Practical VCO

[redhat]

how to design a practical vco? is it a trial and error process? i tried
to design a resonator, calculated its values and simulated it using ADS
and it worked finw when using ideal components, but, when using vendor
components like coil from coilcraft and capacitors from murata , the
output was a signal that has no specific shape,the circuit didn't
resonate at the specified frequency and even not near it although i
have recalculated the component values according to Q unloaded of the
coil.

regards

 

[Dave M]

how to design a practical vco? is it a trial and error process? i tried
to design a resonator, calculated its values and simulated it using ADS
and it worked finw when using ideal components, but, when using vendor
components like coil from coilcraft and capacitors from murata , the
output was a signal that has no specific shape,the circuit didn't
resonate at the specified frequency and even not near it although i
have recalculated the component values according to Q unloaded of the
coil.

regards

My experience with vco is a lot has to do with the frequency in question,
the higher the frequency the more drift and instability and location of components
effecting the frequency becomes more critical.

Dave






www.community4all.com
www.debsajoker.com

 

[redhat]

i am working on 950MHz center frequency,using LC resonator

 

[Bob Liesenfeld]

how to design a practical vco? is it a trial and error process?

What are you using to look at the waveform? Assuming it is a scope that
has that kind of bandwidth, what kind of probe are you using? At these
frequencys, probe issues become paramount. Also, what are you doing with
the DC control line during your test? Unless it is terminated in a *very*
low noise circuit, you will see a lot of garbage on the signal.
How is the circuit laid out? Is it on an etched PC board, or a
protoboard or what? Again at these kinds of frequencys, layout isHope this helps.

 

[redhat]

i am still simulating using aglient ADS software.

 

[maxfoo]

i am working on 950MHz center frequency,using LC resonator

Why don't you just get a free sample from Maxim and see how that compares.
The MAX2624 covers 947 to 998 MHz


http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2061/ln/

 

[Robert Baer]

Why don't you just get a free sample from Maxim and see how that compares.
The MAX2624 covers 947 to 998 MHz


http://www.maxim-ic.com/quick_view2.cfm/qv_pk/2061/ln/

Forget *anything* from Maxim; most of what they advertise is vaporware.

 

[Rene Tschaggelar]

i am working on 950MHz center frequency,using LC resonator

Ah. At theses frequencies, a coils definitely is not
just a coil anymore and a cap is not just a cap anymore.
You should take the effort to actuallt measure the
parts before simulation. Plus, the unspecific waveform
can be described in therms of harmonics. How far down
is the second, how far down is the third ?
A scope is not the right tool for this job.

Rene

 

[redhat]

so, how does anyone design a vco? if the equations of the resonator is
for ideal conditions, the parasitic capacitance has a large effect on
the resonator; how to take into account the parasitic capacitances
effect?

 

[Rene Tschaggelar]

so, how does anyone design a vco? if the equations of the resonator is
for ideal conditions, the parasitic capacitance has a large effect on
the resonator; how to take into account the parasitic capacitances
effect?

Well, either by simulating with actual parts, meaning
including parasitics, or by tweaking the circuit
afterwards. I suggest to do a simulation with actual
parts "to come close" and tweaking afterwards. Don't
forget to spend 1nH per mm of copper trace.

IMO, the gain from getting a cheaper VCO than available
as actual part has its price that ought to be paid back
by the numbers of units sold.

Rene

 

[redhat]

ok, i am going to put a varactor diode instead of one of the
capacitors, should i try to tweak the frequency of the resonator
without the varactor first and then add the varactor and tweak again,
will it be easier and more accurate or it doesn't matter put a varactor
and then tweak.

 

[Rene Tschaggelar]

ok, i am going to put a varactor diode instead of one of the
capacitors, should i try to tweak the frequency of the resonator
without the varactor first and then add the varactor and tweak again,
will it be easier and more accurate or it doesn't matter put a varactor
and then tweak.

You're going to a have a varactor anyway, how else
would you get a voltage dependent frequency ?
Depending on your requirements in terms of
harmonics, one varactor may not be sufficient.
A twin varactor arrangement has lower harmonics
since the waveform itself changes the capacitance
of just one diode. Twin diodes can be arranges
such that the selfinfluence is cancelled out.


Rene

 

[redhat]

ok, what about the reflection amplifier should i put it also then start
tuning or it is better to put it after the resonator outputs the wanted
frequency and which analysis technique is better and easier negative
resistance or feedback?

regards

 

[Rene Tschaggelar]

ok, what about the reflection amplifier should i put it also then start
tuning or it is better to put it after the resonator outputs the wanted
frequency and which analysis technique is better and easier negative
resistance or feedback?

While I did recently read a lot about that subject,
I tended to purchase VCOs for 20$ or so in single
unit quantities.


Rene

 

[Kevin Doyle]

I'm just after building a VCO myself but on 100Mhz.
I used the design and analysis procedures layed out in the book "Complete
wireless design" by Cotter w. Sayer. It covers VCO and Amp design from HF to
UHF.

Kevin.
P.S.
I spent a while building the circuits before I got it right