# Radio Intermediate Frequencies

Discussion in 'Electronic Design' started by James Douglas, Feb 15, 2006.

1. ### PN2222AGuest

I was hoping that I could simulate a circuit with two input frequencies,
Here's a different way to observe the heterodyning phenom.

You got MS Excel, right?
Generate a column of numbers equal to sin(1KHZ*t), a column of numbers
sin(5KHZ*T). A third column is the product of the two.

Generate a graph of the third column to see the sum / difference
frequencies.

Get ambitious. Design a fourth column that eliminates the high-frequency or
low-frequency components.

Parameterize the frequencies.

Have some fun.

PN2222A

2. ### Rich GriseGuest

<snip>

They are not necessarily mutually exclusive characteristics.
[/QUOTE]

A fact which sometimes confronts those who endeavour to bullshit
the bullshitters. ;-P

Cheers!
Rich
(OBTW, there's no apostrophe in ordinary plurals, like 'guys'.)

3. ### John Jardine.Guest

Question comes up quite often and I must admit it's surprisingly awkward
putting some representative bits together without throwing the baby out
with the bathwater. (fun though!).
The maths is minor. It's the visualisation and implemenation where the art
and complexity resides. This circuit is about as simple as it gets and
should go into Circuitmaker OK. I've also stuck on a Switchercad listing of
the same thing.

___ IN914
.--------|___|-o-->|---o--------o---------o
| 1k | | CC | |
| 2Vpp | | .-. .-.
/+\ 1kHz DD| | | | | |
( ) | | | | | |
\-/ | | '-'1M '-'1M
| _1k | .-. AA | BB|
| .-|___|- | | .-o-. .-o--.
| | | | | | | |
| /+\ '-' C| | C| |
| ( )4Vpp 100k| C| --- C| ---
| \-/ 4kHz | C| --- C| ---
| | | | |28nF | |10nF
o------o---------------o------o---o-----o----o--
0V 100mH 100mH

(created by AACircuit v1.28 beta 10/06/04 www.tech-chat.de)

Point of interest is that the waveform at "DD" is actually a true 'mixing'
of the 2 sinewaves. The voltage waveform may look distinctly interesting and
'modulated' but it isn't, it's still just two boring added sinewaves.
The point at "CC" (after the diode rectifier) is the needed true
'multiplication' of the 2 sinewaves, i.e. they really have been modulated.
Notice also that the waveform at "CC" looks nothing remotely like the 2,
neat, sum and difference frequencies the books talk about.
The critical point is that the "CC" diode created, massively distorted
undulating waveform, can now be broken down into -numerous- individual
sinewaves. Two particularly strong and useful ones, being a sum frequency of
5kHz and a difference frequency of 3kHz. (each about 90mVac).
There's also even bigger amounts of the original 1kHz and 4kHz in there,
along with shedloads of other rubbish and is the reason for the added the
tuned circuits.
Theses are sharply resonant at 3kHz ("AA") and 5kHz ("BB") and will for
demo' purposes act as a simple spectrum analyser by cleanly picking out
these particular frequencies from the general bedlam, just the same as a
radio 'I.F. strip' would.
If Circuitmaker has an 'FFT' option then the tuned filters are of no
consequence.
I'd also add, that the 2 input sinewaves don't have to be sines or those
particular voltages. They were just picked for clarity and in reality can
be any old waveshape.
john

(Switchercad *.asc drawing follows. Do a '.TRAN' of about 0.3secs)...

