Maker Pro
Maker Pro

V and I not in phase at resonance Frequency in RLC network?

D

Diego Stutzer

Jan 1, 1970
0
Hi,
Well, I'm really confused.
I simulate a simple serial R-C-L-Network (all in series).

As far as I know the total (input-)Impedance of the network is:
Z = R + jwL - j/(wC) resp. the resonance frequency (where Zin=R) is
1/sqrt(L*C).
At resonance frequency, the Impedance should be real and therefore in
my hummel opinion Voltage and Current schould be in phase.

The funny thing is, when i build up such a network in Schematics
(Cadence PSD 14.1/Orcad 9.2) and simulate it with the PSpice A/D
Simulator, the current is displaced (relative to the voltage) about
lambda/4 - obviously not in phase!?

Can anyone tell my where I made a mistake?
Or why this Problem is showing up?
Thanks to anyone reading this and especially to those who post
answers.
D. Stutzer
 
D

Don Pearce

Jan 1, 1970
0
Hi,
Well, I'm really confused.
I simulate a simple serial R-C-L-Network (all in series).

As far as I know the total (input-)Impedance of the network is:
Z = R + jwL - j/(wC) resp. the resonance frequency (where Zin=R) is
1/sqrt(L*C).
At resonance frequency, the Impedance should be real and therefore in
my hummel opinion Voltage and Current schould be in phase.

The funny thing is, when i build up such a network in Schematics
(Cadence PSD 14.1/Orcad 9.2) and simulate it with the PSpice A/D
Simulator, the current is displaced (relative to the voltage) about
lambda/4 - obviously not in phase!?

Can anyone tell my where I made a mistake?
Or why this Problem is showing up?
Thanks to anyone reading this and especially to those who post
answers.
D. Stutzer

The impedance should be R + jwL + 1/(jwC)

d

_____________________________

http://www.pearce.uk.com
 
J

John Fields

Jan 1, 1970
0
Hi,
Well, I'm really confused.
I simulate a simple serial R-C-L-Network (all in series).

As far as I know the total (input-)Impedance of the network is:
Z = R + jwL - j/(wC) resp. the resonance frequency (where Zin=R) is
1/sqrt(L*C).
At resonance frequency, the Impedance should be real and therefore in
my hummel opinion Voltage and Current schould be in phase.

The funny thing is, when i build up such a network in Schematics
(Cadence PSD 14.1/Orcad 9.2) and simulate it with the PSpice A/D
Simulator, the current is displaced (relative to the voltage) about
lambda/4 - obviously not in phase!?

Can anyone tell my where I made a mistake?
 
D

Diego Stutzer

Jan 1, 1970
0
John Fields said:
---

1
f = -------------
2pi(sqrt(LC))

WOW!! Thank you soo much. The world is sound and the laws of physics apply
again. I just confused w (omega) and f.
Best Regards
D. Stutzer
 
M

Michael

Jan 1, 1970
0
The impedance should be R + jwL + 1/(jwC)
You sure?, how do the j parts cancel at resonance if they are both added?
 
T

Tom Bruhns

Jan 1, 1970
0
Which voltage and which current? Clearly the voltage across the
capacitor will always be pi/2 relative to the current in that
capacitor, and the same (but opposite sign) for an inductor. Also,
are you doing an AC or a transient analysis? If transient, are the
transients settled, are you really on the resonance frequency, and are
you simulating with sufficient time resolution? Oh, and I'm not quite
sure what you mean by "lambda/4" phase shift. Is that degrees or
radians, and what's lambda?

Cheers,
Tom
 
R

Reg Edwards

Jan 1, 1970
0
Multiply top and bottom of 1/jwC by j (This does not change its value) and
you get 1/jwC = minus j/wC.

Back to school with your algebra.
 
B

budgie

Jan 1, 1970
0
Multiply top and bottom of 1/jwC by j (This does not change its value) and
you get 1/jwC = minus j/wC.

Back to school with your algebra.

His algebra looks perfectly fine to me. But as others have pointed
out, he's left the 2pi out.
 
R

Reg Edwards

Jan 1, 1970
0
budgie said:
His algebra looks perfectly fine to me. But as others have pointed
out, he's left the 2pi out.

------------------------------------------

Yes. I apologise for my remark about school. I gained the incorrect
impression from the previous replies.

