Modeling an FM transmitter with LTspice?
dave.harper
the circuit I'm trying to model is shown here:
http://www.sentex.ca/~mec1995/circ/fmt1.htm
It takes a while to start oscillating, but when it does reach a steady
oscillation, the lower half of the waveforms look a little distorted.
Interestingly, when I reduce the inductor value (to say .01uH), the
waveforms are less distorted, but when I only decrease the capacitor
(to say 3pF), the waveforms become more distorted. Any insight into
the source of this distortion?
Could it be that I need to model an equivalent antenna?
I've copied the LTspice .asc file below.
Thanks in advance!
Dave
---------------------------------
Version 4
SHEET 1 928 680
WIRE 160 -144 48 -144
WIRE 272 -144 272 -176
WIRE 272 -144 160 -144
WIRE 416 -144 272 -144
WIRE 528 -144 416 -144
WIRE 624 -144 528 -144
WIRE 704 -144 624 -144
WIRE 416 -128 416 -144
WIRE 704 -128 704 -144
WIRE 528 -112 528 -144
WIRE 624 -112 624 -144
WIRE 48 -96 48 -144
WIRE 160 -96 160 -144
WIRE 272 -96 272 -144
WIRE -384 -80 -384 -128
WIRE -336 -80 -384 -80
WIRE -224 -80 -256 -80
WIRE -144 -80 -144 -128
WIRE -144 -80 -224 -80
WIRE -144 -48 -144 -80
WIRE 416 -48 368 -48
WIRE 528 -48 416 -48
WIRE -384 -32 -384 -80
WIRE -224 -32 -224 -80
WIRE 336 -16 272 -16
WIRE 144 0 96 0
WIRE 624 0 624 -48
WIRE 704 0 704 -48
WIRE 704 0 624 0
WIRE 336 16 336 -16
WIRE 368 16 336 16
WIRE 144 32 144 0
WIRE 160 32 160 -16
WIRE 160 32 144 32
WIRE 208 32 160 32
WIRE 416 48 416 -48
WIRE 560 48 416 48
WIRE 48 64 48 -16
WIRE 96 64 48 64
WIRE -384 80 -384 48
WIRE -224 80 -224 48
WIRE -144 80 -144 16
WIRE 160 96 160 32
WIRE 272 96 272 80
WIRE 704 96 704 64
WIRE 704 96 624 96
WIRE 48 176 48 144
WIRE 160 176 48 176
WIRE 272 176 160 176
WIRE 416 176 272 176
WIRE 512 176 416 176
WIRE 624 176 624 96
WIRE 624 176 592 176
WIRE 416 192 416 176
FLAG -384 80 0
FLAG -384 -128 +9V
FLAG -144 80 0
FLAG -224 80 0
FLAG -144 -128 +4.5V
FLAG 272 -176 +9V
FLAG 416 192 0
SYMBOL voltage -384 -48 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
WINDOW 0 -66 54 Left 0
WINDOW 3 47 51 Left 0
SYMATTR InstName V1
SYMATTR Value 9
SYMBOL res -352 -64 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R6
SYMATTR Value 1k
SYMBOL res -240 -48 R0
SYMATTR InstName R7
SYMATTR Value 1k
SYMBOL polcap -160 -48 R0
SYMATTR InstName C3
SYMATTR Value 10µ
SYMBOL res 32 -112 R0
SYMATTR InstName R2
SYMATTR Value 10k
SYMBOL res 144 -112 R0
SYMATTR InstName R3
SYMATTR Value 1000k
SYMBOL res 256 -112 R0
SYMATTR InstName R8
SYMATTR Value 10k
SYMBOL res 144 80 R0
SYMATTR InstName R9
SYMATTR Value 100k
SYMBOL res 256 80 R0
SYMATTR InstName R11
SYMATTR Value 100
SYMBOL res 400 -144 R0
SYMATTR InstName R12
SYMATTR Value 10k
SYMBOL polcap 512 -112 R0
SYMATTR InstName C1
SYMATTR Value .01µ
SYMBOL polcap 608 -112 R0
SYMATTR InstName C2
SYMATTR Value 30pF
SYMBOL cap 352 -48 R0
SYMATTR InstName C5
SYMATTR Value .1µF
SYMBOL npn 208 -16 R0
SYMATTR InstName Q1
SYMATTR Value 2N3904
SYMBOL cap 80 0 R0
SYMATTR InstName C6
SYMATTR Value .1µF
SYMBOL npn 560 0 R0
SYMATTR InstName Q2
SYMATTR Value 2N3904
SYMBOL res 496 192 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R13
SYMATTR Value 1k
SYMBOL ind 688 -144 R0
SYMATTR InstName L1
SYMATTR Value .1µH
SYMBOL cap 688 0 R0
SYMATTR InstName C4
SYMATTR Value 4.7pF
SYMBOL res 32 48 R0
SYMATTR InstName R1
SYMATTR Value 10
TEXT -440 -464 Left 0 !.tran 0 5u 0 .0001u
TEXT -440 -424 Left 0 !.ic V(in1)=2.7
TEXT -440 -512 Left 0 ;FM transmitter
Tim Wescott
> I'm trying to model an FM transmitter with LTspice. The schematic of
> the circuit I'm trying to model is shown here:
>
> http://www.sentex.ca/~mec1995/circ/fmt1.htm
>
> It takes a while to start oscillating, but when it does reach a steady
> oscillation, the lower half of the waveforms look a little distorted.
