Talk:Pierce oscillator

From Wikipedia, the free encyclopedia

[edit] Load capacitance

The average capacitance (C1+C2)/2 is called the "load capacitance". (this article)

  • "In the popular Pierce oscillator circuit, which has a capacitor to ground on either side of the crystal, the load capacitance is equal to the series combination of the two capacitors plus Cstray."[1](James B. Northcutt)
  • "So what load is your Pierce-gate oscillator presenting to the crystal? A simple calculation illustrated with Figure 2 will tell you: Cload = {[Cin+C1][C2+Cout]/[Cin+C1+C2+Cout]} + pcb strays (2~3pF)"[2][3] (Ramon Cerda)
  • "to implement a Pierce oscillator ... The load capacitance is given by: CL = ((C1 x C2) / (C1 + C2)) + Stray"[4](D. Ibarra)
  • Load capacitance is calculated as follows: Pierce circuit CL = (C1 x C2) / (C1 + C2) + Cstray[5](Abracon)
  • The CX crystal calibration tolerance is influenced by the effective circuit capacitances, specified as the load capacitance (CL.) CL is approximately equal to: CL = (C1 x C2) / (C1 + C2) + CS[6](Euroquartz)
  • CA and CB are chosen such that their series combination capacitance equals the load capacitance specified by the manufacturer, ie 20 pF or 32 pF as mentioned.[7](fairchild)
  • It should be noted that the actual loading capacitance to the crystal equals CL1 || CL2 plus the parasitic capacitance of board and the terminals of the inverters.(Maxim)

What is it really, for our purposes? Provide a reference. — Omegatron 19:36, 18 May 2007 (UTC)

Oh wait. I just realized that "series combination" means the same thing as (C1 x C2) / (C1 + C2), which makes these all agree with each other, more or less, and the article is the only thing that disagrees. So the stray capacitance is in parallel and therefore summed with a normal addition sign? — Omegatron 19:50, 29 May 2007 (UTC)

Yes, you are right. "series combination" means the same thing as all those other formulas (except the incorrect formula I originally posted).

The incorrect formula came from a hasty mis-reading of this text:

"The total value of capacitance at the crystal's terminals is (Ca+Cp)/2, where Ca is the actual value of capacitor, per pin, that you place at the OSC1 and OSC2 pins, and Cp is the per pin parasitic capacitance. Cp is usually about 8pF or so. So, if your crystal wants to see a 20pF load, you will need to put 32pF capacitors at both OSC1 and OSC2: (32+8)/2 = 20pF." -- http://techref.massmind.org/techref/clocks.htm

Sorry about that. --76.209.28.72 05:28, 13 June 2007 (UTC)

I'm pretty sure there are 2 "stray capacitances". One stray capacitance from one pin of the crystal to GND (including the gate input capacitance), in parallel with C1. And another stray capacitance from the other crystal pin to GND (including the gate output capacitance). So we have

C_L = (C_1 + C_{s1}) \| (C_2 + C_{s2}) = {1 \over {1 \over C_1 + C_{s1} } + {1 \over C_2 + C_{s2} } }

For the very common case where (approximately) C1 ≈ C2 and C_s1 ≈ C_s2, we have

C_L \approx (C_1 + C_{s1})/2 = {C_1 \over 2} + {C_{s1} \over 2} = {C_1 \over 2} + C_{stray}

Since both C_s1 and C_s2 (and their series combination, C_stray) are so small that they are difficult to measure -- and since the crystal will still oscillate very close to the nominal frequency even if your estimate of that capacitance is off by an order of magnitude -- then the slightly simplified rule-of-thumb equations are more than adequate. --76.209.28.72 05:28, 13 June 2007 (UTC)

[edit] Move

Someone left a comment in the text:

perhaps move some of this text to Crystal_oscillator#Series_or_parallel_resonance ?

— Omegatron 14:56, 4 June 2007 (UTC)

[edit] Series vs parallel

There's no physical difference between the two, is there? They just vibrate at slightly different frequencies depending on the type of circuit they are used in, so they are tuned and specified for a particular type of circuit. — Omegatron 14:21, 6 June 2007 (UTC)

Yes, the physical construction is the same. The only difference between "series" and "parallel" is whether they write the series resonant frequency on it, or they write the frequency it oscillates at in a Pierce oscillator. The crystal oscillator article should make that clear. --76.209.28.72 05:28, 13 June 2007 (UTC)