better 4046 PLL

My assertions confirmed. There's no hope for you. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

You can start an LC oscillator instantly, with the right initial conditions. And you can kill one pretty dead in less than one cycle by shunting the LC with the right resistor.

A free-ringing LC circuit, absent new external inputs, runs at a constant frequency from this instant forward. It doesn't settle.

--

John Larkin         Highland Technology, Inc 

jlarkin at highlandtechnology dot com 
http://www.highlandtechnology.com 

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom laser drivers and controllers 
Photonics and fiberoptic TTL data links 
VME thermocouple, LVDT, synchro   acquisition and simulation

Except that you usually ask it of regular posters who have just shown you up as not knowing what you were talking about.

I really hate agreeing with Jim Thompson, but on this point he's got back in touch with reality.

--
Bill Sloman, Sydney

.

next cycle will be bigger.

ong?)

OK, I use to have this mistaken idea, that if you had an oscillator and ch anged something (say L or C) (could be mass or torque constant too.) that i t would take Q cycles for the change to be seen. But about five years ago when getting data from a real oscillator I came to understand that the chan ge is visible right away. It sorta changed my perspective. I was just sha ring in case there was someone else who had my same 'mistaken' idea.

George H.

The response of a system with constant coefficients to a continuous forcing function is speed limited at about Q/f0. Parametric changes show up right away, because they change the coefficients of the differential equation.

Step changes in V_C or I_L cause immediate phase and amplitude shifts, but require delta functions of either current or voltage.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

That kind of "thing you know that ain't so" haunts all of us. Posting here is a cheap way finding such misconceptions.

--
Bill Sloman, Sydney

I'm not sure about this. All stable oscillators have phase memory, else they'd have phase reversals at zero crossings. When the C voltage in an RC or LC oscillator is at its midrange, how else does the oscillator know if it's on the upswing or downswing?

In the case of a '555 RC oscillator, the Schmitt trigger/latch is the phase memory (OK, its only one bit of memory, but it's fully static!). In an LC oscillator, the flux in the inductor is the 'memory'. In both cases, there are two degrees of freedom, so you can map both of them (C charge and latch state for the RC, C charge and L current for the LC) and draw a little gyrator vector.

The instantaneous phase of the voltage and current in the inductor. Think about it--the differential equation for an undamped harmonic oscillator is

d**2 V/dt**2 = -V/LC

These are all instantaneous quantities--where's the memory? Past values of the voltage and current don't appear in the equation.

Phasors are a simple way of visualizing the Fourier decomposition of a sinusoidal or nearly sinusoidal signal, and that smuggles in the idea of linearity and time invariance, but what we're talking about is not a time invariant circuit.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Phil Hobbs wrote:

Yes, of course you are right. Hajimiri describes it very elegantly in Oscillator Phase Noise: A Tutorial,

He shows the impulse responses of LC tank in Fig. 4, and examples of impulse sensitivity function (ISF) for (a) LC oscillator and (b) ring oscillator in Fig. 5.

The ISF for a Colpitts is sinusoidal and only goes to zero at the peaks.

I show in the LTspice file for a multivibrator (appended at the end) that the ISF is similar to a ring oscillator. It has spikes at the transitions and is zero elsewhere.

In theory, the lc oscillator should be sensitive to noise over most of the cycle, and the RC oscillator is only sensitive at the transitions.

So why is an RC oscillator so sensitive in a pll, and often goes into limit cycle oscillations, where a simple Colpitts cures the problem?

That only works for the MC4044/4046. The pullup section goes into saturation since it lacks a simple Baker clamp to prevent it. With the series diode and resistor feeding an op amp, like the circuit JL and I were discussing, biasing the loop off zero will increase the amplitude of the pulses into the op amp and potentially cause internal saturation. That circuit needs to remain on time. It is much better to add a ripple filter at the input to the op amp in the event it doesn't like fast pulses

Here's the ISF noise test for a multivibrator. Please correct the wrap at the end.