The w in wC stands for omega = 2*Pi*F, the angular frequency.
 
A

Active8

Jan 1, 1970
0
------------------------------------------

Yes. I apologise for my remark about school. I gained the incorrect
impression from the previous replies.

The w in wC stands for omega = 2*Pi*F, the angular frequency.
now that were all done playing with j...

don't forget

Z = sqrt{R^2 + [(wL) - (1/wC)]^2]}

and

Z(s) = R + Ls + 1/Cs

which is just plain easier to deal with 'til you need to journey back
into time domain land. no need to leave it f(t) for this deal, though.

all that j stuff... that was scary shit. so easy to make a mistake.
swapping w and f is another good one. only works for f/f stuff.

mike
 
P

Paul Burridge

Jan 1, 1970
0
now that were all done playing with j...

don't forget

Z = sqrt{R^2 + [(wL) - (1/wC)]^2]}

and

Z(s) = R + Ls + 1/Cs

which is just plain easier to deal with 'til you need to journey back
into time domain land. no need to leave it f(t) for this deal, though.

all that j stuff... that was scary shit. so easy to make a mistake.

Especially so given the limited typography of this particular medium.
I suspect few of us would have a problem if we could only view these
formulae in a suitably appropriate typeface!!!
 
L

Laycock, Christopher

Jan 1, 1970
0
R + jwL + 1/(jwC)
= R + jwL -j/(wC)
so at resonance wL=1/(wC)
ie w=1/sqrt(LC)

Chris
 
J

John Fields

Jan 1, 1970
0
Which voltage and which current? Clearly the voltage across the
capacitor will always be pi/2 relative to the current in that
capacitor, and the same (but opposite sign) for an inductor. Also,
are you doing an AC or a transient analysis? If transient, are the
transients settled, are you really on the resonance frequency, and are
you simulating with sufficient time resolution? Oh, and I'm not quite
sure what you mean by "lambda/4" phase shift. Is that degrees or
radians, and what's lambda?

---
Since he calculated the resonant frequency of the circuit using

f = 1/sqrt(L*C),

his answer will always be a frequency 6.28 times higher than it should
be, so the reactance of the inductor will be greater than the reactance
of the capacitor, making the phase angle positive.

"Lambda" is usually taken to mean wavelength, so "lambda/4" would mean a
quarter wavelength; in this context, 90°, the approximate phase
difference between his simulated voltage and current.
 
A

Active8

Jan 1, 1970
0
now that were all done playing with j...

don't forget

Z = sqrt{R^2 + [(wL) - (1/wC)]^2]}

and

Z(s) = R + Ls + 1/Cs

which is just plain easier to deal with 'til you need to journey back
into time domain land. no need to leave it f(t) for this deal, though.

all that j stuff... that was scary shit. so easy to make a mistake.

Especially so given the limited typography of this particular medium.
I suspect few of us would have a problem if we could only view these
formulae in a suitably appropriate typeface!!!
what pre tell, is a typeface that would make it hard? wing-dings?

i admit, even in HTML with arial or fixed-pitch, where you have sub and
super tags and greek letter codes, you can't do much. i tried a few free
math notation tools for HTML and wasn't happy. plus you need a plugin
for most if not all. W^3C has a standard. i'd like to see more math
capabiliies in browsers. we could attach formulae without violating the
no bianaries rule. of course microshaft and standards will probably
never be seen in the same line of code.

i still don't see why text based math legibility is font dependant. i
see i could have eliminated some clarifying parens by

- -
| 2 1 2 |
sqrt | R + ----- | or a 1/2 power instead of sqrt
| jwC |
- -
created by Andy´s ASCII-Circuit v1.22.310103 Beta www.tech-chat.de

does that also foul up?

brs,
mike
 
F

Fred Abse

Jan 1, 1970
0
WOW!! Thank you soo much. The world is sound and the laws of physics
apply again. I just confused w (omega) and f. Best Regards


Improbability level now zero. Normality restored. Anything you still can't
cope with is your problem.
(Douglas Adams)
 
M

Michael

Jan 1, 1970
0
all that j stuff... that was scary shit. so easy to make a mistake.
swapping w and f is another good one. only works for f/f stuff.

mike

I get a bit scared of all the j's too, I prefer polar. Usually you
want a magnitude and angle in the end result anyway.
 
Top