AM radio receiver - design

It's been MANY years since I've tinkered AM radio. But I still have, in my junk box, 455kHz ceramic resonators for interstage filtering with bipolar transistors. There also were emitter bypass resonators, allowing for very tight skirts. I don't know if they still make these... or, for that matter, if the ones I have are even any good now... they date to the early '60's ;-)

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
 I love to cook with wine     Sometimes I even put it in the food

Google the GE SuperRadio III schematic?

If at all possible, see if you can borrow a spectrum analyzer and take a look at whether or not there's really any signal to noise ratio to be had. In some cases it doesn't matter

The AM and SW bands are so dominated by noise that almost any receiver is plenty good enough. The only thing that will improve audibility may be bandwidth limiting or a directional antenna.

John

Yup. Somehow my message was chopped off, but your first sentence there was the jist of what was deleted... if there's no SNR to be had, the best receiver in the world isn't going to help.

For sensitivity, more antenna, preferably vertically polarized; possibly an antenna tuner:

For selectivity, a Q multiplier:

Hope This Helps! Rich

One then adds a noise canceling (phasing) circuit (with two antennas feeding it) in front of the above radio (or its equivalent).

And one then regrets the whole experiment.

Conclusion: it can be done but it is a major PITA and the results are iffy.

The antenna is where you control noise. Google magnetic loop antenna. There are commercial versions, such as made by Wellbrook. There are also home brew designs, though generally tuned rather than broadband like the Wellbrook.

If this is for voice, you can get some improvement by band limiting the audio.

If you make the AM radio a phase lock loop, you can get some improvement in the interference rejection. The simplest way to do this is to build a normal AM radio up to the detector stage. At the detector, use something like a HC4046 to phase lock onto the 455KHz signals carrier. The filter in the PLL needs to be quite slow and you have to use the XOR phase detector.

The VCO of the PLL needs to run at at least 2x the IF. You use some flip flops to make a 90 degree version of the IF for the XOR and a in part to drive a HC4053.

The 4053 sync demodulates the IF giving you the audio. Noise energy that happens to land at 90 degrees to the long term average of the carrier is rejected.

Spoken like someone with actual experience building antennas. ;-)

Take a look at the datasheet for the LM3820 AM Receiver IC

The chip is long obsolete, but if you're just wanting a hobby project, this would probably be a good place to start. I have several of those chips left in my junque bin, so if you're interested, I'll be glad to send you a few. BCB receivers typically are built with approx. 40 uV sensitivity. In-band noise will pretty much swamp out any gains from higher sensitivity. You might be able to get better performance by tight IF bandwidth control with ceramic filters, but the noise will still be there. A good magnetic loop antenna will probably be more value than higher sensitivity.

--
Dave M
MasonDG44 at comcast dot net  (Just substitute the appropriate characters in the 
address)

Life is like a roll of toilet paper; the closer it gets to the end, the faster 
it goes.

What makes you think that?

A PLL detector can give THRESHOLD reduction but that is not the same as interference rejection.

OP.. For broadcast AM reception, a directional magnetic loop antenna is a good start as others have said. Where/ what is causing the interference? Get rid of the lamp dimmers in your house.

Mark

If they are that old, they are probably still good. They still built forever back then. Besides, what would the failure mechanism be?

They're in a molded _plastic_ package :-(

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

If we knew what we were doing, it wouldn\'t be called research...
                    -- Albert Einstein

The PLL locks onto the carrier and has a filter with a low bandwidth. This means that the phase of the VCO won't change due to noise that is far from the carrier. For purposes of understanding why it works, assume that the VCO is running exactly in step with the carrier and that in effect the PLL has zero band width.

Now consider the side bands that get demodulated. A noise component that can be thought of as resulting in a signal like:

Y =3D f(t) * cos(wt) + sin(wt)

where: Y =3D the signal f(t) =3D some random (noise) function with no DC sin(wt) =3D the carrier cos(wt) =3D 90 degrees to the carrier

is the important one for the argument. This signal will result in noise from a standard AM demodulation but no noise from the PLL and sync demodulation one.

Now consider the intelligence on the AM signal. It is a function like:

Y =3D f(t) * sin(wt) + sin(wt)

The two methods of doing the AM demodulation give equal values for this input.

The result is that the PLL based method makes the same signal and less noise on its output.

Maybe in theory you are right. But anyone with a radio that has synchronous demod will testify that synch demod doesn't help with a noise source. It is good for rejecting noise from the adjacent channel if you passband shift.

The magnetic loop really does the job. The designs that float the loop above ground are more effective. [The Wellbrook does that.]

The good old superhet design can be used as the after-stage of a tuned front-end, thereby making a TRF front end to the mixsr in the superhet.

would probably be a good place to start. I have

glad to send you a few.

noise will pretty much swamp out any gains from higher

bandwidth control with ceramic filters, but the noise will

sensitivity.

TCA440 was a popular choice as well.

Anyone know of a modern variant IC that does AM (linear AGC IF) decently. All I can see now are FM IF receivers and some completely integrated radios (FM+AM+PLL etc) that are too complex and draw lots of current (>100mA).

Mark

3dB of improvement may be hard for people to notice but it is none the less real.

It prevents the carrier on the adjacent channel from multiplying with anything so it keeps all of its side bands at high frequencies so yes it does a good job of that.

huh?

You don't state whether your intereference is... "QRM" (man-made, i.e. co-channel interference from other transmitters, or even on-channel interference as a result of long- range skip, etc. or from a myriad of non-"radio" sources around the house or neighborhood) or "QRN" (natural interference from atmospherics because of a low signal to noise ratio from trying to detect a very weak signal or trying to listen during an electrical storm, etc.)

You also don't state whether you are "DX"ing (i.e. attempting to listen to station(s) outside the designed coverage area.) "Interference" when trying to listen to a local station is likely very different than "interference" when trying to listen to a station hundreds of km away.

The solution for each kind of "interference" is not the same, so it might be prudent to do some more detailed evaluation and reporting of exactly what kind of "interference" you are experiencing.

Effective solution(s) may very likely turn out to be antenna-related (or even mitigation of the source), and have little or nothing to do with the receiver circuitry itself.

A vendor here in the US (C.Crane) does a brisk business in after-market AM antennas to improve performance of your existing receiver. This page has a list of suggestions for improving AM reception and lists dozens of sources of QRM that you should suspect as causing your "interference"....

Effective antenna solutions don't necessarily have to be expensive store-bought gadgets, either. They could be long-wire antennas or home-made tuned loops, etc. There are many sources of information about AM receiver antennas that could be turned up spending a few quality milliseconds with your friend Google.