Version 4 SHEET 1 1260 800 WIRE -160 -96 -208 -96 WIRE -112 -96 -160 -96 WIRE 64 -96 -112 -96 WIRE 416 -96 144 -96 WIRE 512 -96 416 -96 WIRE 608 -96 512 -96 WIRE 640 -96 608 -96 WIRE 640 -80 640 -96 WIRE -208 -48 -208 -96 WIRE -112 -48 -112 -96 WIRE 416 -48 416 -96 WIRE 512 -48 512 -96 WIRE 640 16 640 0 WIRE -208 48 -208 32 WIRE -160 48 -208 48 WIRE 80 48 -160 48 WIRE 416 48 416 32 WIRE 464 48 416 48 WIRE 704 48 464 48 WIRE -144 80 -400 80 WIRE -112 80 -112 32 WIRE -112 80 -144 80 WIRE 16 80 -112 80 WIRE 80 80 80 48 WIRE 480 80 208 80 WIRE 512 80 512 32 WIRE 512 80 480 80 WIRE 640 80 512 80 WIRE 704 80 704 48 WIRE -400 96 -400 80 WIRE -208 96 -208 48 WIRE -208 96 -320 96 WIRE 16 96 16 80 WIRE 208 96 208 80 WIRE 416 96 416 48 WIRE 416 96 288 96 WIRE 640 96 640 80 WIRE -320 112 -320 96 WIRE 288 112 288 96 WIRE -208 160 -208 96 WIRE -112 160 -112 80 WIRE 416 160 416 96 WIRE 512 160 512 80 WIRE -400 208 -400 160 WIRE -320 208 -320 192 WIRE -320 208 -400 208 WIRE -288 208 -320 208 WIRE -272 208 -288 208 WIRE -16 208 -48 208 WIRE 16 208 16 176 WIRE 16 208 -16 208 WIRE 80 208 80 144 WIRE 80 208 16 208 WIRE 208 208 208 160 WIRE 288 208 288 192 WIRE 288 208 208 208 WIRE 336 208 288 208 WIRE 352 208 336 208 WIRE 608 208 576 208 WIRE 640 208 640 176 WIRE 640 208 608 208 WIRE 704 208 704 144 WIRE 704 208 640 208 WIRE -208 272 -208 256 WIRE -112 272 -112 256 WIRE 416 272 416 256 WIRE 512 272 512 256 FLAG 608 -96 VCC FLAG 640 16 0 FLAG -288 208 Q1B FLAG -144 80 Q2C FLAG -16 208 Q2B FLAG -208 272 0 FLAG -112 272 0 FLAG -160 48 Q1C FLAG 336 208 Q3B FLAG 480 80 Q4C FLAG 608 208 Q4B FLAG 416 272 0 FLAG 512 272 0 FLAG 464 48 Q3C FLAG -160 -96 Noise SYMBOL Voltage 640 -96 R0 WINDOW 3 28 84 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value +5V SYMATTR InstName V1 SYMBOL Npn -48 160 M0 WINDOW 3 57 81 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q2 SYMBOL Npn -272 160 R0 WINDOW 3 51 58 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q1 SYMBOL res -224 -64 R0 SYMATTR InstName R1 SYMATTR Value 1k SYMBOL res -128 -64 R0 SYMATTR InstName R2 SYMATTR Value 1.05k SYMBOL res 32 192 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R3 SYMATTR Value 20k SYMBOL res -336 208 M180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R4 SYMATTR Value 20k SYMBOL cap 64 80 R0 SYMATTR InstName C1 SYMATTR Value 100pf SYMBOL cap -416 96 R0 SYMATTR InstName C2 SYMATTR Value 100pf SYMBOL Npn 576 160 M0 WINDOW 3 57 79 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q4 SYMBOL Npn 352 160 R0 WINDOW 3 49 60 Left 2 SYMATTR Value 2N3904 SYMATTR InstName Q3 SYMBOL res 400 -64 R0 SYMATTR InstName R5 SYMATTR Value 1k SYMBOL res 496 -64 R0 SYMATTR InstName R6 SYMATTR Value 1.05k SYMBOL res 656 192 R180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R7 SYMATTR Value 20k SYMBOL res 272 208 M180 WINDOW 0 36 76 Left 2 WINDOW 3 36 40 Left 2 SYMATTR InstName R8 SYMATTR Value 20k SYMBOL cap 688 80 R0 SYMATTR InstName C3 SYMATTR Value 100pf SYMBOL cap 192 96 R0 SYMATTR InstName C4 SYMATTR Value 100pf SYMBOL Voltage 160 -96 R90 WINDOW 3 45 -112 VRight 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 WINDOW 0 -48 40 VRight 2 SYMATTR Value PULSE(0 2 6.6us 1n 1n 100n 0 1) SYMATTR InstName V2 TEXT 16 -248 Left 2 ;'Multivibrator Step Change In Phase TEXT 24 -224 Left 2 !.tran 20u TEXT 136 -224 Left 2 !.options plotwinsize=0 TEXT -80 -176 Left 2 ;Set V2 Delay to 6.5uS or 6.7uS to see no change in phase

