# CLapp LC oscillation

Discussion in 'Electronic Design' started by jason, Mar 19, 2005.

1. ### jasonGuest

Hello All

I am studying about Clapp LC oscillator.
Given below link is the circuit of interests
http://www.geocities.com/jason_class/ClappLC.jpg

May I know how do we calculate by hand if this circuit gonna oscillate
or not?

I learnt in lecture that I must find Av multiply by beta(feedback
factor). Av is always given by gm multiply by R efficient , which is
gmReff
Reff is the sum of series and parallel resistances of all emitter
resistance, ac emitter resistance , load resistance and also reflected
resistances from the tap of capacitors. Emitter is mentioned here
because I learnt about bipolar transistor configuration.

For the circuit in the link given, I do not understand why the same
approach is use?
Why we can use gmReff as the Av since it is a common drain
configuration? Is it because we have already sum all the net
resistances in Reff , therefore we can use gmReff as the Av?

So for other type of configuration of mosfet such as common gate and
common source? We can use the same Av=gmReff ?
If possible show me the correct way of checking if the circuit can
oscillate by manual calculation..at least the steps or methodology.

Kindly enlighthen

rgds and thanks
Jason

2. ### Don PearceGuest

First can you tell us the purpose of C1, C2 R5 and R6. The capacitors
are unnecessary, and the resistors will degrade the quality of the
oscillator.

d

Pearce Consulting
http://www.pearce.uk.com

3. ### jasonGuest

Dear Don

Thanks for the response.
C1 and C2 should be for coupling purposes. Then R5 and R6 should be for
high input and output impedance to and from the amplifier.
That's my guess. If wrong, kindly correct me.

My main question is why my lecturer can use Av=gmReff for both the case
of bipolar transistor and also mosfet for different
configuration(common gate, common drain, common source, EC, CB or CC)

rgds and thanks
jason

4. ### John WoodgateGuest

In oscillator circuits, it's not correct to use the terms 'common
drain', common gate, common source'. This is because the apparent
'common' electrode is determined by the *external* circuit, usually
where the earth/ground connection is made, and that doesn't affect the
conditions for oscillation.
Yes, if you choose to take for the 'input' and 'output' of the
amplifier, the points in the circuit which makes the equation true.
An oscillator circuit is a loop, and you can break the loop at *any*
point in order to determine the loop gain. But the calculation is easier
for some places in the loop than for others.

5. ### John WoodgateGuest

There is no DC path through the Clapp network, so you don't need C1 and
C2.
You don't need them either. The three capacitors in the Clapp network
are not only for tuning; they are also for impedance transformation. If
the inductor were tapped at two places, you would see it as a
transformer. Well, you can do (almost) exactly the same by connecting
two or more capacitors (three in the Clapp circuit) in series across an
untapped inductor and taking taps off between the capacitors.

6. ### jasonGuest

Dear John

Thank you so much. You have taught so much I will never have in class
room.
But can you further explain about the followings?
I do not understand it

<<In oscillator circuits, it's not correct to use the terms 'common
drain', common gate, common source'. This is because the apparent
'common' electrode is determined by the *external* circuit, usually
where the earth/ground connection is made, and that doesn't affect the
conditions for oscillation. >>

The common gate drain or source is refer to the amplifier stage(the
simple mosfet in my case). Is that incorrect? Or you mean amplifer is
the external circuit?

<<Yes, if you choose to take for the 'input' and 'output' of the
amplifier, the points in the circuit which makes the equation true. >>

The main problem is I do not know which is the point of the "input" and
"output" that makes the equation true.
Or if you have any online document thats explains this well and comes
with example.

Kindly enlighthen
Thank you all

rgds
Jason

7. ### jasonGuest

Dear All

designer but a student taking IC design course. I am new in this.
Kindly explain in details if possible.
I know I am bold and ashame for asking this but I hope you can give me
a helping hand. Please share me me if u have any good online material
on this topic
Thank you so much in advance
My apology for any inconveniences caused
Thank you

rgds
Jason

8. ### Reg EdwardsGuest

All LC and crystal oscillators are the same.

