Photodiode Detecting Mains-Powered Filament Light Bulb Output as 100Hz ?!

[MikeNye]

Hi everyone,

I've been playing around with some photodiode devices (Texas Instruments OPT101) lately for a hobby project that I'm working on, and I've come across a weird phenomenon that had me baffled for a couple of days... Not quite as weird as Dave's weird oscilloscope phenomenon, but still pretty confusing for a beginner. I think I have figured out why this is occurring, but would be keen for someone more experienced to verify my findings... Read on for more info!

I have a very simple photodiode circuit (figure 5 from the OPT101 data sheet if anyone is interested), with my oscilloscope measuring the voltage created as a result of light level exposure to the photodiode.

When I illuminate the photodiode with a stable light source, such as a battery powered LED, then the voltage generated by the photodiode device is flat DC - this is what I expected.

The weirdness occurs when I illuminate the photodiode with a mains-powered halogen desk lamp. Initially, the scope trace showed a sinusoid with a DC offset, which I also expected. The DC offset for the brightness of the halogen lamp, with the sinusoid being caused by the very small changes in brightness (faster than the eye can recognise) as a result of the 50Hz mains alternating current. So, I expected the sinusoid to have a frequency of 50Hz - identical to mains AC frequency.

However, when I actually measured the frequency of the sinusoid, it was 100Hz - double mains frequency! Check it out:


This had me really confused, because I would've expected the sinusoid to have a frequency of 50Hz, however, upon further consideration, I think the problem can be answered with the following reasoning:

Mains AC voltage is at 50 Hz, with a period of 20ms. Therefore, the filament will have current flowing through it in one direction for 10ms, and the other direction for 10ms. Because the filament will glow regardless of the direction of current, we get peak light output twice per AC cycle (once during positive current, once during negative current), giving us the resultant sinusoid with a period of 10ms, or a frequency of 100Hz.

I hope my explanation makes sense, and I would be very interested if someone more knowledgeable than I could verify my thought process...

Thanks for reading!

-Mike

[TheWelly888]

Have you tried looking at other light sources such as fluorescent lamps? And ordinary light bulbs ( the now banned sort )? Also driving an LED with different waveforms?

[rossmoffett]

You're correct.

[DavidDLC]

The mains have a frequency of 50 Hz, for each complete sine wave the light will vary the intensity twice on the positive and the negative part of the wave, that is what you are detecting.

[kc1980]

Yup, you're right.  We've done a few wire arc simulations using 400Hz aircraft power.  During the arc event, we would always hear a constant pitch which, after very little thought, we originally believed to be 400Hz.  However, when I generated a pure tone via my mp3 player just for kicks, I found that the pitch of the arc was in fact 800Hz.  Duh!

[xani]

Why i have a feeling that in month or 2 ill see a post on hackaday.com about "Bulb-synchronized alarm clock" ? 

[MikeNye]

Wow, so many responses in less than 24 hours! Thanks to all who replied. Glad to know my reasoning is correct.

Posted by: TheWelly888
Have you tried looking at other light sources such as fluorescent lamps? And ordinary light bulbs ( the now banned sort )? Also driving an LED with different waveforms?

Not yet, however this is something I plan on doing over the next few weeks. I also have some TLV5618A DAC chips, which I have connected to an Arduino. I've written some code to display sine waves, square waves, saw waves, ramps, steps, etc - a very basic arbitrary waveform generator. Here are a few traces:

Posted by: kc1980
Yup, you're right.  We've done a few wire arc simulations using 400Hz aircraft power.  During the arc event, we would always hear a constant pitch which, after very little thought, we originally believed to be 400Hz.  However, when I generated a pure tone via my mp3 player just for kicks, I found that the pitch of the arc was in fact 800Hz.  Duh!

Interesting stuff. I had no idea aeroplane power had such a higher frequency than normal mains power. Why is this? Alternators driven from faster rotating turbines?

Thanks to everyone.

Cheers.

-Mike

[kc1980]

Interesting stuff. I had no idea aeroplane power had such a higher frequency than normal mains power. Why is this? Alternators driven from faster rotating turbines?