[Transient Analysis] { Npanes: 2 { traces: 3 {589827,0,"V(noise)"} {589826,0,"V(q1b)"} {589829,0,"V (q3b)"} X: ('µ',0,0,2e-006,2e-005) Y[0]: (' ',1,-3.2,0.8,5.6) Y[1]: ('_',0,1e+308,0,-1e+308) Volts: (' ',0,0,0,-3.2,0.8,5.6) Log: 0 0 0 GridStyle: 1 }, { traces: 2 {524292,0,"V(q1c)"} {524294,0,"V(q3c)"} X: ('µ',0,0,2e-006,2e-005) Y[0]: (' ',1,0,0.5,5) Y[1]: ('_',0,1e+308,0,-1e+308) Volts: (' ',0,0,1,0,0.5,5) Log: 0 0 0 GridStyle: 1 } }

John Larkin wrote:

I don't believe we are talking about the same thing. You cannot use a

74HC74 in AC analysis.

An XOR has terrible triangle ripple voltage to the VCO. This causes huge sidebands at the clock frequency.

Speaking about sidebands, here is a updated version of the PLL Analysis using Bessel and Gaussian LC filters at the input to the op amp instead of RC filters.

These reduce the amplitude of the spikes almost an order of magnitude. This could be extended to more sections to reduce the sidebands further.

Incidentally, I had very poor luck with the free AADE filter design program. I don't know how, but the filter values were way off. The free Elsie filter program was much easier to use and gave the correct values. If you are interested the url is

JK

Here's the LC filter version. As usual, watch for the wrap at the end.

