SPICE simulation of finite Q inductors

[Joel Kolstad]

Those of you who simulate things like passive LC filters... when you want a
simulation that includes the finite Q effect of the inductor, do you tend to
use a model that computes the equivalent parallel resistance, Rloss, based on
the Q specified on a data sheet at a given frequency (and then use that at
*all* frequencies of your simulation, effectively creating Q(freq) )? Or do
you simply specify Q and then use that at all frequencies (creating
Rloss(freq) )?

For the following, assume you first calculated Rloss at a frequency f0:

The actual Q of an inductor varies with sqrt(freq) until you start approaching
self-resonance, so assuming Q is fixed, you underestimate Q above f0 and
overestimate below f0. Using just Rloss, you overestimate below f0 and
underestimate above f0.

---Joel

 

[Mark]

Those of you who simulate things like passive LC filters... when you want a
simulation that includes the finite Q effect of the inductor, do you tend to
use a model that computes the equivalent parallel resistance, Rloss, based on
the Q specified on a data sheet at a given frequency (and then use that at
*all* frequencies of your simulation, effectively creating Q(freq) )? Or do
you simply specify Q and then use that at all frequencies (creating
Rloss(freq) )?

In most cases, the Q is important only near the resonant frequency. I
use a single fixed value resistor in series with the L that sets the Q
at the frequency corresponding to the circuit resonance frequency.

Mark

 

[Jim Thompson]

In most cases, the Q is important only near the resonant frequency. I
use a single fixed value resistor in series with the L that sets the Q
at the frequency corresponding to the circuit resonance frequency.

Mark

In critical applications I'm inclined to use a multi-lump model to
make sure I catch any "squirrels" ;-)

...Jim Thompson

 

[Michael A. Terrell]

In critical applications I'm inclined to use a multi-lump model to
make sure I catch any "squirrels" ;-)

...Jim Thompson

Too bad you can't make that work on Usenet. We're overrun with
squirrels.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Jim Thompson]

Too bad you can't make that work on Usenet. We're overrun with
squirrels.

I've suddenly acquired jack rabbits in my front yard. Can't figure
out where they're nesting.

...Jim Thompson

 

[Michael A. Terrell]

I've suddenly acquired jack rabbits in my front yard. Can't figure
out where they're nesting.

Jim Thompson

Be VERY careful, Jim. Eeyore sent them to spy on you!


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Jim Thompson]

Be VERY careful, Jim. Eeyore sent them to spy on you!

THAT explains it ;-)

...Jim Thompson

 

[Michael A. Terrell]

THAT explains it ;-)

Jim Thompson

You better oil up your Uzi, he wants them to build a roundabout in
your yard.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Jim Thompson]

You better oil up your Uzi, he wants them to build a roundabout in
your yard.

I think the greeting sign (in Hebrew) out on the front patio will
probably drive him back ;-)

...Jim Thompson

 

[Michael A. Terrell]

I think the greeting sign (in Hebrew) out on the front patio will
probably drive him back ;-)

...Jim Thompson

Do you think that I should tell the little moron that I was trained
to kill with the M16, the M60, and how to use the M72, yet I've never
had to kill anyone?


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[RRogers]

Those of you who simulate things like passive LC filters... when you want a
simulation that includes the finite Q effect of the inductor, do you tend to
use a model that computes the equivalent parallel resistance, Rloss, based on
the Q specified on a data sheet at a given frequency (and then use that at
*all* frequencies of your simulation, effectively creating Q(freq) )? Or do
you simply specify Q and then use that at all frequencies (creating
Rloss(freq) )?

For the following, assume you first calculated Rloss at a frequency f0:

The actual Q of an inductor varies with sqrt(freq) until you start approaching
self-resonance, so assuming Q is fixed, you underestimate Q above f0 and
overestimate below f0. Using just Rloss, you overestimate below f0 and
underestimate above f0.

---Joel

I can't find the original message so I will continue here. The first
order model should include the winding resistance as a series resistor,
and then compute a parrallel resistor due to the core losses. Verify
over the frequencies your interested in, materials vary a lot. In fact
if a series resistance is sufficiently accurate, you can use
pre-distortion (right term?), tilting the frequency axis, and design
the filter having perfect (inf Q) components; and then undistort the
results to get values. That sounds a little jumbled, but these days I
presume that technique is obsolere. I reliably designed a 9 pole
filter in the 70's this way; with both time and frequency requirements.


Ray

 

[Leon]

Those of you who simulate things like passive LC filters... when you want a
simulation that includes the finite Q effect of the inductor, do you tend to
use a model that computes the equivalent parallel resistance, Rloss, based on
the Q specified on a data sheet at a given frequency (and then use that at
*all* frequencies of your simulation, effectively creating Q(freq) )? Or do
you simply specify Q and then use that at all frequencies (creating
Rloss(freq) )?

For the following, assume you first calculated Rloss at a frequency f0:

The actual Q of an inductor varies with sqrt(freq) until you start approaching
self-resonance, so assuming Q is fixed, you underestimate Q above f0 and
overestimate below f0. Using just Rloss, you overestimate below f0 and
underestimate above f0.

---Joel

Coilcraft has some nice SPICE models for their inductors. You could
modify one of those.

Leon

 

[Leon]

Those of you who simulate things like passive LC filters... when you want a
simulation that includes the finite Q effect of the inductor, do you tend to
use a model that computes the equivalent parallel resistance, Rloss, based on
the Q specified on a data sheet at a given frequency (and then use that at
*all* frequencies of your simulation, effectively creating Q(freq) )? Or do
you simply specify Q and then use that at all frequencies (creating
Rloss(freq) )?

For the following, assume you first calculated Rloss at a frequency f0:

The actual Q of an inductor varies with sqrt(freq) until you start approaching
self-resonance, so assuming Q is fixed, you underestimate Q above f0 and
overestimate below f0. Using just Rloss, you overestimate below f0 and
underestimate above f0.

---Joel

Coilcraft has some nice SPICE models for their inductors. You could
modify one of those.

Leon

 

[Charlie Edmondson]

I've suddenly acquired jack rabbits in my front yard. Can't figure
out where they're nesting.

...Jim Thompson

Well, you might as well just accept them, and hope your plants grow
faster than they can eat them...

Charlie

 

[Jim Thompson]

Well, you might as well just accept them, and hope your plants grow
faster than they can eat them...

Charlie

I don't think they eat cactus or mesquite ?

If they do, I'll just get out the shotgun ;-)

...Jim Thompson

 

[Michael A. Terrell]

I don't think they eat cactus or mesquite ?

If they do, I'll just get out the shotgun ;-)

Jim Thompson

If they eat cacti, ship them to Lake County Florida. There is enough
there to feed millions of the willey wabbits down there.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Charlie Edmondson]

I don't think they eat cactus or mesquite ?

If they do, I'll just get out the shotgun ;-)

...Jim Thompson

Young mesquite, probably. Cacti, they usually use to cover their
burrows. The problems are if you have any flowers or food plants
growning. In Borrego, we had several families, but there was only one of
the plants that they liked so well that it couldn't grow. But, once we
planted some marigolds. Lasted two days...

Charlie