The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Open loop VCO modulation
https://designers-guide.org/forum/YaBB.pl?num=1272983076

Message started by aaron_do on May 4^th, 2010, 7:24am

Title: Open loop VCO modulation
Post by aaron_do on May 4^th, 2010, 7:24am

Hi all,

I am curious what the limitations on open loop VCO modulation are. My thinking is that the biggest issue will be process variation which will create some uncertainty in the frequency deviation.

Is there a standard way to figure out for example what is the maximum data rate that such modulation can take based on expected capacitance variation etc? Will some kind of corner simulations be sufficient? I don't have access to monte carlo simulations.

thanks,
Aaron

Title: Re: Open loop VCO modulation
Post by love_analog on May 5^th, 2010, 1:04pm

What do you mean by open loop vco modulation.

as the vctl drifts, the VCO frequency changes.

Or are you talking about how much VCO frequency changes with Process.

Title: Re: Open loop VCO modulation
Post by aaron_do on May 5^th, 2010, 6:15pm

Hi,

I am talking about a transmitter design where the modulation is done by directly applying the modulating signal to the VCO. So in an open loop scheme, you would have two control voltages for the VCO. One for the normal locking of the PLL, and one controlling a different set of varactors which is driven by the modulating signal. That way you can get FSK.
So yeah, I'm curious about the effect of process variation on such a scheme, and what kind of limits could be achieved.

cheers,
Aaron

Title: Re: Open loop VCO modulation
Post by Mayank on Jun 8^th, 2010, 12:37am

Hi aaron_do,

Nice & Simple Idea, BUT

IF YOU DO NOT USE A PLL, INSTEAD JUST AN OSCILLATOR

1. VCO frequency drifts heavily with changes in Process & Temperature.
a.> For Process, you would need sm kind of trimming.
b.> For controlling Temperature Drift, you would need some Bias cktry designed so as to compensate temp effects.

2. Even after doing the above mentioned stuff, You can never be phase aligned in Open-Loop Process. So only FSK & no kind of PSK/OFDM/phase sensitive modulations can be applied.

3. Even if you compensate point no. 1, there will still remain certain drifts in freq with PVT which you need to put a spec on i.e. you need to calculate the limit of unwanted freq variations that your FSK can tolerate.

IF YOU USE GENERATE operating pt from PLL & PUMP SIGNAL INTO VCO VARACTORS;

1. When you will change your signal bits, VCO freq will vary immediately. This kind of variation will compete against VCO's single Period Jitter. You gotta take care of that.

2. PLL will try to negate the effect of Signal that you apply to the varactor. So if your signal is DC (i.e. it sits at a particular value & doesnt change for a long time), then even though you would expect your VCO Freq. output to be Fop_point + ( some gain value * your signal i/p ) , BUT over a long time, PLL will stabilize the vco freq to PLL's op. freq. So you cant see that output.

--
Mayank

Title: Re: Open loop VCO modulation
Post by aaron_do on Jun 8^th, 2010, 7:27am

Quote:

IF YOU USE GENERATE operating pt from PLL & PUMP SIGNAL INTO VCO VARACTORS

yeah that's basically what I intend. It's a common technique actually, but I haven't seen much info on the limitations. Based on what you're saying, it seems there is a minimum limit to data rate, and also the data must have an approximately equal number of zeros and ones in order to ensure the PLL does not lock onto the wrong frequency.

I'm more looking for references at this point on how such a design can be done.

thanks,
Aaron

Title: Re: Open loop VCO modulation
Post by Mayank on Jun 8^th, 2010, 10:36pm

Quote:

Based on what you're saying, it seems there is a minimum limit to data rate, and also the data must have an approximately equal number of zeros and ones in order to ensure the PLL does not lock onto the wrong frequency.

Yeah, Some for of 0/1 Padding OR Frequency Hopping OR Spread Spectrum Teechniques would be required; I Suppose.

References :--

http://www.google.co.in/url?sa=t&source=web&cd=4&ved=0CCUQFjAD&url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F4483492%2F4487950%2F04488025.pdf%3Farnumber%3D4488025&ei=kicPTJ7YE9HBrAfstaClCQ&usg=AFQjCNHghxMjG-qjcsmxgx9DgfRv82mTAw&sig2=077dCMIj-EfMd9iZ-blhpA

http://www.google.co.in/url?sa=t&source=web&cd=3&ved=0CCAQFjAC&url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F4520033%2F4529518%2F04529612.pdf%3Farnumber%3D4529612&ei=kicPTJ7YE9HBrAfstaClCQ&usg=AFQjCNFGpFE29uGg9PmseTW62OJsHPN6zw&sig2=X9dIAWugsZO_H4muCvH91g

& many more on ieeexplore.

i am currently not working on this. BUT do suggest me if you find any good reading material on this.

--
Mayank.

Title: Re: Open loop VCO modulation
Post by aaron_do on Jun 9^th, 2010, 9:09pm

Sure thing. Thanks

Aaron

Title: Re: Open loop VCO modulation
Post by rfcooltools.com on Jun 10^th, 2010, 10:25pm

for a VCO the time varying signal on the varactor will have the following equation
Vout(t)=A*sin(wt+kv*integ(x(t)))

which coincidentally is the same for FM where kv is replaced by "m" for modulation index.

With that in mind wide band fm power spectral density inherently follows a bessel functions of the first kind which are probably solved somewhere on the internet. Where amplitude of x(t) and KV determine the bandwidth and spectral content. I believe this spectral leakage in the transmitter will ultimately be the limiting factor in ACPR and thus be the limitation of the circuit.

http://rfcooltools.com