radio reception interference

Status
Not open for further replies.

a.bisnath

Senior Member
I was recently asked by a farmer to install a 110volt supply and computer in his home ,I ran dedicated circuit breaker,put in an ups for any outages pretty standard stuff,everthing works well .The problem is the farmer also has a 40 year old AM/FM radio close by ,when the computer is started certain stations especially local weather oriented stations no longer can be heard on the radio the antanna for this radio is a single 16 AWG wire that goes to the roof and has always worked yeilding good reception.It is attached to the external antenna screw at back of radio.The customer wants both computer and radio to work at the same time.May I ask for some opinions and suggestion to remove this interfernce?
 

M_A_A

Member
Location
Western MA.
You should try to find the source of the interference by eliminating components. For instance, unplug everything from the UPS but leave it on,then plug in the PC and turn it on,lastly plug in the monitor and power it up. Note which device causes the interference.

Most likely it is a noisy power supply in the PC itself.

The PC is on a separate circuit correct?

The interference may be AC line induced(or RF induced through the antenna). A long shot,but if its from the ac line you might try plugging the radio in circuit fed from the other leg of the breaker panel and see if goes away.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110112-0741 EST

If the enclosure of the computer and similar equipment is well shielded, then most of the RFI (Radio Frequency Interference) is probably conducted out of the devices on the power wires or other cables.

This could be conducted by the power line into the radio, or it could be radiated from the power and other wires and picked up by the radio antenna wire.

An illustration:

I usually listen to a reasonably strong AM station in the car. The car has no direct connection to power wires. As I pull up near the shop building there is background noise that interferes with the AM signal. Starts to be noticable about 15 ft away from the building. Within the building are several computers, peripherials, and five CNC machines. This is an illustration of radiated interference. Probably most of it conducted on power wiring out of the machines and then radiated from the power wiring.

Some suggestions:

1. Do as M_A_A suggested and try to determine which components are the major source of noise. A battery powered AM radio will be a useful test tool for your troubleshooting.

2. In the AM band I had a lot of interference from my monitors. Much less noise from the computers. My computers are never fully enclosed. Almost all covers are off, and therefore not well shielded.

3. Short the antenna terminals of the customer's radio. Does this eliminate most of the interference noise? The noise from computers does not sound quite like hiss (white noise) but is similar. If shorting the antenna terminals has virtually no effect, then move the radio to another part of the room first, and after that another part of the house. How does this effect the noise level? If you do not greatly reduce the problem with the antenna terminals shorted and moved some distance away, then the noise is conducted on the power wires. Reduction of RFI with increased distance from the source will imply it is radiated.

4. If the RFI is getting out of the equipment via the wiring to the equipment, then using various types of filters may help.

5. If the noise is simply getting into the radio via the radio antenna, then you may reduce the problem by using a coax from the antenna terminals outside to the antenna.

.
 

a.bisnath

Senior Member
further checks

further checks

Firstly allow me to say thanks for all the suggestions,
the noise becomes less when the radio is moved futrher from the computer desk
the noise is there with a battery operated radio
placed radio supply on other power leg,good idea but it did not work
the noise seems to be strongest from the internet broadband modem and the flat screen monitor
where can I purchase the RF chokes online?
I like the coax idea,will try, should the braid be sent to ground?
the farmer insists he want his antique radio unmoved(RCA VICTOR brand)
the farmer is 82 years old and and can tote a 45 kg sack of feed on his back
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
I like the coax idea,will try, should the braid be sent to ground?

Yes. Also, if the radio still has a two pronged (ungrounded) AC plug, I'd replace it with a grounded plug and three conductor power cable with the chassis to ground. The house has grounded receptacles, doesn't it?
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110114-0842 EST

Old radios had AC cords with two wires and no EGC. Those without a built-in antenna usually had two terminals for the antenna, in some cases maybe only one. Usually these were an unbalanced system meaning that one terminal was connected to the chassis, call it GND, and this one would be connected to the earth via a ground rod. The other terminal, call it ANT, would connect to a long wire antenna.