The higher frequency allows us to use smaller transformers, generators and motors, which saves quite a bit of weight.  This is fairly common on ships as well.  Yes, the generators are driven from the engines but airplanes typically convert to a fixed 400Hz through mechanical or electrical means.

[MikeNye]

The higher frequency allows us to use smaller transformers, generators and motors, which saves quite a bit of weight.  This is fairly common on ships as well.  Yes, the generators are driven from the engines but airplanes typically convert to a fixed 400Hz through mechanical or electrical means.

Very interesting stuff. I had no idea. Thanks for taking the time to reply.

I'm guessing that the appliances used in aircraft/ships are specially designed to cope with this higher frequency AC... What voltage do they use?

[kc1980]

115Vrms, three-phase.  Again, 3-phase is typical for ships and factories; it provides much smoother power delivery.  Some companies prefer the delta configuration for additional weight savings, while others prefer the y configuration which is slightly better from a safety perspective -- if a motor loses one phase, it tends to run better in the y config.

The mains are at a much lower frequency because the primary concern with long-distance power transmission is efficiency.  At 50/60Hz, the effects of inductance is much less severe.  In addition, as the frequency increases, current becomes restricted to the conductor surface, thus reducing the effective cross-sectional area through which the current flows -- that's called the skin effect.  We don't have to worry about these things as much on airplanes so we can afford to run at 400Hz.  Engineering is all about trade-offs.

[tecman]

115Vrms, three-phase.  Again, 3-phase is typical for ships and factories; it provides much smoother power delivery.  Some companies prefer the delta configuration for additional weight savings, while others prefer the y configuration which is slightly better from a safety perspective -- if a motor loses one phase, it tends to run better in the y config.

Actually there is a bit more here.  Delta is cheaper (fewer windings), less weight (as stated) and is primarily used in industrial apps where the load is line-to-line.  Most industrial 480v is for motor loads, heaters, etc and does not "need" ground.  If floating, you can have a safety issue since there is no ground reference, so some installations use a "wild leg" where a single leg is directly tied to ground, or to ground through a tap off of one phase.  Y configuration is used where you need to have some single phase equipment powered.  In the US, 120/208v is a common example, and Europe most wiring is 220/315v is common, even in homes.  The lower voltage is leg to ground, and the higher is leg-to-leg.

Paul

[Zero999]

Europe most wiring is 220/315v is common

You mean 380V.

In fact nowadays it's all 230V/400V.

[kc1980]

Wow!  I wonder if the higher household voltage affects the prevalence and/or severity of electrical accidents among average citizens -- it must!  Also, it probably makes faulty wiring that much more susceptible to shorting/arcing.  Incredible.

[safarir]

Here in Quebec, there is no 277/480v, we prefer 600v  

[alm]

Wow!  I wonder if the higher household voltage affects the prevalence and/or severity of electrical accidents among average citizens -- it must!

It's called natural selection . The difference in safety philosophy is basically 'don't make it too dangerous to get shocked' versus 'don't make it too dangerous by making sure people don't get shocked', so the total statistics are probably about even. 230V systems (which is +/- 10% or so, so both 220V and 240V are perfectly within spec) are usually designed with more safety measures.

[kc1980]

Well then!  Maybe we should raise our voltage here in the US to weed out some of the morons that are making the movie Idiocracy seem like non-fiction.  Or maybe we should just raise the voltage on Wall Street.   

[qno]

Be aware that in the new Airbus A380 the frequency can be between 380 and 800 Hz.

Just in case your application has to work in an airplane.

[xani]

Well on 230V u need thinner wires coz of lower current and if u like touching "live" wires both 110 and 230V  kills ;]

[elal1862]

Also, it probably makes faulty wiring that much more susceptible to shorting/arcing.  Incredible.

And you make it sound like that's a bad thing. 

That's another difference in safety philosophy:  'go out with a bang' versus 'smoke/smoulder/burn'

I don't see shorting as a problem; it will pop a fuse/ trip the breaker, thus bringing the faulty wiring/ equipment in a state where it should be: de-energised.  Sounds fail-safe to me.
Higher voltages = more powerful arcs = more likely that the arc will either pull so much current that the breaker trips -or - get so hot that it will simply vapourise the fault, thus breaking the current.  Problem solved in milliseconds.