Modelling inductor 'Q' in LTSpice

[Paul Burridge]

Hi,

Say I have a 10uH inductor with a 'Q' of 120, how would I represent
this in LT's inductor properties? There isn't a box for entering Q.
Does one have to arrive at an arbitrary equivalent and stick it in DC
resistance or something?

Thanks,

p
--

The BBC: Licensed at public expense to spread lies.

 

[Mike Engelhardt]

Paul,

Say I have a 10uH inductor with a 'Q' of 120, how would I represent
this in LT's inductor properties? There isn't a box for entering Q.
Does one have to arrive at an arbitrary equivalent and stick it in DC
resistance or something?

You add a parallel resistance to limit inductor Q. Q is the ratio of
resistance and reactance, so the resistance value depends on the
frequency that is Q is specified.

Rpar = 2*pi*freq*L*Q

You can also limit the Q with a series Resistance by this formula:
Rpar = 2*pi*freq*L/Q

But for inductors, it's almost always better to use the parallel
loss because real inductors' Q rolls off with with frequency. You
can get somewhat better accuracy by using a combination of
series and parallel losses.

--Mike

 

[Ian Buckner]

"Mike Engelhardt" <pmte@concentric.net> wrote in message
news:c3kevj$qo9@dispatch.concentric.net...

Paul,

Say I have a 10uH inductor with a 'Q' of 120, how would I
represent
this in LT's inductor properties? There isn't a box for entering
Q.
Does one have to arrive at an arbitrary equivalent and stick it in
DC
resistance or something?

You add a parallel resistance to limit inductor Q. Q is the ratio
of
resistance and reactance, so the resistance value depends on the
frequency that is Q is specified.

Rpar = 2*pi*freq*L*Q

You can also limit the Q with a series Resistance by this formula:
Rpar = 2*pi*freq*L/Q

But for inductors, it's almost always better to use the parallel
loss because real inductors' Q rolls off with with frequency. You
can get somewhat better accuracy by using a combination of
series and parallel losses.

--Mike


Go to http://www.coi1craft.com/models.cfm and pick a suitable

family of inductors. Click on the Spice Model button which will
take you to data sheets with the Spice model, parameters for the
various parts, and information on the valid frequency range.

If you have or can measure the parameters for your particular
part, modify the model values to suit.

Regards
Ian

 

[Ken Smith]

In article <1079951540.187684@cswreg.cos.agilent.com>,
Ian Buckner <Ian_Buckner@agilent.com> wrote:
[...]

If you have or can measure the parameters for your particular
part, modify the model values to suit.

You don't actually have to measure the parameters. Often you have a graph
of Q vs frequency. All you have to do is put the right L, Rs, and Rp
values into spice to get that Q curve and your model will be good enough
in most applications.

--
--
kensmith@rahul.net forging knowledge

 

[Paul Burridge]

On 21 Mar 2004 11:13:39 EST, "Mike Engelhardt" <pmte@concentric.net>
wrote:

You add a parallel resistance to limit inductor Q. Q is the ratio of
resistance and reactance, so the resistance value depends on the
frequency that is Q is specified.

Rpar = 2*pi*freq*L*Q

You can also limit the Q with a series Resistance by this formula:
Rpar = 2*pi*freq*L/Q

But for inductors, it's almost always better to use the parallel
loss because real inductors' Q rolls off with with frequency. You
can get somewhat better accuracy by using a combination of
series and parallel losses.

Thanks, Mike. I'd seen the notation on others' schematics and often
wondered if it was factoring for Q.
--

The BBC: Licensed at public expense to spread lies.