Consider RG58 coax with a stranded center core wire. Connect the shield to GND and the center wire to ANT. Run a separate wire from GND outside to its own ground rod. As a starter just use a screwdriver into the earth. Run the coax to some point away from the house, and connect to the end of the antenna. Possibly having the antenna not start at the house. In other words move the antenna away from the house. You might try a 10 ft vertical rod for an antenna some distance from the house and power lines.

Sometimes the orientation of the radio's AC plug in the outlet might change the noise level.

If this produces a substantial improvement, but not good enough, then you could start playing with filtering on the equipment cables.
http://en.wikipedia.org/wiki/Ferrite_bead
You might try part number 240-2281-ND from DigiKey. This is a solid toroid with a reasonable thru hole.
http://search.digikey.com/scripts/DkSearch/dksus.dll
Here is the page for split cores
http://search.digikey.com/scripts/DkSearch/dksus.dll?Cat=3408554&k=ferrite cores split
And the following part number is for a kit that may have split cores
http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=K-404+EMI+A+LG
K-404 EMI A LG

On CRT or liquid crystal displays I believe much of the RFI radiation is from the display face.

.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110114-0927 EST

ggunn:

The reason I did not suggest using the house EGC for ground is that this actually may be conducting noise. But it is an easy try.

A lot of what may need to be done here is simply trial and error.

.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110114-0956 EST

If neither antenna terminal is obviously connected to the chassis, then do an ohmmeter check to see if the input terminals are floating from chassis.

If they are floating, then the terminal to which you connect the shield should also be connected to the chassis.

.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
110114-0842 EST

Old radios had AC cords with two wires and no EGC. Those without a built-in antenna usually had two terminals for the antenna, in some cases maybe only one. Usually these were an unbalanced system meaning that one terminal was connected to the chassis, call it GND, and this one would be connected to the earth via a ground rod. The other terminal, call it ANT, would connect to a long wire antenna.

I don't know from radios, but I deal with old tube guitar amplifiers a lot. A common mod for them is to replace the original two conductor power cord with a three conductor one, adding an EGC from the chassis to building ground. It helps with noise issues and makes them safer. I thought maybe it would be a good idea in this case. Maybe not?
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
111014-1120 EST

ggunn:

The house EGC can be an effective ground so it is an easy thing to try. Your suggestion is good because it is easy and therefore one does it first. It is not necessary to replace the existing cord. For a test one can just run a wire from the radio chassis to the wall outlet EGC.

A lot of equipment, such as computers, may have input filters that include capacitors to the EGC, and this coupling can introduce noise onto the EGC system.

The AM band is nominally 1 MHz and the inductive reactance of the EGC is greater here than in the power and audio range.

.
 

K8MHZ

Senior Member
Location
Michigan. It's a beautiful peninsula, I've looked
Occupation
Electrician
Lot's of advice here, much of it speculation.

As you may notice from my user name, I am an amateur radio operator. I have had personal experience with RFI and am familiar with the laws (Yes, there are FCC laws about RFI).

The first step is to ID the source of the noise. The recommendation of using a portable AM radio is spot on.

What we are fortunate enough to know is that the RFI is due to the installation of one of the computer's components. While monitoring the RFI, disconnect the components one at a time until the noise goes away. Percentage wise your monitor is the most likely culprit.

Once the source is identified, the reduction or removal of the RFI can be dealt with.

If this is a NEW unit, there should be an FCC sticker on it somewhere. That is the Part 15 sticker which basically says that the device doesn't emit RFI. If it does, take it back the the store and tell them you want a device that is Part 15 compliant and that you have one that was certified as such that wasn't.

If the unit is used, it depends on the device. One trick radio operators used on CRT's was to take them apart and put foil inside the case, all bonded together. However, I don't recommend it unless you are familiar with the HV stuff in a CRT. You can make a nasty situation by getting the foil in the wrong place.

Making the radio immune to RFI is not as easy. First, you have to figure out the ingress path. Is is the AC cord, the antenna or just plain front end overload? Each path will have different approaches.

DO NOT modify the guy's radio. Any modification voids the UL listing for safety.