Version 4 SHEET 1 4472 800 WIRE -176 96 -224 96 WIRE -16 96 -176 96 WIRE 96 96 48 96 WIRE 96 112 96 96 WIRE -576 176 -592 176 WIRE -512 176 -576 176 WIRE -416 192 -448 192 WIRE -384 192 -416 192 WIRE -176 192 -176 96 WIRE -160 192 -176 192 WIRE -688 208 -736 208 WIRE -512 208 -688 208 WIRE -32 208 -48 208 WIRE 176 208 160 208 WIRE 208 208 176 208 WIRE 304 208 288 208 WIRE 336 208 304 208 WIRE 480 208 464 208 WIRE 576 208 560 208 WIRE -160 224 -176 224 WIRE 176 224 176 208 WIRE 304 224 304 208 WIRE -592 256 -592 176 WIRE -304 272 -304 256 WIRE -32 272 -32 208 WIRE -32 272 -304 272 WIRE 464 272 464 208 WIRE 496 272 464 272 WIRE 576 272 576 208 WIRE 576 272 560 272 WIRE 592 272 576 272 WIRE 672 272 656 272 WIRE 688 272 672 272 WIRE 720 272 688 272 WIRE 176 304 176 288 WIRE 304 304 304 288 WIRE 464 336 464 272 WIRE 512 336 464 336 WIRE 544 336 512 336 WIRE -592 352 -592 336 WIRE -416 352 -416 336 WIRE 672 352 672 272 WIRE 672 352 608 352 WIRE 544 368 512 368 WIRE 96 384 96 192 WIRE 128 384 96 384 WIRE 176 384 160 384 WIRE 208 384 176 384 WIRE 304 384 288 384 WIRE 336 384 304 384 WIRE 176 400 176 384 WIRE 304 400 304 384 WIRE 512 400 512 368 WIRE -176 416 -176 224 WIRE -176 416 -224 416 WIRE 96 416 96 384 WIRE 176 480 176 464 WIRE 304 480 304 464 WIRE -560 496 -592 496 WIRE -512 496 -560 496 WIRE 512 496 512 480 WIRE -416 512 -448 512 WIRE -384 512 -416 512 WIRE -16 512 -224 512 WIRE 96 512 96 496 WIRE 96 512 48 512 WIRE -736 528 -736 208 WIRE -512 528 -736 528 WIRE -736 576 -736 528 WIRE -592 576 -592 496 WIRE 128 576 128 384 WIRE 176 576 128 576 WIRE 208 576 176 576 WIRE 304 576 288 576 WIRE 336 576 304 576 WIRE 464 576 464 336 WIRE 464 576 416 576 WIRE -304 592 -304 576 WIRE -192 592 -304 592 WIRE -32 592 -32 272 WIRE -32 592 -192 592 WIRE 176 592 176 576 WIRE 304 592 304 576 WIRE -736 672 -736 656 WIRE -592 672 -592 656 WIRE 176 672 176 656 WIRE 304 672 304 656 FLAG -416 256 VCC FLAG -304 32 VCC FLAG -384 96 VCC FLAG -304 352 VCC FLAG -384 416 VCC FLAG -416 512 VCO FLAG -416 192 DATA FLAG -192 592 CLR FLAG -416 352 0 FLAG 176 480 0 FLAG 304 480 0 FLAG -592 672 0 FLAG -736 672 0 FLAG -592 352 0 FLAG -688 208 Ramp FLAG -576 176 Dly FLAG -560 496 Ref FLAG -480 224 0 FLAG -480 160 VCC FLAG -480 544 0 FLAG -480 480 VCC FLAG 512 336 Vin FLAG 688 272 VDC FLAG 512 496 0 FLAG 576 384 0 FLAG 576 320 VCC FLAG 176 304 0 FLAG 304 304 0 FLAG 176 672 0 FLAG 304 672 0 FLAG 160 208 0 FLAG 160 384 0 SYMBOL Voltage -416 240 R0 WINDOW 3 28 84 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value +5V SYMATTR InstName V1 SYMBOL 74hc74 -304 32 R0 WINDOW 40 20 14 Left 2 SYMATTR SpiceLine2 SPEED=1 SYMATTR InstName U1 SYMATTR SpiceLine VCC=5 TRIPDT=1e-9 SYMBOL 74hc74 -304 352 R0 WINDOW 40 -45 260 Left 2 SYMATTR SpiceLine2 SPEED=1 SYMATTR InstName U2 SYMATTR SpiceLine VCC=5 DELAY=0.1 TRIPDT=1e-9 SYMBOL digital\\74hc00 -112 144 R0 WINDOW 40 -42 152 Left 2 SYMATTR SpiceLine2 SPEED=1 SYMATTR InstName U3 SYMATTR SpiceLine VCC=5 