Phone noise on an audio set, how to explain this to layman

[Kjelt]

How can I tell in layman's terms why an 800MHz or higher portable phone with its 1W power can cause interference on an audio set (20-22kHz), you know the noise just before the phone rings.
I have real problems explaining this to a layman. Who can help ? Perhaps an idea for a Dave's eevblog episode.

[coppercone2]

envelopes and mixing

draw a bunch of sine waves with a line on top like their smoothed by a frequency specific rectifier and cap to make pulses

maybe an equivalent circuit could be a HPF that takes the line noise in, followed by a rectifier and filter to make some kind of pulse wave that is then fed back into the original signal.

app note 96 does a good job explaining dc offset with rfi rectification. just modulate the offset

http://www.analog.com/media/en/training-seminars/tutorials/MT-096.pdf

[bson]

Semiconductor junctions rectify RFI into DC, producing an offset voltage.  When a transmitter is pulsed you end up with a pulsing offset voltage, and if the pulsing frequency is within the bandwidth of the circuit it will manifest itself as an induced squarish wave.  If the circuit is an audio circuit and the pulsing is in the audible portion of the spectrum you will hear it as a square wave, which sounds raspy, especially at low pulsing frequencies.

[Brumby]

The key phrase here is "in layman's terms". 

I would presume that this means any arguments based on electromagnetic theory are going to be less than helpful - so I offer the following explanation:

First, I expect the person can understand how a radio signal might get into an audio circuit - but does not understand how something operating at 800MHz could be heard.  That being the case.....

Imagine you are in a car driving over a smooth roadway and that your car has reasonably good soundproofing.  You pretty much aren't going to hear the tyre noise - and if you did, it would be a constant low level noise that you would soon ignore.  This is like a radio transmitter that just sends out a steady, unchanging signal.  Whether tyre noise or a radio signal, unless you are actually looking for either one, you just aren't going to notice anything.

Now let's say you just drove over an expansion joint.  You will get a little "thump" because of a change in the tyre and road conditions.  You still aren't hearing the tyre noise - but you can hear a (different) noise when something changes quickly.  A single little thump might not catch your attention either - but let's say you stray over a ripple strip.  Now you have a lot of little (sometimes not so little) thumps - and you will certainly notice that.

It is the same sort of thing that happens with a radio signal.  If you have a sudden change in the signal, then you could hear a click.  This is not you hearing 800MHz - but hearing a change in that 800MHz.  Now let's say there are a number of such changes that happen at a rate that we can hear.  Just like the ripple strip, we hear a series of clicks - which we would describe as a buzz.

The 800MHz signal isn't going to get far in the audio chain, but it doesn't have to.  The buzz that gets created becomes the audio signal that fits right in the zone - and will travel along it just like any other audio signal.

For entertaining examples of this, check out some videos by searching for "musical road".

[Richard Crowley]

It is not clear (at least to me) WHY you have to explain it?

Isn't the fact that it happens prima facie evidence?  And that the results are undesirable?
There were problems with RFI long before cell phones came along.

There is a story (not authenticated) that one famous Hollywood director banned the presence of cell phones on (or NEAR) the set.  His assistant had a hammer and bag of nails available any time a cell phone of the cast or crew went off (either audibly or RFI into the audio) during production.  The director would confiscate the offending device and actually nail it to the nearest doorpost (indoors) or fencepost (outdoors).  There have been times when I dearly wished I could do that.   

[tkamiya]

I think you can say something like "because your phone is so close to our equipment, it's signal is enough to cause interference" and leave it at that? 

If you get into signal strength and frequencies, you'll lose people's attention rather quickly.

Come to think of it, I haven't heard that sound in recent years.  My PC's speaker used to pick it up quite readily.  I wonder what changed? 

[Brumby]

It is not clear (at least to me) WHY you have to explain it?

For the same reason that anything is explained to anyone.  To provide understanding.

