Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

about quality factor Q

Status
Not open for further replies.

Fractional-N

Full Member level 1
Joined
Oct 15, 2007
Messages
97
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,288
Activity points
2,063
i know the resonance frequency is a frequency that imaginary part of impedance is zero at that freaquency.

my question is:
consider a general resonance circuit( i mean, not just a simple parallel RLC);
does the impedance has a peak(max) at resonance freq?? and (if yes) why it must has a max at that freq? how you relate the "imaginary must be zoro" condition to "impedance is maximum" condition?

3 questions :D
thank you
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,977
Helped
14,147
Reputation
28,553
Reaction score
12,834
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,292
In a series resonace circuit, the impedance is minimal at resonance. So as the general case, you have a local maximum or minimum of impedance.

Basically your adding susceptance (parallel resonance) respectively reactance (series resonance) amounts of opposite direction, they sum to zero in the resonance case, leaving the loss related real part of admittance respectively impedance.
 

    Fractional-N

    points: 2
    Helpful Answer Positive Rating

insrusingh

Junior Member level 1
Joined
Jul 8, 2008
Messages
19
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,378
In high speed digital circuit frankly speaking componnts having low Q factor has no place.
 

LvW

Advanced Member level 5
Joined
May 7, 2008
Messages
5,835
Helped
1,740
Reputation
3,484
Reaction score
1,340
Trophy points
1,393
Location
Germany
Activity points
39,237
Fractional-N said:
i know the resonance frequency is a frequency that imaginary part of impedance is zero at that freaquency.
my question is:
consider a general resonance circuit( i mean, not just a simple parallel RLC);
does the impedance has a peak(max) at resonance freq?? and (if yes) why it must has a max at that freq? how you relate the "imaginary must be zoro" condition to "impedance is maximum" condition?
You are asking for a "general resonance circuit" - not just "simple parallel RLC".
Well, the situation is not as easy as you probably think.
Sometimes its really surprising.
Example: Take C=100uF in series with 10 ohms and take L=100uH in series with 10 ohms. Then put both series connections in parallel. Now you have a LC parallel circuitry - both elements with some losses.
As a result - the impedance of the circuit exhibits a MINIMUM at some kHz. At this frequency the imaginary part and the phase shift is zero.
 

    Fractional-N

    points: 2
    Helpful Answer Positive Rating

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,977
Helped
14,147
Reputation
28,553
Reaction score
12,834
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,292
The example in fact shows that "general resonance circuit" is a dubious term, however I wouldn't regard a circuit Q of 0.05 as resonance, normally.
 

LvW

Advanced Member level 5
Joined
May 7, 2008
Messages
5,835
Helped
1,740
Reputation
3,484
Reaction score
1,340
Trophy points
1,393
Location
Germany
Activity points
39,237
FvM said:
The example in fact shows that "general resonance circuit" is a dubious term, however I wouldn't regard a circuit Q of 0.05 as resonance, normally.
Is there any lower Q limit for defining the state of resonance ?
 

Kral

Advanced Member level 4
Joined
Mar 28, 2005
Messages
1,326
Helped
280
Reputation
558
Reaction score
85
Trophy points
1,328
Location
USA
Activity points
13,418
Fractional-N,
The resonant frequency is the frequency at which the voltage and current are in phase. This is not necessarily the frequency at which the maximum impedance occurs. At the resonant frequency, the imaginary component of the impedance will be zero, but the magnitude of the impedance will not necessarily be maximum. A network consisting of an ideal inductor in parallel with an ideal capacitor has its peak impedance magnitude at its resonant frequency, but this condition is not true in most cases. Of course with a series resonant circuit, the only maxima are at frequencies of zero and infinity. In the circuit proposed by LvW, there is indeed a minimum that occurs at a frequency well below the resonant frequency and the frequency of maximum impedance.
Regards,
Kral
 

LvW

Advanced Member level 5
Joined
May 7, 2008
Messages
5,835
Helped
1,740
Reputation
3,484
Reaction score
1,340
Trophy points
1,393
Location
Germany
Activity points
39,237
Kral said:
The resonant frequency is the frequency at which the voltage and current are in phase. This is not necessarily the frequency at which the maximum impedance occurs. At the resonant frequency, the imaginary component of the impedance will be zero, but the magnitude of the impedance will not necessarily be maximum.
..............................................
In the circuit proposed by LvW, there is indeed a minimum that occurs at a frequency well below the resonant frequency and the frequency of maximum impedance.
Regards,
Kral
It is very easy to show that this minimum is not "well below the resonant frequency", instead it is identical to the resonant frequency as this is the only frequency with a real impedance.
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,977
Helped
14,147
Reputation
28,553
Reaction score
12,834
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,292
Is there any lower Q limit for defining the state of resonance ?
I think the term could be regarded ridiculous for a 0.05 Q. If you define resonance as cancellation of imaginary impedance component, it would be correct, but not meaningful in a technical sense. I prefer to reserve it for systems with less than aperiodical damping.
 

