 50 Hz noise. And differences in amplitudes are related to different viewing angles.

100 Hz

100,000 Hz

10,000,000 Hz - Now the LED gets noticably darker, probably because it can't keep up with the PWM speed anymore.

So it seems we don't even need a battery, just connecting/wiring/soldering a resistor to a BPW34 is enough. And then attaching the probes left and right of the resistor, right next to it.

martin

KM Thank you, could you please again share a photo how the photodiode and the rezistor is attached together without the battery? Does the orientation of both matter? Does it also matter which clips of the osc go where?
I have downloaded picoscope, have hantek 6022BE and all the gear, I just need to put it together now.

KM

martin Orientation does not matter without the battery. If the signal is negative, below zero, just switch the probes. I will upload a photo in some minutes.

KM

martin PicoScope is a software only for PicoScope oscilloscopes. You need either:

Linux: OpenHantek (https://github.com/OpenHantek/openhantek)

Windows: The official software from Hantek for Windows (http://www.hantek.com/en/productdetail_2_31.html) or one of the alternative softwares for Windows. PCSCOPE was recommended by Wootever: http://www.eevblog.com/forum/testgear/hantek-6022be-20mhz-usb-dso/msg1099490/#msg1099490. There is also BasicScope, (http://pididu.com/wordpress/basicscope/) which looks a little strange: http://www.eevblog.com/forum/testgear/hantek-6022be-20mhz-usb-dso/?action=dlattach;attach=281643;image. There is also Open6022BE which I don't find the link for. It is buried somewhere in above eevblog thread. All those Windows versions might not run on Windows 10 as some users reported.

Android: HScope (http://hscope.martinloren.com/)

Hanzebenger

KM
I set it up this way. Worked nice on my CCFL monitor. Here is a picture:
But I was surprised that my laptop doesn't show the 1000Hz PWM I expected (Chi Mei N173HGE), it shows constant voltage there (no picture attached). It's possible it has no PWM because there are many versions, but there are certainly devices which have 1000Hz.
My question now is: Can I conclude from the steep downfall in the picture that my setup would detect 1000Hz PWM although the PWM of my CCFL monitor is only 180Hz?
I would says "yes" because the downfall is only 0,4ms long. What do you think?

Viktor

KM Have you tried connecting the OPT101 like shown here?

KM

Viktor Yes, I tried to do that. But the battery had no impact at all. Maybe my soldering was too bad. So I took my second OPT101 and used it without the attached board, effectively overheating it within seconds.
However, I think the OPT101 may be too slow due to the official bandwidth specification of 14 kHz.

Edit: Viktor I realized I soldered the OPT101 to its board the opposite way. Would this make any difference? I had to switch the probe tips to get a positive voltage.

martin

Viktor On your first photo of the photodiode orientation, and on the second, the diode is reversed. Which one is the right orientation please? The notch is first pointed towards the place where oscilloscope gets connected, then it is pointed to the other side. I am a bit confused. Also how do you connect the 1m/OUT and -V/COM together, soldering it together?

This image is my first attempt. The wires are from battery. Is this the correct way to connect it?

Here is the reversed orientation of the diode and I soldered some of the pins together as per the diagram. Is this the correct way?

This is how it works with my soundcard oscilloscope.

Could you please share photos how your battery and oscilloscope are connected to the board so it works? Thank you!

Viktor

KM There is a small notch on the photodiode to help with the orientation, this should be the correct layout:

Edit:
Test with a Nokia 6 at <17% brightness:

result from notebookcheck.com:

Edit2:
I tried to create a pwm signal with the arduino status led:

Code:
void setup()
{
pinMode(13, OUTPUT);
}

void loop()
{
digitalWrite(13, HIGH);
delayMicroseconds(100);
digitalWrite(13, LOW);
delayMicroseconds(100);
}

The opt101 can also be connected without soldering:

KM

My recent LED measurements (see above) which showed no perfect square waves bothered me a little and I began to wonder whether the LED or the photodiode was too slow for the PWM signal. So I wanted to know how much impact the 9V battery really has, and as you can see in this screenshot, it speeds up the rise time a little.