> Interestingly, when I reduce the inductor value (to say .01uH), the
> waveforms are less distorted, but when I only decrease the capacitor (to
> say 3pF), the waveforms become more distorted. Any insight into the
> source of this distortion?
>
> Could it be that I need to model an equivalent antenna?
>
> I've copied the LTspice .asc file below.
>
> Thanks in advance!
> Dave
>
What are you trying to find out with your simulation? Simulation is
usually either pointless or time consuming unless you have some narrow
thing to look for, and have a good idea what it is.
I suspect that the waveform distortion is related to the impedance of the
tank more than the resonant frequency. Reducing the inductor value
lowers impedance and raises frequency; reducing the capacitor value
raises impedance and frequency together.
Yes, you probably need to model an equivalent antenna. In fact, for many
things you may want to find out from simulation you would need to model
several different equivalent antennas.
Roy Lewallen
> I'm trying to model an FM transmitter with LTspice. The schematic of
> the circuit I'm trying to model is shown here:
>
> http://www.sentex.ca/~mec1995/circ/fmt1.htm
>
> It takes a while to start oscillating, but when it does reach a steady
> oscillation, the lower half of the waveforms look a little distorted.
> Interestingly, when I reduce the inductor value (to say .01uH), the
> waveforms are less distorted, but when I only decrease the capacitor
> (to say 3pF), the waveforms become more distorted. Any insight into
> the source of this distortion?
>
> Could it be that I need to model an equivalent antenna?
>
> I've copied the LTspice .asc file below.
>
> Thanks in advance!
> Dave
In order to start, an oscillator has a loop gain greater than one. This
causes the oscillation amplitude to continually increase. Eventually,
the amplitude is limited by something nonlinear, since it can't increase
forever. Oscillators designed for low noise, high stability, or other
precise characteristics have some stable limiting mechanism such as
diodes to limit the amplitude at a known value. Most, like this one,
depend on distortion caused by overdriving one or both transistors to
limit the amplitude. Consequently, distortion virtually always
occurs(*). To get a clean output requires filtering, and the more
distortion in the basic feedback waveform, the better the filtering
required. All this circuit has is a single tank circuit in one
transistor collector. The loaded Q of this circuit is just too low and
the fundamental distortion too great to get a clean waveform. Adding an
antenna might help or hurt the distortion, depending on the type of
antenna -- it might filter some harmonics but not others, and at
resonance it'll further load the tank. You might try tapping the
collector and antenna down on the tank, but I'd be prone to add some
additional filtering between the tank and antenna. Even that might not
work as you expect because of the complex way the antenna impedance
changes with frequency, but it'll help.
(*) It's usually not practical to filter audio oscillators, so some low
distortion audio oscillators are made by using feedback limiting that
has a very slow time constant -- one that's many cycles of the waveform.
A light bulb is often used for the limiter. In general, you can trade
loop gain for distortion -- the closer you make the loop gain in the
linear region to 1, the less distortion you get. But the slower it
starts, and it won't start at all if anything like temperature or
component value variation drops the loop gain below 1.
Roy Lewallen, W7EL
JIMMIE
>
> - Show quoted text -
Whoever said to expect something like a clean signal from a
transmitter like that in the first place. I ve build a few of those
little toy FM transmitters and the output is far from clean. Unless
this is just an exercise in Spice modeling
its probably a waste of time to do so. Intead of worrying about the
antenna I would try to figure out what was the best load for the
circuit and design an antenna to meet that requirement otherwise I
would just hang a 8 inch piece of hookup wire off it aand call it
done.
Jimmie
Brian Howie
<452cfa58-f0c3-4544...@z34g2000vbl.googlegroups.com>,
dave.harper <dave....@gmail.com> writes
>I'm trying to model an FM transmitter with LTspice. The schematic of
>the circuit I'm trying to model is shown here:
>
>http://www.sentex.ca/~mec1995/circ/fmt1.htm
>
>It takes a while to start oscillating, but when it does reach a steady
>oscillation, the lower half of the waveforms look a little distorted.
>Interestingly, when I reduce the inductor value (to say .01uH), the
>waveforms are less distorted, but when I only decrease the capacitor
>(to say 3pF), the waveforms become more distorted. Any insight into
>the source of this distortion?
>
>Could it be that I need to model an equivalent antenna?
>
>I've copied the LTspice .asc file below.
I ran the file. It looks like the output voltage is forward biasing the
collector-base junction of the output transistor Q2 , causing some
clipping. Q2 is saturating.
Brian GM4DIJ
--
Brian Howie
Tim Wescott
I was going to suggest that if he wants a clean signal he should probably
filter the snot out of the signal, and maybe even buffer it.
But he didn't ask...
brian whatcott
> I'm trying to model an FM transmitter with LTspice. The schematic of
> the circuit I'm trying to model is shown here:
>
> http://www.sentex.ca/~mec1995/circ/fmt1.htm
>
> It takes a while to start oscillating, but when it does reach a steady
> oscillation, the lower half of the waveforms look a little distorted.
> Interestingly, when I reduce the inductor value (to say .01uH), the
> waveforms are less distorted, but when I only decrease the capacitor
> (to say 3pF), the waveforms become more distorted. Any insight into
> the source of this distortion?
>
> Could it be that I need to model an equivalent antenna?
>
> I've copied the LTspice .asc file below.
>
> Thanks in advance!
> Dave
>
circuit Q way high. Tapping the antenna into the coil might serve you
better, and placing a small resistor in the emitter path before the
feedback cap would hold the gain down, but planting a resistor to
represent the antenna load might do the trick right there??
Brian W