TRIPDT=1e-9 SYMBOL res 112 96 M0 SYMATTR InstName UP SYMATTR Value 1k SYMBOL res 112 400 M0 SYMATTR InstName DOWN SYMATTR Value 1k SYMBOL cap 160 400 R0 SYMATTR InstName C1 SYMATTR Value 36pf SYMBOL cap 288 400 R0 SYMATTR InstName C2 SYMATTR Value 5.6pf SYMBOL cap 560 256 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C3 SYMATTR Value 10n SYMBOL cap 656 256 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C4 SYMATTR Value 1n SYMBOL res 576 192 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R5 SYMATTR Value 1k SYMBOL voltage -592 560 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value 2V SYMBOL voltage -736 560 R0 WINDOW 3 4 149 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(0 4 500n 1u 0 0 1u) SYMATTR InstName V3 SYMBOL voltage -592 240 R0 WINDOW 3 -102 146 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value PULSE(1.8 2.2 0 50u 1ns 1n) SYMATTR InstName V4 SYMBOL opamps\\1pole -480 192 R0 SYMATTR InstName U5 SYMATTR Value2 Avol=1Meg GBW=1e9 Slew=1e9 SYMBOL opamps\\1pole -480 512 R0 SYMATTR InstName U6 SYMATTR Value2 Avol=1Meg GBW=1e9 Slew=1e9 SYMBOL Voltage 512 384 R0 WINDOW 3 28 84 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR Value 2.5V SYMATTR InstName V5 SYMBOL opamps\\1pole 576 352 R0 WINDOW 123 -92 181 Left 2 SYMATTR Value2 Avol=1Meg GBW=1e6 Slew=1e6 SYMATTR InstName U4 SYMATTR SpiceLine2 en=0 enk=0 in=0 ink=0 Rin=10Meg SYMBOL diode 48 496 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName D1 SYMBOL diode -16 112 R270 WINDOW 0 32 32 VTop 2 WINDOW 3 0 32 VBottom 2 SYMATTR InstName D2 SYMBOL ind 192 400 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L1 SYMATTR Value 15µh SYMBOL cap 160 224 R0 SYMATTR InstName C5 SYMATTR Value 360pf SYMBOL cap 288 224 R0 SYMATTR InstName C6 SYMATTR Value 56pf SYMBOL res 432 368 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 1k SYMBOL ind 192 224 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L2 SYMATTR Value 150µH SYMBOL res 432 192 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value 1k SYMBOL cap 160 592 R0 SYMATTR InstName C7 SYMATTR Value 36pf SYMBOL cap 288 592 R0 SYMATTR InstName C8 SYMATTR Value 4.3pf SYMBOL ind 192 592 R270 WINDOW 0 32 56 VTop 2 WINDOW 3 5 56 VBottom 2 SYMATTR InstName L3 SYMATTR Value 13µH SYMBOL res 432 560 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R3 SYMATTR Value 1k TEXT -376 -64 Left 2 ;'PFD and Op Amp Analysis Bessel and Gaussian Filters TEXT -376 -40 Left 2 !.tran 0 50u 0 TEXT 160 32 Left 2 !.include 74hc.lib TEXT 160 -56 Left 2 !.options plotwinsize=0 TEXT 144 128 Left 2 ;NOTE1: Filter values are for illustrative purposes only TEXT 160 -32 Left 2 !.options nomarch TEXT 160 8 Left 2 !.ic V(VDC) = 2.5 TEXT 160 -8 Left 2 !.ic V(Vin) = 2.5 TEXT 224 256 Left 2 ;1MHz TEXT 208 432 Left 2 ;10MHz TEXT 208 624 Left 2 ;10MHz TEXT 216 664 Left 2 ;Gaussian TEXT 224 464 Left 2 ;Bessel TEXT 216 304 Left 2 ;Bessel TEXT 144 152 Left 2 ;NOTE2: Ground unused filters

