AM radio

Hi guys (and gals), I hope all you Hams won=92t laugh at me too much.

So as a final piece to my current project we want to make an AM radio receiver. (This is mostly to give a =91real world=92 mixing application.) Now the only AM radio I=92ve played with was a long piece of wire as antenna to coil, variable cap, Ge diode to ear bud. The current idea is to NOT do the frequency selection with an LC circuit but with a mixer. So, antenna to HF amp to mixer to audio amp. I banged this together yesterday and it worked OK, but far from great. I was able to see some nice ~10kHz wide side bands around a central carrier. (on a spectrum analyzer) But when I tuned the LO to be centered on the carrier and listened to the sound it was disappointing. The sound level was bouncing around at about a 1 Hertz rate. I could see the same thing on the =91scope, the HF signal from the antenna was bouncing around.

My antenna was just a piece of wire maybe three meters long stuffed into the BNC jack of an amp with a 1 kohm input impedance. My first thought was that I need some sort of balun transformer on the input. (I=92m guessing the bouncing around of the input was because the other half of my antenna was ground?)

This then raises several questions. What=92s the impedance of my piece of wire. (I=92ve got to keep the transformer impedance well above this.)

OK I=92ve got several more questions, like should I use some other sort of antenna. But I=92ll wait...perhaps my first assumption is wrong!

Thanks for any advice. (discrete chuckling from Hams is OK.)

George H.

Reply to
George Herold
Loading thread data ...

It has to be phase locked, not 'tuned the LO to be centered on the carrier' if you want a clean and stable output. Any offset will cause the problem you have.

Find a copy of the ARRL handbook, or their design guides.

--
You can't have a sense of humor, if you have no sense.
Reply to
Michael A. Terrell

I guess you mean discreet. A *discrete* chuckle would sound rather like the 1 Hz bouncing you experienced :-)

Anyways. I'm no expert at this, but the "bouncing" could be caused by an AGC circuit with too short a time constant. Or an overdriven mixer stage.

Reply to
Pimpom

An example of the synchronous detection that Michael mentions...

formatting link

(at 60kHz).

...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
 Click to see the full signature
Reply to
Jim Thompson

What you're describing is a direct conversion radio, which does not work on an AM signal without considerable help.

A far easier path is to have a signal processing chain that goes

antenna ==> mixer ==> IF strip ==> detector ==> audio

I assume that you're working in the AM broadcast band: for this, an RF amplifier is unnecessary. Atmospheric noise is bad enough in the AM band that you need a pretty crappy mixer - antenna combination before the mixer noise dominates the atmospherics.

Keep the mixer, but use a local oscillator signal that's offset from the desired receive frequency by a fixed amount (the intermediate frequency). Then follow the mixer by one or two stages of amplification and filtering at this intermediate frequency, and follow _that_ with your plain old diode, and either an earbud or an audio amplifier. Traditional AM radios use a 455kHz intermediate frequency with the LO set to the sum of the receive and intermediate frequencies, but that practice was established when gain was considerably harder to achieve in the MHz, and it mandates a tunable filter on the front end to reject image frequencies (or it mandates a fixed front-end filter and a narrow reception range).

I'd be tempted to go with that fixed front-end filter, narrow range, and

455kHz IF frequency, just because it's easy. As an alternative you can use an IF anywhere from 2MHz on up. You can get pre-made ceramic resonators for 455kHz, and you can get crystal filters with good bandwidths for AM reception at 10.7MHz -- with a 10.7MHz IF frequency you could easily make the radio tune the AM band, or with a few switchable front-end filters you could get everything from 0 to 9MHz.

Unless you're interested in high fidelity AM reception (don't laugh -- folks do it) then a plain old diode detector is the way to go. There are some minor improvements that can be had by using a synchronous detector, but that's only for people who want to wring the most out of a receiver, not someone who just wants a technology demonstrator.

Your Very Best Bet of all is probably to find a block diagram or schematic of an AM radio on the web, scab your mixer into it, and have fun.