Adding a grounding conductor to the chassis may make things worse, or even unsafe. If the radio came with a two prong cord it was NOT designed to have the chassis earthed unless there is a specific earthing terminal accessible without taking the radio apart.

Using coax to feed the antenna is a waste of time. Coax is unbalanced feed line and the radio and presumably the antenna are both balanced. Coax will just add loss to the system. Feeding the antenna with balanced twin lead is the way to go in order to get the antenna away from the computer.

You say the antenna is a single wire. Are there two terminals and only one wire? If so, that is where the feed line needs to be connected so the antenna can be moved away from the computer.

The current antenna is too close to the computer.

If the ingress path is determined to be via antenna and adding proper feed line almost, but not completely removes the RFI, a current or voltage transformer may (or may not) take the rest out.

All that being said, I had one CRT that no matter what we did, you couldn't operate a radio within 25 feet of it. I tried everything and finally just gave up. Well, I didn't try painting the inside of the case with copper paint, but did make attempts at shielding and isolation that worked on other monitors.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110114-1230 EST

Whether or not a radio input is balanced or not is only a function of whether or not the input coil is floating or one end is connected to common (chassis or ground). The input RF amplifier coil may simply be tapped near the common end of the input tuning coil (an auto-transformer), or a separate coil may be used as the primary to couple to the input tuning coil (an isolated transformer).

I stand corrected on whether or not you should connect the radio chassis to earth. If it is an AC DC radio, then you do not want to connect the chassis to earth or EGC.

Based on the description that this is an antique radio and in combination with the use of an external antenna I suspect that the power supply is from a power transformer. This provides isolation from the AC line. Radios with an antenna terminal are with very great probability going to have isolation and have two terminals, GND and ANT. I do not believe AC-DC radios came into existence until about 1935 to 1940.

A coax cable is essentially self shielding whereas twin-lead needs an additional shield to have shielding. Generally these are of different impedance levels. Most coax is about 50 ohms and typical twin-lead is 300 ohms. This is simply a matter of geometry.

Why use twin-lead if you are going to simply connect a single antenna wire? How does this differ from just the antenna wire? By a small shunt capacitance between the two wires in the twin-lead and limited noise cancellation if the non-antenna wire of the twin-lead is grounded. Whether coax or twin-lead is used it is unlikely that there is an impedance match with the antenna. Thus, some sort of standing waves will exist on the transmission line.

.
 

rattus

Senior Member
Try a battery powered radio to see if that solves the problem. If it does, that would indicate that the the transient suppression caps between the hot wire and radio chassis and between the neutral and chassis may be bad. One could make such an external network to plug into a 3-prong receptacle using the EGC as the common point. Plug the radio into the same receptacle.

Anyone remember the sizes? My memory says 0.05 microfarad, 600V ceramic or paper.

It would be cheaper to just buy a new radio.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
DO NOT modify the guy's radio. Any modification voids the UL listing for safety.

Adding a grounding conductor to the chassis may make things worse, or even unsafe. If the radio came with a two prong cord it was NOT designed to have the chassis earthed unless there is a specific earthing terminal accessible without taking the radio apart.

As I said, I don't know from radios, but we quite often modify older tube amps by installing a three conductor power cord to replace the stock two conductor unpolarized one and grounding the chassis to the ECG. Some of them also have a switch that connects a capacitor between the chassis to one leg or the other of the AC power, which we disable. This is quite effective in reducing shocks between, say, a guitarist and a microphone connected to a PA system. It appears to help with noise issues as well, though I don't have any hard data to support that. I realize that it is a design change, but the design of the power distribution system has changed since this equipment was built.

So tell me; how can earthing the chassis make a radio unsafe? I'm not trying to be provocative; I want to better understand what is going on with grounding issues.
 
Last edited:

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
xxx So tell me; how can earthing the chassis make a radio unsafe?
I suspect what is being referred to was called an ?All-American 5?. The ones I worked were mfg in the 50s to early 60s. There were dozens (at least lots) of mfgs, all were about the same with the exception of the plastic housing and the name. No transformer, five tubes, filaments in series, one side of the line cord was connected to the metal chassis. Some had an antenna connection (screw terminal) on the back. None had a ground terminal. Most had a loop inside of the case. Touching the chassis and a grounded object could easily get one a deliciously nasty surprise. Another was if the volume control knob was missing, touching the metal pot shaft could also get one a decent shock.

