Cell Phone Interference

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

mattnj

Well-known member
Joined
Nov 27, 2007
Messages
89
Location
NJ
We all know what that sound when cell phones are near a computer speaker. It's like, "a funky analogy-synthy rhythm consisting of repeat Eighth-Sixteenth-Sixteenth, with an accent on the Eighth note and the sixteenths are staccato."

Are there any current audio designs that are better at rejecting this interference?

My friend can't play out of his bass amp if it's in the same room as a GSM cell phone. It's a cheapo Traynor, but still it's unbearable. He took it all apart and tried shielding the amp with foil tape.

Any thoughts?
 
Maybe you could try some round ferrit beads on power cables or little ferrit bead inside, on signal wire ?
For the bass amp, you could even add a passive low pass filter
 
are there designs that can reject his annoying interference? It seems to get more and more prevalent today. I attended a presentation with jbl self powered speakers and they were getting hit constantly. i feel like audio designs may need to take this problem with more caution.
 
It's extremely difficult owing to the power and frequencies involved.

Once I inadvertently designed a three-piece computer speaker product (for Harman in fact) that wound up being immune. The input signal processing was done with discrete devices which for noise considerations ran at rather higher-than-usual currents. There was a lot of unbroken ground plane as well.

Often the input stage of a given opamp will be the most sensitive point for rectification/detection of RF. Some manufacturers found that a ceramic disc cap with precisely trimmed lead length formed a series L-C low-Z network and could be fairly effective tacked across the input pins. I seem to recall they were 39pF parts. But this worked for one particular frequency.
 
wow. that LC network thing is cool. So if I understand correctly, this type of interference is just like any other RF interference, and depending on the circuit design, it may be more or less susceptible to the interference. I thought that this type of interference seemed much harder to shield against since so many products are affected by it. Clock radios, computer speakers, self powered monitors, home theaters etc.
Thanks for the insight.

-Matt
 
Yes, it's just like other interference.

To do a thorough job against a wide range of frequencies takes pretty aggressive approaches, with electric and magnetic shielding, all sorts of feedthrough filters, common-mode chokes, etc. etc. The shielding is bad enough, but you usually still have to get signals in and out :grin:

The military folks have to be quite diligent. Some of the standards developed are quite demanding. This reference mentions for example that a 20 lb. computer meeting Tempest standards can weigh about 75 lb.! :


http://mae.pennnet.com/articles/article_display.cfm?Section=ARCHI&C=News&ARTICLE_ID=72293&KEYWORDS=TEMPEST&p=32
 
GSM interference is due to trace lengths on the affected circuit as well as the burst-y nature of the data. The data bursts are around 217hz with the actual modulated signal being in the mid MHz to low GHz. So for 800-900 Mhz a trace of around 7cm will be a near perfect 1/4 wave antenna! 2.5GHz will be picked up by around 3cm long traces!

US GSM phones can transmit up to 30dbm, or 7V at 2GHz. That's 1W. Heck, at 900MHz they can almost do 2W!
 
Thanks for the info guys. 75 lb desktop seems ridiculous, but I guess security deems it necessary.

1W to 2W unwanted signals have to be pretty hard to shield against, which I guess is why it costs so much to do.

I've read a little about the 217hz thing, but never really understood that the board traces could pick it up.
I learned a lot from this thread so far....if only I can finish this paper which is due in a few hours!
 
Slightly OT.

I've always found the cell phone rhythm to be very much like a stilted, choked swung eighths pattern. Being a jazz drummer myself, it became a running joke among my friends that if you were "swinging like a cell phone" it meant you might need to dial down the cheese knob and find yourself some soul. :cool:

On a slightly more serious note, we have a strict cell phones off policy in the studio for this very reason. The power these devices output is rather large and can be especially loud when one of the lobes of the transmitter is pointed right at a piece of equipment.

I used to be able to tell if someone was calling me a couple seconds before the phone actually rang because the interference was so bad in my previous car. Tricked a couple less audiocentric friends that way before they caught on (which didn't take long with that loud "dat-dat-dadat").

-Matt
 
another OT story as well..

I was sweeping a product with a hand wand and a spectrum analyzer the other day. I'm specifically checking for harmonics of a 50mhz oscillator(square) around the 100-200Mhz region when I see a blip on the screen.

I wait a second and I see it again.

And again.

And again.

And again.

every second, at 124.75 Mhz.

Very high spur, around -10dbmv, noise floor around -80dbmv.

I hold the wand up in the air and this spur goes off of the screen!

Of course my director walks in and asks me why I have the wand held up like the statue of liberty..

I let him in on my discovery.

So there we were.. standing in awe of this regular blip. It's definately a beacon of some kind. We thought that perhaps it was a portable device like a pager/phone/blackberry/etc, so we cleared the room of everything, including a clock that received atomic time data to keep itself up-to-date.

Still blipping away.

I finally look up an FCC spectrum table and find that 124.75 is a low power beacon for a nearby civil airport, 1 second intervals!

Sheesh. The airport is at least 10 miles away and will still picked it up like it was nothing. I switched to another wand and the spur wasn't even half of what it was previously.

It just goes to show that the winding in the wand was just right to pick that RF up even at low powers.
 

Latest posts

Back
Top