Blue=9V battery, Red=no battery; Frequency: 5000 Hz, 100K resistors

However, reducing the resistance from 100K down to 10K speeds up the diode almost proportionally by the factor 10:

Blue=10K resistor, Red=100K resistor; No batteries

The downside is that 10K is not only 10 times faster but also delivers a 10 times smaller voltage, so I believe for now 100K are a good trade-off. If anyone has ideas how to make this faster or can measure his OPT101 rise times, let me know. With the Raspberry Pi How-To you can create your own PWM-controlled LED circuit and measure the speed of your oscilloscope setup: https://ledstrain.org/d/375-howto-create-pwm-led-flicker-with-a-raspberry-pi

KM

OPT101 without battery, 5000 Hz PWM:

So it seems the OPT101 gives up much earlier.

martin

Viktor Can you connect the OPT101 only with the rezistor like in @KM s post? I tried but it doesn work, it doesnt show anything. When I use it with my soundcard, it works, but then the waveform is bad, thats why I wanted to fix it with the Hantek. Or could I just use the BPW34 instead with the rezistor? Are they that different?
I have the opt101 with the board and everything, but Im afraid that when soldering it to the board, ill just connect all the legt together with the tin and it wont work.

KM

It looks like both the working OPT101 + 9V battery solution and the BPW34 + 9V battery + 100K resistor share the same responsiveness, with the OPT101 delivering a higher voltage.
Both suffer from their slow rise and fall times.

I am thinking about a new approach the whole day now: connecting multiple BPW34 in parallel to maximize the photocurrent, and using a low impedance resistor - 10K or lower - to maximize the voltage. High responsiveness and still decent voltage output.

Viktor

Using the photodiode circuit wizard at: http://www.analog.com/designtools/en/photodiode/ it shows that a BPW34 + AD8066 should be able to output a bandwith with 500kHz:

The AD8066 can be found here: https://www.ebay.de/itm/261101423440

KM

OK somebody else must take over from here. I don't have the electronic knowledge to connect such devices and build circuits and I realized learning it in an autodidactive way takes a lot of time and endurance which I don't have.

KM

Hanzebenger I think you should even see 10 MHz PWM - as long as it's full on/off PWM. As you can see in my pictures in which I compared battery vs non-battery waves, a few posts above: https://ledstrain.org/d/312-homemade-oscilloscope-to-detect-pwm-diy-guide/52
The wave will not look like rectangular PWM anymore, and it will float above zero, but you can see spikes that have the correct frequency.

When the flicker is just a small ripple riding on a DC signal though, as seen in "flicker-free" displays, it might not be displayed. But such small flicker may matter: FRC/temporal dithering, unresolved software eye strain - all those have just a very tiny flicker. So I imagine if a backlight has a similar small flicker, it might trigger eye strain, too.
You can see such small waves a little easier when you switch your PicoScope oscilloscope to "AC" mode. The Hantek 6022BE doesn't have such a mode.
In fact I think there is a higher frequency riding on top of your waves (where there is no thin line but distortion) which you could make visible by adjusting the time axis.

I believe the largest part of your rise times is related to the slow CCFL response. You should easily detect 1000 Hz PWM.

KM

I have read that you can also connect a photodiode to speakers, which is exactly what my expensive LiFli flicker detection device does. I just plugged my headphone into its oscilloscope output which happens to be a regular headphone jack and I could hear the humming 100 Hz flicker of my bulbs and lower frequency LEDs. It should be possible to connect a BPW34 to a standard headphone jack, both cheap parts, and then plug in headphones and go hunt for flicker-free devices. At least flicker-free in a sense that the flicker, if any, is in the audible range of below 20 kHz. This should be an ultra-cheap but precise setup with your ears as oscilloscope.

KM

Here you can see notebookcheck's PWM setup in action as they test the iPhone Xs:

https://youtu.be/sE6OjuRAHpE

Oscilloscope: Rigol DS1052E, ~300 €
Photosensor: Thorlabs PDA100A-EC, 327,29 € (PDA100A2) excluding VAT, https://www.thorlabs.com/thorproduct.cfm?partnumber=PDA100A-EC
The latter mentioned here on their German website: https://www.notebookcheck.com/Wie-wir-testen.1948.0.html

« Previous Page Next Page »