Prev: Passive Earthquake Proofing w/ Dry Stone
Next: Best Light for reading?
From: Daku on 10 May 2010 22:28
Thanks for the feedback. My question, basically is:
How do I control the oscillation frequency of a ring oscillator based
voltage
controlled oscillator? I have created SPICE code for both varactor-LC
tank
and ring oscillator VCOs, and while I am testing the ring oscillator
based
code, I am not sure what parameters I need to tweak to see the
oscillation
frequency change.

On May 10, 8:52 pm, Tim Wescott <t...(a)seemywebsite.now> wrote:
> In general, LC tanks don't have control voltage ports, so their
> resonance frequency can't be controlled by a voltage. In VCO usage,
> there's a voltage-controlled part -- nearly always a varactor diode,
> although there are other ways to control the frequency of an LC
> oscillator, if you're feeling perverse.
Yes, as I said before, I do have a working SPICE model of a varactor
-LC VCO.
>
> > In a
> > ring oscillator based voltage controlled oscillator, the oscillation
> > frequency is independent of a control voltage. The control voltage
> > basically switches the inverters on and off. So how does one control
> > the oscillation frequency ??
>
> Like the LC oscillator, a ring oscillator is not an inherently voltage
> controlled device. Like the LC oscillator, if one wants to base a VCO
> on a ring oscillator, one must modify one's oscillator to respond to a
> control voltage.
Exactly. Consider the simplest ring oscillator consisting of 3 CMOS
inverters, wirh a PMOS and NMOS at the top and bottom of each, so
that each of the PMOS acts a pull-up and the NMOS as the pull-down.
Each of the pull-up PMOS have their gates tied, and each of the NMOS
have their gates tied as well. By applying an appropriate gate
voltage, I
can pull down each of the inverters and start the ring oscillator to
oscillate.
Now if the voltage applied to each of the NMOS is below the NMOS
threshold, there is no oscillation. However, once the start the
oscillator,
how do I vary the frequency ??

>
> There must be a real question buried in here, but you've worked so hard
> to abstract your question from what you're doing that you've lost the
> sense of it. Try again, starting with what you want to _do_ -- even if
> it's to understand why professor X and university Y has published an
> article about phase locked loops using ring oscillators.
>
> With a sensible question, perhaps we can make sensible answers.
>
> With a more detailed but still mangled question, perhaps we can help you
> make a sensible question.
>
> --
> Tim Wescott
> Control system and signal processing consultingwww.wescottdesign.com

From: Tim Wescott on 10 May 2010 23:11
Daku wrote:
> Thanks for the feedback. My question, basically is:
> How do I control the oscillation frequency of a ring oscillator based
> voltage
> controlled oscillator? I have created SPICE code for both varactor-LC
> tank
> and ring oscillator VCOs, and while I am testing the ring oscillator
> based
> code, I am not sure what parameters I need to tweak to see the
> oscillation
> frequency change.
>
> On May 10, 8:52 pm, Tim Wescott <t...(a)seemywebsite.now> wrote:
>> In general, LC tanks don't have control voltage ports, so their
>> resonance frequency can't be controlled by a voltage. In VCO usage,
>> there's a voltage-controlled part -- nearly always a varactor diode,
>> although there are other ways to control the frequency of an LC
>> oscillator, if you're feeling perverse.
> Yes, as I said before, I do have a working SPICE model of a varactor
> -LC VCO.
>>> In a
>>> ring oscillator based voltage controlled oscillator, the oscillation
>>> frequency is independent of a control voltage. The control voltage
>>> basically switches the inverters on and off. So how does one control
>>> the oscillation frequency ??
>> Like the LC oscillator, a ring oscillator is not an inherently voltage
>> controlled device. Like the LC oscillator, if one wants to base a VCO
>> on a ring oscillator, one must modify one's oscillator to respond to a
>> control voltage.

