how to measure Q value?

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
ezpcb
Posts: 98
Joined: Tue Apr 19, 2005 1:01 am
Contact:

how to measure Q value?

Post by ezpcb »

Can any one tell me how to measure Q value of a capacitor, inductor or a R-L-C network? thanks!
http://www.EzPCB.com
High Quality PCB for Electronics Hobbists, Pay for Chrokee, Get Land Rover
rshayes
Posts: 1286
Joined: Tue Mar 04, 2003 1:01 am
Contact:

Re: how to measure Q value?

Post by rshayes »

If the part has moderate to high Q, the usual method is to add an inductor or capacitor to form a resonant circuit and then measure either the bandwidth or the increase in voltage due to the resonance. The Boonton or Hewlett-Packard Q meters made this type of measurement and would indicate Q directly.<p>For low Q parts, it is usually easier to use a bridge to maesure the resistive and reactive parts of the impedance and calculate the Q. For example, measuring the inductance of a wire-wound resistor. The Boonton 250 R-X Meter was well suited for this.
User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: how to measure Q value?

Post by Bob Scott »

I'm assuming that the L and C and R are wired in parallel.
Q is "Quality".
D=1/Q where D is the damping factor. (This sentence has been edited due to errors)<p>(Paragaph removed due to errors.)<p>To put it another way, the R in your circuit is probably installed there a a Damping resistor. The less resistance, the more damping.<p>Xc of the capacitor drops with increasing frequency. Xl rises with increasing frequency.<p>Xc=1/(2*pi*f*C)
Xl=2*pi*f*L<p>A parallel LC circuit resonates at a frequency where Xl and Xc are equal. (Xl is inductive reactance and Xc is capacitive reactance.)<p>CORRECT ME IF I'M WRONG but I think that if you are using ideal components, and at the resonant frequency, the Q is equal to 1 when a resistor of the value of Xl or Xc is placed in parallel with L and C. <p>Q is also related to bandwidth.
Q=f/BW where f is the resonant frequency and BW is the -3Db bandwidth. This means that the AC voltage amplitude drops on both sides of the resonant frequency and the points where the amplitude drops to 0.707 of the measured resonant voltage mark the bandwidth limits. Subtract the lower -3DB bandwidth frequency limit from the upper -3Db bandwitdth frequency limit. This gives you the value of BW. Then the value of Q is f/BW.<p>Estimate the resonant frequency by using the formula f=1/(2*pi*SQURRT(LC)) where f is in Hz, L is in Henries, and C is in Farads. Someone correct me if I'm wrong. This is from memory.<p>Then you can drive the parallel circuit with a high impedance sine wave generator and while sweeping the circuit, use an oscilloscope to find the resonant frequency and the -3Db frequency values.<p>By "high impedance sine wave generator" I mean use a resistor in series with the generator that is at least 10 or more times the value of the R in the LRC circuit.<p>Bob :cool: <p>[ October 13, 2005: Message edited by: Bob Scott ]<p>[ October 13, 2005: Message edited by: Bob Scott ]<p>[ October 13, 2005: Message edited by: Bob Scott ]</p>
-=VA7KOR=- My solar system includes Pluto.
Robert Reed
Posts: 2277
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: how to measure Q value?

Post by Robert Reed »

EZPB-<p>In LC circuits, 'Q' is directly related to resistance. It does not matter how it is positioned in the circuit, or if it is parellel or series LC configuration. Theoretically, with absolutely no resistance present, one could inject a single hi-speed pulse into a tank circuit and it would ring (resonate) forever. Of course in the real world, this is not possible as resistance is present everywhere. This resistance can be in parellel or in series with the LC components. and even internal to the L or C itself--no matter it still eats up energy and this is what starts lowering 'Q' from an otherwise infinate value. The more resistance, the less 'Q' and vice versa. 'Q's of 50 to 100 are considered good. For most of us, the cheap dirty coils we wind probably end up at half of those values. Bob made an interesting comment about the word behind 'Q'.
(quality) . Its seems not many people know the word behind the symbol (actually more properly Quality Factor).As to 'Q' measurements vs. sweeping, it might be simpler to note the -3 Db points spread and divide that into the center frequency. Example---Fo=100Mhz ; -3 Db points= 98 Mhhz & 102 mhz ; spread = 4 MHz ; 100 / 4 = 25 (Q). If you need higher 'Q' you can cascade circuits to acheive this with amplifiers inserted to cover losses. Some I.F. struips can attain an overall 'Q' of several thousand this way. If resonant frequencys will enter into your calculations-- Here is one neat formula that I have always used----F(resonance)=159.2 / Sq. root of LC . F in Mhz ; L and C in microunits. This is not as cumbersome as the usual formula for resonance. BTW there was an informative discussion on the subject of 'Q' some posts back but I cannot not recall its title.<p>[ October 09, 2005: Message edited by: ROBERT REED ]</p>
Post Reply

Who is online

Users browsing this forum: Ahrefs [Bot] and 38 guests