RF-OM

Advanced Member level 2
Joined
Aug 3, 2007
Messages
584
Helped
123
Reputation
246
Reaction score
17
Trophy points
1,298
Location
The Earth
Activity points
5,751
In general any circuit that has reactance can resonate at some frequencies. The resonance condition “sum of reactance is equal to zero” can occur in different part of electrical network and at different frequencies. When reactive part of impedance goes to minimum it does not mean that the total impedance goes to maximum. For parallel resonance (anti-resonance) impedance will be equal to the equivalent parallel resistance and will be maximal whereas for series resonance equivalent impedance will be equal to series resistance and will be minimal. There is variety of resonant circuits. The most simple of them are classified as resonant tanks of the 1st, 2nd and 3rd kind in electrical network theory. Even these simple circuits can have more than one resonance in the range of interest. Very often in RF and microwave analyses at least three resonances are considered and analyzed.
Regarding Q about 0.05 it is against the Q-factor definition. In this case we have power loss about 20 times higher than the total energy stored per one period. Definitely this is impossible.
 

Kral

Advanced Member level 4
Joined
Mar 28, 2005
Messages
1,326
Helped
280
Reputation
558
Reaction score
85
Trophy points
1,328
Location
USA
Activity points
13,418
My calculations show a minimum Z of approximately 7.5-j2.6 at a frequency of 57.3 KHz. This corresponds to a phase lag of approximately 19 degrees. I'll double check my calculations when I get a chance. In your reply, there is some undecipherable text “…is identical…”. Perhaps this is due to my browser.
Regards,
Kral

Added after 2 minutes:

This is in regards to FVM's post at 14:41
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
47,977
Helped
14,147
Reputation
28,553
Reaction score
12,834
Trophy points
1,393
Location
Bochum, Germany
Activity points
278,292
This is in regards to FVM's post at 14:41
I'm not aware of.

Regarding Q about 0.05 it is against the Q-factor definition. In this case we have power loss about 20 times higher than the total energy stored per one period. Definitely this is impossible.
I applied the Q definition usual in electronics Q = |Im(Z)|/|Re(Z)| to the proposed circuit (that has 1 ohm reactance and 20 ohm resistance):
Example: Take C=100uF in series with 10 ohms and take L=100uH in series with 10 ohms. Then put both series connections in parallel. Now you have a LC parallel circuitry - both elements with some losses.
The problem is, that the circuit doesn't store energy for one period, cause it isn't oscillating at all. Thus I refused to regard it a resonant circuit. For circuits with usual Q values, both definitions have similar results.
 

LvW

Advanced Member level 5
Joined
May 7, 2008
Messages
5,835
Helped
1,740
Reputation
3,484
Reaction score
1,340
Trophy points
1,393
Location
Germany
Activity points
39,237
Kral said:
My calculations show a minimum Z of approximately 7.5-j2.6 at a frequency of 57.3 KHz. This corresponds to a phase lag of approximately 19 degrees.
Perhaps there is an error in your calculation ?
My results: Minimum of Z at Fo=1.593 kHz with Zmin=(L+R1*R2*C)/(R1+R2)*C=5.05 ohms and Zmax(w=0 and w=infinite)=10 ohms.

To FvM and RF-OM:
I agree that the classical definition of Q for the case under discussion is not meaningful. However, why should we call the state with IMG(Z)=0 not "resonance"?

Another remark: Don´t blindly trust any simulation program!!! I have simulated the circuit as described in my first reply with a current input of 1 A - and, therefore, the created voltage should be identical to the impedance. As a result, for ac analysis as well as for tran analysis a phase shift of 180 deg was displayed for the "resonant" frequency, which simply is not true. (This would be identical to a negative real resistance).
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top