# capacitor nonlinearity again

Started by May 9, 2013
```Does anybody have an LT Spice example that models the C-V dependence of a
ceramic cap? I was thinking about the parametric frequency divider that I did
with a diode, and it occurs to me that a capacitor's c (probably) falls off
symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
parametric divider is different.

It might be fun to play with that.

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
```
```On 5/9/2013 12:08 AM, John Larkin wrote:
>
> Does anybody have an LT Spice example that models the C-V dependence of a
> ceramic cap? I was thinking about the parametric frequency divider that I did
> with a diode, and it occurs to me that a capacitor's c (probably) falls off
> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
> parametric divider is different.
>
> It might be fun to play with that.
>

Isn't capacitor voltage dependence relatively slow physical effect ?

VLV

```
```Should be pretty easy to take the standard nonlinear inductor model and
swap out integral(V dt) for integral(I dt).

Everything else is the same, including "core" loss.  There should still be
a cutoff frequency where losses take over.  Let's see, eddy current
becomes "eddy voltage", but not "eddy", which implies a parallel loop, it
has to be in series now.  That would be the voltage dropped across the
ESR.  Should be easy to estimate from Z plots, as well as ESL.

Anything you can do with a saturable reactor you can do with a saturable
cap, obviously.  It'll be smaller, but probably comparable loss, given the
low Q of type 2 ceramics.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://seventransistorlabs.com

"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> wrote in
message news:qgbmo810a6hsg31pk3b0hg11hhuq5cndm6@4ax.com...
>
> Does anybody have an LT Spice example that models the C-V dependence of
> a
> ceramic cap? I was thinking about the parametric frequency divider that
> I did
> with a diode, and it occurs to me that a capacitor's c (probably) falls
> off
> symmetrically with +- voltage, but a diode is asymmetric. So the
> capacitor based
> parametric divider is different.
>
> It might be fun to play with that.
>
>
> --
>
> John Larkin                  Highland Technology Inc
> www.highlandtechnology.com   jlarkin at highlandtechnology dot com
>
> Precision electronic instrumentation
> Picosecond-resolution Digital Delay and Pulse generators
> Custom timing and laser controllers
> Photonics and fiberoptic TTL data links
> VME  analog, thermocouple, LVDT, synchro, tachometer
> Multichannel arbitrary waveform generators

```
```On 2013-05-09 07:08, John Larkin wrote:
>
> Does anybody have an LT Spice example that models the C-V dependence of a
> ceramic cap? I was thinking about the parametric frequency divider that I did
> with a diode, and it occurs to me that a capacitor's c (probably) falls off
> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
> parametric divider is different.
>
> It might be fun to play with that.
>
>

Maybe a series connection of a normal capacitor and a behavioral
voltage source. Make the voltage of the BV k times an odd power
of the voltage across the capacitor.

Jeroen Belleman
```
```On Thu, 09 May 2013 00:26:27 -0500, Vladimir Vassilevsky <nospam@nowhere.com>
wrote:

>On 5/9/2013 12:08 AM, John Larkin wrote:
>>
>> Does anybody have an LT Spice example that models the C-V dependence of a
>> ceramic cap? I was thinking about the parametric frequency divider that I did
>> with a diode, and it occurs to me that a capacitor's c (probably) falls off
>> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
>> parametric divider is different.
>>
>> It might be fun to play with that.
>>
>
>Isn't capacitor voltage dependence relatively slow physical effect ?

It's fast enough to cause distortion at audio frequencies, but it does seem to
be very, very messy.

This guy did a lot of testing...

http://www.cliftonlaboratories.com/capacitor_voltage_change.htm

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
```
```On 5/9/2013 8:50 AM, John Larkin wrote:
> On Thu, 09 May 2013 00:26:27 -0500, Vladimir Vassilevsky <nospam@nowhere.com>
> wrote:
>
>> On 5/9/2013 12:08 AM, John Larkin wrote:
>>>
>>> Does anybody have an LT Spice example that models the C-V dependence of a
>>> ceramic cap? I was thinking about the parametric frequency divider that I did
>>> with a diode, and it occurs to me that a capacitor's c (probably) falls off
>>> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
>>> parametric divider is different.
>>>
>>> It might be fun to play with that.
>>>
>>
>> Isn't capacitor voltage dependence relatively slow physical effect ?
>
> It's fast enough to cause distortion at audio frequencies, but it does seem to
> be very, very messy.
>
> This guy did a lot of testing...
>
> http://www.cliftonlaboratories.com/capacitor_voltage_change.htm

This guy assumes capacitor nonlinearity to be static; while it is time
dependent effect. That's why his simulations don't match with his own
measurements.

I'd model capacitor nonlinearity by gyrator with lowpass in control
path. First order approximation would probably be enough for practical
purpose.

DSP and Mixed Signal Designs
www.abvolt.com