Version 4
SHEET 1 880 680
WIRE 0 208 0 32
WIRE 0 336 0 288
WIRE 96 208 96 160
WIRE 96 336 96 288
WIRE 144 32 0 32
WIRE 144 160 96 160
WIRE 240 32 224 32
WIRE 240 160 224 160
WIRE 240 160 240 32
WIRE 272 32 240 32
WIRE 384 32 336 32
WIRE 384 112 384 32
WIRE 384 320 384 192
WIRE 496 192 496 176
WIRE 496 288 496 272
WIRE 528 32 384 32
WIRE 528 64 528 32
WIRE 528 176 496 176
WIRE 528 176 528 144
WIRE 528 288 496 288
WIRE 528 320 528 288
WIRE 560 176 528 176
WIRE 560 192 560 176
WIRE 560 288 528 288
WIRE 560 288 560 256
WIRE 640 192 640 176
WIRE 640 288 640 272
WIRE 672 32 528 32
WIRE 672 64 672 32
WIRE 672 176 640 176
WIRE 672 176 672 144
WIRE 672 288 640 288
WIRE 672 320 672 288
WIRE 704 176 672 176
WIRE 704 192 704 176
WIRE 704 288 672 288
WIRE 704 288 704 256
FLAG 0 336 0
FLAG 96 336 0
FLAG 528 320 0
FLAG 672 320 0
FLAG 384 320 0
SYMBOL voltage 0 192 R0
WINDOW 3 -84 122 Left 0
WINDOW 123 -85 152 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value SINE(0 1000m 1000)
SYMBOL voltage 96 192 R0
WINDOW 3 -18 122 Left 0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V2
SYMATTR Value SINE(0 2000m 4000)
SYMBOL diode 272 48 R270
WINDOW 0 32 32 VTop 0
WINDOW 3 0 32 VBottom 0
SYMATTR InstName D1
SYMATTR Value 1N914
SYMBOL res 240 16 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R2
SYMATTR Value 1k
SYMBOL res 240 144 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R3
SYMATTR Value 1k
SYMBOL ind 480 176 R0
SYMATTR InstName L2
SYMATTR Value 100m
SYMBOL cap 544 192 R0
SYMATTR InstName C2
SYMATTR Value 28000pF
SYMBOL res 544 160 R180
WINDOW 0 36 76 Left 0
WINDOW 3 36 40 Left 0
SYMATTR InstName R1
SYMATTR Value 1meg
SYMBOL ind 624 176 R0
SYMATTR InstName L1
SYMATTR Value 100m
SYMBOL cap 688 192 R0
SYMATTR InstName C1
SYMATTR Value 10000pF
SYMBOL res 688 160 R180
WINDOW 0 36 76 Left 0
WINDOW 3 36 40 Left 0
SYMATTR InstName R4
SYMATTR Value 1meg
SYMBOL res 368 96 R0
SYMATTR InstName R5
SYMATTR Value 10k
TEXT -90 26 Left 0 !.tran .3
TEXT 368 8 Left 0 ;CC
TEXT 488 160 Left 0 ;AA
TEXT 632 160 Left 0 ;BB
TEXT 224 8 Left 0 ;DD

4. ### Bob MastaGuest

<snip>

Note that an ideal multiplier would output only the sum and difference
frequencies, not the originals. But it is much simpler to simply
build a non-linear amplifier that has strong distortion components
(quadratic distortion) at the sum and difference frequencies, and
it is OK since everything but the diference is filtered out by the IF
stage.

Best regards,

Bob Masta

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com
Home of DaqGen, the FREEWARE signal generator

5. ### Anthony FremontGuest

Ok, thanks for the correction and extra info.

6. ### GuestGuest

: I believe a Direct Conversion design, which eliminates the need for an IF,
: makes separating the modulation from the RF signal quite simple. Direct
: Conversion uses demodulation rather than envelope detection, so the signal
: is baseband audio straight out of the mixer. A simple low pass filter (RC)
: will prevent RF from affecting the following audio stage(s). If one wishes,
: wideband and narrowband audio filters can be switched in/out for "music" and
: CW.

Don,

While direct conversion has the advantages that you mention, it
also has some serious disadvantages. One of the biggest disadvantages is
that Direct Conversion Receivers suffer from large DC offsets caused by
self-mixing of the LO (i.e. LO couples into RF signal.) Most of the
solutions to this problem are not very elegant -- they involve
calibration/trimming/etc. It's also difficult to prevent the coupling of
the LO into the input signal, because of the way the mixers, themselves
are typically designed (I've only worked with CMOS mixers -- this may not
apply as much to Bipolar designs.)

Joe

7. ### Don BoweyGuest

The problem you describe is certainly possible with a poor design, such as
using a single diode mixer and not using an isolation stage between the
signal input and the mixer. But these are avoidable.

More importantly, the OP asked about methods (I assumed modern ones) that do
not use an IF. Direct Conversion can easily provide good results, just as a
superhet can provide good results *if properly designed*.

Don

8. ### James DouglasGuest

I am glad I checked back on my post T H A N K S. I also recently
downloaded still another simulation package, although not as easy
to use as CircuitMaker yet it's seem Ok,

Anybody using B2 SPICE v5?