Tubes or transistors

They all behave in the same way.

Changing the point of the ground connection has no electrical effect.
It merely changes the name.

The only way to understand how they work is to study the simple
elementary arithmetic involved. Otherwise one must suffer in
ignorance.

Unfortunately, in the Western World's schools, the standards of
education in elementrary arithmetic during the last 30 years has
fallen to abysmally low levels.

I once met a prospective member of parliament, no doubt with visions
of becoming the Chancellor of the Exchequer, who was unable to
calculate that 7 times 9 equals 63. I'm pleased to report he
forfeited his £500 deposit.

No chance of HIM ever understanding a Clapp or a Collpits or a
Hartley.

9. ### Fabio G.Guest

Linearize it with the small signal analysis and write the expression of
the loop gain in the Laplace domain.

If a frequency in which the phase shift is zero exists, the circuit may
oscillate.
If at that frequency the loop gain is >= 1, the circuit will oscillate.
This is the basic methodology to analyze oscillators.

Bye

10. ### Jim ThompsonGuest

I think that's a necessary condition for oscillation, but not a
sufficient condition for starting. Probably requires application of
one of Lyopanov's tests.

...Jim Thompson

11. ### GenomeGuest

Hi Cheang Long, can I take a guess that you are Chinese?
Absolutely no offence meant. It's just for the next bit....
Perhaps, if you are Chinese you should be referring to it as the Crapp
Oscillator. Go on, smile.....?

I know nothing about them either but I was so intrigued that I thought I'd
have a go.

Mind you, you have to promise that your real name is not Paul Burridge....
that's an 'in' joke by the way.

I notice, from your website schematic that you are using LTspice. That is
way cool because I do too. I am the most excellent person on the planet so
you can't be too bad yourself.
Well, if I figure it out I'll tell you.
That sounds about right or rather, that's the sort of stuff the old bloke at
the front of the room has being burbling on about for so long that he missed
the point in the first place and isn't very creative sexually. I guess he
doesn't excite students either.
But, you don't want to do that.....

What you want to do is a loop analysis that runs around things from the
input back to the input (the loop) without quite getting there. Scary stuff
but that's the way it is.... Anything else is chips cooked in old vegetable
oil.
Ah but,

First you have to look at your circuit and figure out what it thinks it's
doing. I am anally retentive so I like pretty circuit diagrams.

Since you have LTspice you will be able to do something with this.....I have
faith.

Select the text. Copy it to the clipboard. Open Notepad and paste the
results into it. Save the file with a suitable name with a .asc extension

Version 4
SHEET 1 880 680
WIRE 496 80 528 80
WIRE 528 496 336 496
WIRE -32 496 -32 464
WIRE -32 272 -32 208
WIRE -32 128 -32 80
WIRE -32 80 336 80
WIRE -32 80 -160 80
WIRE -160 80 -160 400
WIRE -160 464 -160 496
WIRE -160 496 -32 496
WIRE -160 0 368 0
WIRE 368 0 368 80
WIRE 368 80 416 80
WIRE -160 528 -160 496
WIRE 528 304 528 80
WIRE 528 384 528 496
WIRE -32 336 -32 368
WIRE -32 368 112 368
WIRE -32 368 -32 400
WIRE 336 496 112 496
WIRE 336 304 336 80
WIRE 336 80 368 80
WIRE 336 400 336 496
WIRE 112 400 112 368
WIRE 112 368 256 368
WIRE 112 480 112 496
WIRE 112 496 -32 496
WIRE 256 368 256 32
WIRE 256 368 288 368
WIRE 256 32 -160 32
FLAG -160 0 VD
IOPIN -160 0 Out
FLAG -160 528 0
FLAG -160 32 VG
IOPIN -160 32 Out
SYMBOL res 512 64 R90
WINDOW 0 -27 62 VBottom 0
WINDOW 3 -30 61 VTop 0
SYMATTR InstName R1
SYMATTR Value 1K
SYMBOL ind -48 112 R0
WINDOW 0 42 45 Left 0
WINDOW 3 41 69 Left 0
SYMATTR InstName L1
SYMATTR Value 100µ
SYMBOL cap -176 400 R0
WINDOW 0 45 23 Left 0
WINDOW 3 45 45 Left 0
SYMATTR InstName C1
SYMATTR Value 220p
SYMBOL cap -48 272 R0
WINDOW 0 40 24 Left 0
WINDOW 3 41 47 Left 0
SYMATTR InstName C2
SYMATTR Value 1n
SYMBOL cap -48 400 R0
WINDOW 0 47 23 Left 0
WINDOW 3 48 47 Left 0
SYMATTR InstName C3
SYMATTR Value 1n
SYMBOL voltage 528 288 R0
WINDOW 0 40 43 Left 0
WINDOW 3 46 67 Left 0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName VDD
SYMATTR Value 20V
SYMBOL njf 288 304 R0
WINDOW 0 56 37 Left 0
WINDOW 3 56 62 Left 0
SYMATTR InstName J1
SYMATTR Value 2N3819
SYMBOL res 96 384 R0
WINDOW 0 39 43 Left 0
WINDOW 3 39 63 Left 0
SYMATTR InstName R2
SYMATTR Value 1E6
TEXT -168 576 Left 0 !.tran 0 200u 0 uic