[Transient Analysis] { Npanes: 3 Active Pane: 2 { traces: 1 {524293,0,"V(vin)"} X: ('µ',0,0,5e-006,5e-005) Y[0]: (' ',3,2.486,0.002,2.512) Y[1]: ('m',1,1e+308,0.0003,-1e+308) Volts: (' ',0,0,3,2.486,0.002,2.512) Log: 0 0 0 GridStyle: 1 }, { traces: 2 {34603010,0,"I(Up)"} {34603011,0,"I(Down)"} X: ('µ',0,0,5e-006,5e-005) Y[0]: ('m',1,-0.0016,0.0004,0.0024) Y[1]: ('m',0,1e+308,0.01,-1e+308) Amps: ('m',0,0,1,-0.0016,0.0004,0.0024) Log: 0 0 0 GridStyle: 1 }, { traces: 1 {589828,0,"V(vdc)"} X: ('µ',0,0,5e-006,5e-005) Y[0]: (' ',1,1.5,0.1,2.6) Y[1]: ('m',0,1e+308,0.01,-1e+308) Volts: (' ',0,0,1,1.5,0.1,2.6) Log: 0 0 0 GridStyle: 1 } }

Just fore reference, here's some examples

Popodi Pulsed Oscillators 1966

Davies Pulse Triggered Oscillator

Davies Start Stop Colpitts 1968

Gercekci Start Stop with Fixed Phase 1981

Lewis Switch DRO On and Off 1986

Graham High Speed Switched Oscillator 1986

Doty Fixed Starting Phase 1988

Buhler Instant Startup Oscillator IBM 1989

Greene Gated LC Oscillator 1990

Haight Zero Phase Startup 1990

Buhler looks like it may give the best control over the startup amplitude.

JK

Thanks for the nice paper! I'm very far from an expert. But I'll guess th e LC Colpitts is better because it has a much higher Q. RC oscillators are like the ring oscillator.. (low Q) as the Authors say it's like mud compar ed to the ringing in a fine wine glass. (LC oscillator.)

George H. (I really like the idea of timing the added energy so it only adds amplitud e noise and little phase noise.)

What's an MC4046? The so called 4046 is CMOS. There is/was an MC14046 (Motorola) CMOS PLL.

The MC4044 is gold-doped TTL... gold doping kills lifetime.

[snip] ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| San Tan Valley, AZ 85142   Skype: Contacts Only  |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
I love to cook with wine.     Sometimes I even put it in the food.

Well, the multivibrator and ring oscillator are wide bandwidth, so noise arrives at sensitive nodes with full amplitude.

The LC oscillator is narrow bandwidth, so theoretically the noise energy gets filtered and only the energy in the noise bandwidth can influence the cycle.

Hajimiri takes this into account when discussing thermal noise (EQ 11), but he uses an impulse in Fig. 3 when discussing the impulse sensitivity function (ISF). In fact, he uses a lossless tank, which implies infinite Q. According to intuition, the tank phase should not shift, but his figures clearly show it does.

So we need some more insight into what is happening.

Is there any reason why your news client double spaces?

JK

Only if there is a huge divisor in the loop. A 1:1 PLL will have the ripple at 2F, which won't make sidebands.

But design the loop filter so that doesn't happen.

--

John Larkin         Highland Technology, Inc 

jlarkin at highlandtechnology dot com 
http://www.highlandtechnology.com 

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom laser drivers and controllers 
Photonics and fiberoptic TTL data links 
VME thermocouple, LVDT, synchro   acquisition and simulation

The triangle wave is at 1F.

If you are measuring phase noise, you need the spurs down as far as possible, ideally below the noise. You will not get there with an XOR.

JK

John K wrote:

noise

energy

sensitivity

infinite

I finally broke down and put Hajimiri's thesis to the test. It doesn't work.

So much for peer review.