> Exactly. Consider the simplest ring oscillator consisting of 3 CMOS
> inverters, wirh a PMOS and NMOS at the top and bottom of each, so
> that each of the PMOS acts a pull-up and the NMOS as the pull-down.
> Each of the pull-up PMOS have their gates tied, and each of the NMOS
> have their gates tied as well. By applying an appropriate gate
> voltage, I
> can pull down each of the inverters and start the ring oscillator to
> oscillate.
> Now if the voltage applied to each of the NMOS is below the NMOS
> threshold, there is no oscillation. However, once the start the
> oscillator,
> how do I vary the frequency ??
>
>> There must be a real question buried in here, but you've worked so hard
>> to abstract your question from what you're doing that you've lost the
>> sense of it. Try again, starting with what you want to _do_ -- even if
>> it's to understand why professor X and university Y has published an
>> article about phase locked loops using ring oscillators.
>>
>> With a sensible question, perhaps we can make sensible answers.
>>
>> With a more detailed but still mangled question, perhaps we can help you
>> make a sensible question.

If you don't have any clue how to vary the frequency of your ring
oscillator, why do you want to use one?

Have you searched the literature? Have you Googled?

As far as I know a ring oscillator is a really crappy oscillator that
semiconductor companies build for bragging rights about how fast their
new processes go. To what practical use are you planning on putting
your ring oscillator, why won't something like an RC oscillator work,
and why would you want to control the speed of a ring oscillator if all
they're good for is going fast?

If you have no clue, if you can't find the answer from colleagues or
fellow students, and if you really do need to make the thing work -- how
about "bleeder" FETS at each stage, to lightly pull the node up or down?
You'll kill gain, which will make the thing not want to oscillate, but
you'll also slow it down.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
From: whit3rd on 11 May 2010 02:01
On May 10, 9:37 am, George Herold <ggher...(a)gmail.com> wrote:

[on the subject of VCO circuitry]

> Hey, does CMOS have a time delay that depends on the supply voltage?
> (It seems like it might.. or you could add some R's between stages.)
> Then you could control the frequency somewhat by changing the supply
> voltage.

The output impedance of CMOS does depend on the power supply
voltage, of course, as well as on temperature and unit-to-unit
variations. This quality is better developed, IMHO, in OTA
(operational transconductance amplifier) circuits.
CMOS impedance can vary by a factor of six (3 to 18V range of
4000 series gates), but the old CA3080 was good for a factor
of 1000.
From: John Fields on 11 May 2010 10:15
On Mon, 10 May 2010 09:37:56 -0700 (PDT), George Herold
<ggherold(a)gmail.com> wrote:


>Hey, does CMOS have a time delay that depends on the supply voltage?
>(It seems like it might.. or you could add some R's between stages.)
>Then you could control the frequency somewhat by changing the supply
>voltage.

---
From TI's 1989 High Speed CMOS Logic Data Book:

news:t6piu55sqenrovms7k5fuhrl8m7666thng(a)4ax.com

From: Daku on 11 May 2010 23:39
Dear Sir,
I do not intend to use a ring oscillator based VCO.
However, as you have said below, literature search
shows a lot of people using the ring oscillator based
voltage controlled oscillator in phase locked loop
designs, which is sounds strange to me, since in a
phase locked loop, the VCO oscillation frequency
must be very sensitive to the input voltage level.
Maybe the ring oscillator is designed to oscillate
at the central frequency, but frequency variation is
difficult and tricky.

> If you don't have any clue how to vary the frequency of your ring
> oscillator, why do you want to use one?
>
> Have you searched the literature? Have you Googled?
>
> As far as I know a ring oscillator is a really crappy oscillator that
> semiconductor companies build for bragging rights about how fast their
> new processes go. To what practical use are you planning on putting
> your ring oscillator, why won't something like an RC oscillator work,
> and why would you want to control the speed of a ring oscillator if all
> they're good for is going fast?
>
> If you have no clue, if you can't find the answer from colleagues or
> fellow students, and if you really do need to make the thing work -- how
> about "bleeder" FETS at each stage, to lightly pull the node up or down?
> You'll kill gain, which will make the thing not want to oscillate, but
> you'll also slow it down.
>
> --
> Tim Wescott
> Control system and signal processing consultingwww.wescottdesign.com

First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: Passive Earthquake Proofing w/ Dry Stone
Next: Best Light for reading?