```
```On Thu, 09 May 2013 09:15:36 -0500, Vladimir Vassilevsky
<nospam@nowhere.com> wrote:

>On 5/9/2013 8:50 AM, John Larkin wrote:
>> On Thu, 09 May 2013 00:26:27 -0500, Vladimir Vassilevsky <nospam@nowhere.com>
>> wrote:
>>
>>> On 5/9/2013 12:08 AM, John Larkin wrote:
>>>>
>>>> Does anybody have an LT Spice example that models the C-V dependence of a
>>>> ceramic cap? I was thinking about the parametric frequency divider that I did
>>>> with a diode, and it occurs to me that a capacitor's c (probably) falls off
>>>> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
>>>> parametric divider is different.
>>>>
>>>> It might be fun to play with that.
>>>>
>>>
>>> Isn't capacitor voltage dependence relatively slow physical effect ?
>>
>> It's fast enough to cause distortion at audio frequencies, but it does seem to
>> be very, very messy.
>>
>> This guy did a lot of testing...
>>
>> http://www.cliftonlaboratories.com/capacitor_voltage_change.htm
>
>This guy assumes capacitor nonlinearity to be static; while it is time
>dependent effect. That's why his simulations don't match with his own
>measurements.
>
>I'd model capacitor nonlinearity by gyrator with lowpass in control
>path. First order approximation would probably be enough for practical
>purpose.
>
>DSP and Mixed Signal Designs
>www.abvolt.com
>

Or use a table, as in...

http://www.analog-innovations.com/VVC.zip

but, as you suggest, add a low-pass "driving" the table.

You could likewise tabularize the dissipation factor.

Wonder how you would go about characterizing the "frequency response"?

...Jim Thompson
--
| James E.Thompson                                 |    mens     |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

I love to cook with wine.     Sometimes I even put it in the food.
```
```On Thu, 09 May 2013 09:15:36 -0500, Vladimir Vassilevsky <nospam@nowhere.com>
wrote:

>On 5/9/2013 8:50 AM, John Larkin wrote:
>> On Thu, 09 May 2013 00:26:27 -0500, Vladimir Vassilevsky <nospam@nowhere.com>
>> wrote:
>>
>>> On 5/9/2013 12:08 AM, John Larkin wrote:
>>>>
>>>> Does anybody have an LT Spice example that models the C-V dependence of a
>>>> ceramic cap? I was thinking about the parametric frequency divider that I did
>>>> with a diode, and it occurs to me that a capacitor's c (probably) falls off
>>>> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
>>>> parametric divider is different.
>>>>
>>>> It might be fun to play with that.
>>>>
>>>
>>> Isn't capacitor voltage dependence relatively slow physical effect ?
>>
>> It's fast enough to cause distortion at audio frequencies, but it does seem to
>> be very, very messy.
>>
>> This guy did a lot of testing...
>>
>> http://www.cliftonlaboratories.com/capacitor_voltage_change.htm
>
>This guy assumes capacitor nonlinearity to be static; while it is time
>dependent effect. That's why his simulations don't match with his own
>measurements.
>
>I'd model capacitor nonlinearity by gyrator with lowpass in control
>path. First order approximation would probably be enough for practical
>purpose.

It looks messy enough that it's hardly worth modeling, at least to use in a
product. Even if you decide it's safe, the next batch of caps may be worse.

--

John Larkin                  Highland Technology Inc
www.highlandtechnology.com   jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME  analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
```
```John Larkin <jjlarkin@highnotlandthistechnologypart.com> wrote:
> On Thu, 09 May 2013 00:26:27 -0500, Vladimir Vassilevsky <nospam@nowhere.com>
> wrote:
>
>>On 5/9/2013 12:08 AM, John Larkin wrote:
>>>
>>> Does anybody have an LT Spice example that models the C-V dependence of a
>>> ceramic cap? I was thinking about the parametric frequency divider that I did
>>> with a diode, and it occurs to me that a capacitor's c (probably) falls off
>>> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
>>> parametric divider is different.
>>>
>>> It might be fun to play with that.
>>>
>>
>>Isn't capacitor voltage dependence relatively slow physical effect ?
>
> It's fast enough to cause distortion at audio frequencies, but it does seem to
> be very, very messy.
>
> This guy did a lot of testing...
>
> http://www.cliftonlaboratories.com/capacitor_voltage_change.htm

very nice page there.

mylar wins!

I should send that guy some polypropylene, polycarbonate and polystyrene
caps to test.

```
```On Thu, 09 May 2013 14:32:21 +0200, Jeroen <jeroen@nospam.please>
wrote:

>On 2013-05-09 07:08, John Larkin wrote:
>>
>> Does anybody have an LT Spice example that models the C-V dependence of a
>> ceramic cap? I was thinking about the parametric frequency divider that I did
>> with a diode, and it occurs to me that a capacitor's c (probably) falls off
>> symmetrically with +- voltage, but a diode is asymmetric. So the capacitor based
>> parametric divider is different.
>>
>> It might be fun to play with that.
>>
>>
>
>Maybe a series connection of a normal capacitor and a behavioral
>voltage source. Make the voltage of the BV k times an odd power
>of the voltage across the capacitor.
>
>Jeroen Belleman

G source, output in parallel with the capacitor, output current made a
function of both the current thru the capacitor and the voltage across
it.

...Jim Thompson
--
| James E.Thompson                                 |    mens     |