They worked fairly well if you were in town. They were easy to troubleshoot and repair. Probably be considered horribly dangerous today.


xxx It would be cheaper to just buy a new radio.
Of course it would be. He is 82, and can still pack a 100lb sack of grain. And, he wants his radio.

Sounds reasonable to me.

ice
 

S'mise

Senior Member
Location
Michigan
Gar said "some sort of standing waves will exist on the transmission line"
I thought he had an AM/FM radio? (Not a transeiver)

Keep in mind the two screws on the back of most old radios are for FM only. The AM antenna is usually mounted inside the radio.

If the antenna is built into the radio, the only solution I know is to move it away from the computer.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
110321-1956 EST

Most radios predating 1950 were AM only. Very few FM stations existed. Most AM radios predating the mid 40s used an external antenna. In other words prior to the end of WWII.

However, I have a 1940 Zenith portable with a WaveMagnet antenna. There was no RF amplifier, and the WaveMagnet was a coil of wire about 10" by 16" that was both the antenna, and the RF mixer coil. In other words, part of the mixer tuned circuit. This antenna was normally in the radio box, but could be pulled out and oriented to maximize the signal level. The tubes in the set were loctal vs octal with 1.5 V filaments. The set was AC-DC and battery powered. With battery weighed about 20 #. Also the standard dry cell battery was a large part of the cost of the set.

Most of the tubes of the 20s and 30s used 4 or 5 pin sockets. The octal and loctal bases, 8 pins, did not appear until the late 30s. Tetrode or pentode tubes with 5 pin bases had the grid terminal on top of the tube. Some tubes had no separate cathode, but directly used the filament as both the heater and cathode. With 5 base pins and a separate cathode the control grid was forced to be somewhere other than the base.

See http://en.wikipedia.org/wiki/Pentode for a very limited discussion of the pentode. Note: the constant current characteristics of the pentode. By contrast a triode is not a constant current device.

Was the noise problem of the original post of this thread ever solved?

.
 

hurk27

Senior Member
So tell me; how can earthing the chassis make a radio unsafe? I'm not trying to be provocative; I want to better understand what is going on with grounding issues.

Very easily

Many older tube amps, had one side of the power cable connected to the chassis sometimes through caps, sometimes not, most used a switch called a polarity switch that was supposed to get rid of any hum if there was some, it also would get rid of the shock you got from the pickups or strings of a guitar, mic, or anything else that might be tied to the un-balanced shielded 600 ohm cable that was also tied to this chassis, installing a 3-wire cord and plug tying the EGC to the chassis, you now have put parallel neutral current on the EGC's, the problem comes when one switches the polarity switch, now you have placed a dead short between the EGC and the hot, to do this you would have to rewire the whole chassis grounding since it is used as one side of the power supply through out the equipment, you would go a lot farther if you would just make sure the plug used is polarized and correctly wired, and the polarity switch defeated.

With that said, back to the OP, as was said, if it is the AM side of the radio having the problem then I doubt very much your going to find much of a way to rid this interference, computers and monitors, power supply's ETC... all produce RF at various frequencies, and AM is so prone to this kind of interference, and back in the 40's and 50's the front ends of these radios were not filtered very well, AM car radios had some of the best filtering, but even then if the interference is on the frequency that is being tuned too, there is nothing you can do to eliminate it, except moving the antenna away from the source, many of these old radios had 4 antenna screws on the back, one for the AM, a ground and 2 for the FM for 300 ohm balance feed line, I have done a 300 ohm balance line feed on stand offs with a twist every 1/4" that helped eliminate some interference then just run a long wire balance di-pole on the roof on stand off's, but if this radio has an internal antenna also without a external/internal switch to defeat it, your still going to get this interference.

I'm not quit sure how you went from being an electrician to a radio technician? but if you have completed your scope of work you need to get paid and he needs to hire a radio technician.
 
Status
Not open for further replies.
Top