I know that you could drive a bus through some holes in my effort, but it wasn't meant to be a full engineering explanation - just to try and illustrate how the base carrier would not be heard, but the modulation would.  Even with my lame explanation, a lay person could then be taken further along this line of discussion for basic understanding of other phenomenon.  If they ever arrive at the point of wanting to get into the technical nitty gritty, then we get off the musical road and into the lab.

[Richard Crowley]

The pulsed RF signal from cell phones is rather similar to the kinds of signals developed by the "Evil Empire" back during the Cold War for jamming international radio broadcasting. 

Note that the interference happens at regular intervals, not just before/during conversation connections.  GSM cell phone interference is especially bad when the subject is wearing a wireless microphone and has an active cell phone in their pocket.  The unblanced, low-level connection between the microphone head and the "belt-pack" transmitter is especially vulnerable to that kind of RFI.

https://youtu.be/aOMVdOc9UbE

https://youtu.be/uPYyrS7fc7g

[bson]

Note that the interference happens at regular intervals, not just before/during conversation connections.

Any idea what it's doing?  I would guess it's synchronizing with the tower for the UMTS WCDMA frequency hopping and code sequencing, but it seems only GSM does this kind of pulsing.

[janoc]

It is not clear (at least to me) WHY you have to explain it?

Isn't the fact that it happens prima facie evidence?  And that the results are undesirable?
There were problems with RFI long before cell phones came along.

There is a story (not authenticated) that one famous Hollywood director banned the presence of cell phones on (or NEAR) the set.  His assistant had a hammer and bag of nails available any time a cell phone of the cast or crew went off (either audibly or RFI into the audio) during production.  The director would confiscate the offending device and actually nail it to the nearest doorpost (indoors) or fencepost (outdoors).  There have been times when I dearly wished I could do that.   

That's a good solution. Some years ago I have been invited to a wedding where the bride-to-be made sure they confiscated every phone of everyone outside the church - there were a lot of businessmen that tend to think they are indispensable and the world would collapse if they didn't answer the phone. Some had even two or three cellphones on them. It looked like the typical movie scene where a guy gets arrested and has to hand over his guns - "OK, the one on the belt, the one in the pocket - and yes, the one in the boot as well!"

All good and well - and then in the middle of the ceremony a phone goes off, with that trademark RFI into the PA system.

It was the priest's phone.

J.

[Kjelt]

Thanks all, quite usefull answers.
So what is the "opening" this gets in the audiosystem, I suspect the interlinks which if not balaced are poorly shielded ? Or are the holes in the chassis for convection already large enough?

[Brumby]

While openings in a case might provide some avenue for RF penetration, my observation is that wires running to the amplifier are a prime source.  They act as antennae, picking up any EM radiation around.  Inputs are normally shielded, but given a strong RF signal, even those shields may not be sufficient.  Power leads can also pick up RF which can make its way into the circuitry - and then there are the speaker leads.  These are usually unshielded, making them a good antenna.  They can pass RF into the amplifier circuitry and may be enough for it to find its way to a susceptible part of the circuitry.

[Kjelt]

Yes I also have doubts about speakerwire, I thought since they are driven by low impedance outputs they would be unsusceptible, if not it makes me wonder about companies selling these kinds of "litze style" speakercables, excellent antennas.
https://goo.gl/images/HrUSXy

[Richard Crowley]

Audio gear is generally susceptible to RFI penetration especially through input connections, and especially low-level (high-gain) microphone inputs.  It is possible to add RF filtering to most input (and output) ports, but most audio gear doesn't implement RFI filtering since it has been a relatively rare problem, at least before the cell-phone era.

One of the most popular brands of audio connectors (Neutrik) even makes a range of connectors with RF filtering built-in.  NC3MXX-EMC

[Marco]

How can I tell in layman's terms why an 800MHz or higher portable phone with its 1W power can cause interference on an audio set (20-22kHz), you know the noise just before the phone rings.

If you want to put in effort then show don't tell. Try to construct a discrete audio circuit with high input impedance and input protection prone to mixing (low capacitance RF Schottkys?) and make a phone call. Then you can point to the diodes and try to explain mixing and say rectifying junctions are present in all amplifiers.