JK

Version 4 SHEET 1 1260 800 WIRE -144 -80 -208 -80 WIRE -64 -80 -144 -80 WIRE 304 -80 240 -80 WIRE 384 -80 304 -80 WIRE -64 -64 -64 -80 WIRE 384 -16 384 -80 WIRE -208 16 -208 0 WIRE -64 16 -64 0 WIRE 0 16 -64 16 WIRE 32 16 0 16 WIRE 240 16 240 0 WIRE -208 32 -208 16 WIRE -64 32 -64 16 WIRE 240 32 240 16 WIRE -208 128 -208 112 WIRE -64 128 -64 112 WIRE 112 128 112 112 WIRE 240 128 240 112 WIRE 384 128 384 48 FLAG 0 16 Noise FLAG -64 128 0 FLAG -208 128 0 FLAG -208 16 L1R1 FLAG -144 -80 L1C1 FLAG 384 128 0 FLAG 240 128 0 FLAG 240 16 L2R2 FLAG 304 -80 L2C2 FLAG 112 128 0 FLAG 112 32 0V SYMBOL Voltage -64 16 R0 WINDOW 3 206 144 Right 2 WINDOW 123 0 0 Left 2 WINDOW 0 50 34 Right 2 SYMATTR Value PULSE(0 2 6.2us 1n 1n 100n 0 1) SYMATTR SpiceLine Rser=1u SYMATTR InstName V1 SYMBOL ind -192 -96 M0 SYMATTR InstName L1 SYMATTR Value 22µH SYMATTR SpiceLine Rser=1u SYMBOL res -224 16 R0 SYMATTR InstName R1 SYMATTR Value 0.5 SYMBOL cap -48 -64 M0 SYMATTR InstName C1 SYMATTR Value 8.84nF SYMATTR SpiceLine Rser=1u SYMBOL ind 256 -96 M0 SYMATTR InstName L2 SYMATTR Value 22µH SYMATTR SpiceLine Rser=1u SYMBOL res 224 16 R0 SYMATTR InstName R2 SYMATTR Value 0.5 SYMBOL cap 400 -16 M0 SYMATTR InstName C2 SYMATTR Value 8.84nF SYMATTR SpiceLine Rser=1u SYMBOL res 96 16 R0 SYMATTR InstName R3 SYMATTR Value 1k TEXT 16 -248 Left 2 ;'LCR Step Change In Phase TEXT 24 -224 Left 2 !.tran 20u TEXT -80 -176 Left 2 ;Set V1 Delay to 6.2us, 6.6us or 7.6us to see changes in zero crossing TEXT 336 -224 Left 2 !.ic V(L1R1) = 10 TEXT 336 -208 Left 2 !.ic V(L2R2) = 10 TEXT -80 -152 Left 2 ;Set R1, R2 to 0.5 Ohms for Q = 100 TEXT -80 -128 Left 2 ;Set R1, R2 to 1.25 Ohms for Q = 40

[Transient Analysis] { Npanes: 2 Active Pane: 1 { traces: 1 {589827,0,"V(noise)"} X: ('µ',0,0,2e-006,2e-005) Y[0]: (' ',1,-0.2,0.2,2.2) Y[1]: ('_',0,1e+308,0,-1e+308) Volts: (' ',0,0,1,-0.2,0.2,2.2) Log: 0 0 0 GridStyle: 1 }, { traces: 3 {524290,0,"V(l1c1)"} {524292,0,"V(l2c2)"} {524293,0,"V (0v)"} X: ('µ',0,0,2e-006,2e-005) Y[0]: (' ',0,-10,2,10) Y[1]: ('_',0,1e+308,0,-1e+308) Volts: (' ',0,0,0,-10,2,10) Log: 0 0 0 GridStyle: 1 } }

Sorry, you are correct. The XOR analysis on page 2 shows the output frequency is twice the reference:

Even with 2F ripple, you are still not going to get the spurious sidebands down very far.

Also, the zero phase error location is not as easy to verify. The signals are 90 degrees out of phase and zero error is sensitive to the duty cycle of the input signals. The PFD aligns the leading edges so you can tell when you are at zero error.

The XOR needs helper circuitry to find and maintain lock. Usually this is a PFD.

JK

What sidebands will you get if you have an oscillator at F and you FM modulate it at 2F? Besides, in most cases a loop filter will take out 2F and a lot of other noise. And most VCOs have an inhernet lowpass filter in the varicap coupling circuit.

Depends on the acquisition range. A narrow-pull VCXO might be happy with an XOR phase detector. One example would be a 10 MHz VCXO with +-100 PPM pull range, which is +-1 KHz. An xor detector and a lowpass of a couple of KHz would work fine. The 2F component from the xor, 20 MHz, would be attenuated to nil by a 2 KHz loop filter.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation 
Picosecond-resolution Digital Delay and Pulse generators 
Custom timing and laser controllers 
Photonics and fiberoptic TTL data links 
VME  analog, thermocouple, LVDT, synchro, tachometer 
Multichannel arbitrary waveform generators