--
www.wescottdesign.com
Reply to
Tim Wescott

On a sunny day (Wed, 16 Nov 2011 09:34:51 -0800 (PST)) it happened George Herold wrote in :

An AM radio with some gain needs an AGC (Automatic Gain Control) circuit. AM signals in the medium wave and short wave range change amplitude a lot, especially far away ones (fading). The antenna impedance, unless you have a quarter wave or something, which is unlikely due to the long wavelength, will be rather high.

Most AM medium wave receivers these days use a ferrite rod as magnetic antenna (direction sensitive).

Reply to
Jan Panteltje

application.)

I'm sure he meant what he said. Heh. Heh. Heh.

(Actually, I had not realized that "discrete" and "discreet" had different spellings -- I learned to spell late, from spell checkers that went "boing" at me, so I'm still a little weak at keeping homophones, well, discrete. So I have these little unconnected errors that crop up. But I try to be discreet about discrete little lacks of ability -- when I get lax, though, I screw up).

--
www.wescottdesign.com
Reply to
Tim Wescott

That's inherent in the math. The LO is not exactly the frequency of the incoming carrier.

John

Reply to
John Larkin

Grin, yeah no spell checker can fix all my mistakes.

George H.

Reply to
George Herold

.
g
z

Ahh, OK.

-
Reply to
George Herold

.
g
z

OK that's it! Thanks Michael. I tweaked the local oscillator by 1 Hz and that changed the period of the bouncies. (I made a little transformer but that made no difference.)

Yup I've been reading the ARRL.

George H.

-
Reply to
George Herold

h.

ng

tz

m
t

=A0 =A0 ...Jim Thompson

=A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 | =A0 =A0 et =A0 =A0 =A0|

=A0|

=A0 =A0 =A0 |

ide quoted text -

Thanks Jim, that's a few more transistors than I like to deal with. :^)

George H.

Reply to
George Herold

If you want to do it that way, you have two options:

  1. Phase lock your local oscillator to the incoming carrier to get rid of the beat.

  1. Make it an I/Q radio, square the outputs of the mixers, add them together and take square root of the sum. This is doable in the analog world, but somewhat tedious. If the mixer outputs are digitzed, the processing is quite easy in a DSP.

--

Tauno Voipio, OH2UG (no chuckles for a good question)
Reply to
Tauno Voipio

.

ion.)

on

o

ht

of

nd

n

al

re

Thanks Tim, This is all for a new apparatus about Fourier methods... and mixing is a part of it. I'm not out to build the best or even a good AM radio. The idea was to take the pieces we had (mixer's, frequency source, some amplification )and 'do something' with them.

We are 'stupid physicst's' here some days, with ideas that only work 'in theory'. (In theory, practice and theory are the same, in practice, they are different.)

George H.

Reply to
George Herold

h.

.
m
t
.

Yeah, even those who don't speak English as a native language, have better English than I do. But I will agree, that the spell checkers that underlines my mispelled words and let's me go back and try and fix them has helped my spelling to no end. (imagine how bad it use to be :^)

George H.

Reply to
George Herold

,

tenna (direction sensitive).- Hide quoted text -

Thanks Jan, I was guessing something like 1 k ohm for an impedance. I wound a little transformer on top of a 1 mH inductor and it seemed to work OK... well it did nothing to improve my bouncies.

I'll have to search on the web for a ferrite antenna. I was looking through the digikey and mouser catalog, but found nothing.

George H.

Reply to
George Herold

I think you need the AGC working properly

formatting link

Reply to
halong

He has no AGC.

Reply to
John S

Well, what you did should work as expected -- as long as you bend enough theory toward it.

--
Tim Wescott
Control system and signal processing consulting
 Click to see the full signature
Reply to
Tim

.............[snip]...............

Well, it's not really an "even" thing. We non-native users have the advantage of having had to learn the language systematically, starting with speaking and spelling simple words at the same time. This is particularly true in countries like India where the study of English is introduced at the primary school level. The B side is that our sentences are often stiff and stilted.

Reply to
Pimpom

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.