And you should be able to open it up in LTSpice and run it.
Now, once you have seen MY circuit oscillate you should begin to realise
that I have an Humungeously (JW) Enormous Penis and I am Gods gift to
TWomanKind.....

OK, I was lucky.

I knew a bit and fiddled a bit and it didn't work. So I thought something
else and took a guess and it had a go. So I changed something else and off
it went. And that reminded me of something else.

What I will say for now is L1 and C1 form a parallel resonant network.
You may beat me up later.

Go on, write me the impedance of a parallel LC network. I'm not being a
bastard here, if you can't do it then you haven't been taught it....

I've been taught it but, in the existensial sense.... it doesn't (make
sense).

And yes, I am making this up as I go along. I apologise for my liberal use
of English thingummyjigs.

DNA

12. ### Reg EdwardsGuest

======================================

Touch an RF point with a damp finger. If it stops oscillating then it
shows it's working and there's not much to worry about.

======================================
======================================

I do wish you guys would stop referring to Laplace. It's an indication
you are not fully aware of what you are waffling about.

It was Heaviside's revolutionary mathematics, 100 years later than
Laplace, which solved the trivial yet-to-appear problem of how
oscillators work - and a great number of other fundamental, physical,
hitherto insoluable engineering problems.

It was Laplace's relatively trivial piece of work, of not much
consequence at the time, which by chance provided the link between
Heaviside's masterpiece and the rest of the World of Mathematics. It
was the missing link, discovered more or less by accident by the early
practical communications engineers, needed to convince the university

But it was the blockheads, confidence tricksters, obtaining money
under false pretences, not the engineers, who wrote the subsequent
text books. And so Laplace's s highjacked Heaviside's p and D.

Laplace, honoured, of the Napoleonic era, who survived Madam
Guilotine, is amongst the world's greatest mathematicians. No doubt an
honest man. He must be turning over in his grave at the injustice.

Heaviside, a self-educated young genius, died with little recognition
of the fundamental mathematical and engineering importance of his work
after many years as a lonely recluse in 1925 aged 75.

He will be remembered for his mathematical prediction of "The
Heaviside Ionospheric Layer" via which world-wide HF communications
were to blossom a few years afterwards.

Quotation (to the blockhead university professors) : "Shall I refuse
to eat my dinner because I do not understand the processes of
digestion? "

13. ### jasonGuest

I would like to thank people who gives constructive ideas. Thank you

14. ### Robert BaerGuest

C10 and C14 do the DC blocking, so C1 and C2 are redundant.
R5 *decreases* the coupling (drive) to the LC network, thereby
increasing the feedback loss.
R6 is non-functional unless the value approaches 10% of R1 or more -
in which case it also would increase the feedback loss *and* add a phase
shift that can be undesirable.