Update: Varistor/MOV Spice Modeling

Jim Thompson

unread,

May 10, 2015, 11:46:12 AM5/10/15

to

John Larkin

unread,

May 11, 2015, 1:34:30 PM5/11/15

to

On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>Varistor/MOV Spice Modeling has been updated.
>
>See VaristorMOV.zip on the Device Models & Subcircuits Page of my
>website.
>
> ...Jim Thompson

Encrypted! Not very general.

Given how imprecise MOVs are, seems like a couple of zeners and
resistors, and maybe a capacitor, would be plenty good enough.

MOVs "wear out", too. Did you include that?

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

Jim Thompson

unread,

May 11, 2015, 4:23:57 PM5/11/15

to

On Mon, 11 May 2015 10:34:26 -0700, John Larkin
<jla...@highlandtechnology.com> wrote:

>On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
><To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>
>>Varistor/MOV Spice Modeling has been updated.
>>
>>See VaristorMOV.zip on the Device Models & Subcircuits Page of my
>>website.
>>
>> ...Jim Thompson
>
>Encrypted!

To conceal my techniques. Did you read... 8 lines versus 34 ??

>Not very general.

Parameterizable to any MOV... read!

>
>Given how imprecise MOVs are, seems like a couple of zeners and
>resistors, and maybe a capacitor, would be plenty good enough.
>

Good enough for you, perhaps >:-}

>MOVs "wear out", too. Did you include that?

Typical Larkin sand-in-air... you can't help but be an asshole.

John Larkin

unread,

May 11, 2015, 5:55:25 PM5/11/15

to

On Mon, 11 May 2015 13:23:52 -0700, Jim Thompson

<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Mon, 11 May 2015 10:34:26 -0700, John Larkin
><jla...@highlandtechnology.com> wrote:
>
>>On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
>><To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>>
>>>Varistor/MOV Spice Modeling has been updated.
>>>
>>>See VaristorMOV.zip on the Device Models & Subcircuits Page of my
>>>website.
>>>
>>> ...Jim Thompson
>>
>>Encrypted!
>
>To conceal my techniques. Did you read... 8 lines versus 34 ??
>
>>Not very general.
>
>Parameterizable to any MOV... read!
>
>>
>>Given how imprecise MOVs are, seems like a couple of zeners and
>>resistors, and maybe a capacitor, would be plenty good enough.
>>
>
>Good enough for you, perhaps >:-}

Good enough for anybody. MOVs are not precise parts.

>
>>MOVs "wear out", too. Did you include that?
>
>Typical Larkin sand-in-air... you can't help but be an asshole.

Did I hurt your feelings by not admiring your encrypted MOV model?

I really can't imagine why anyone needs to Spice a MOV. The things
that matter (power dissipation, joule capacity, wearout) aren't in the
model.

Tim Williams

unread,

May 11, 2015, 6:07:39 PM5/11/15

to

"Jim Thompson" <To-Email-Use-Th...@On-My-Web-Site.com> wrote
in message news:dp32lat5ilmiqir9k...@4ax.com...

> To conceal my techniques. Did you read... 8 lines versus 34 ??

8 lines? Looks more like 20 to me.

Tim

--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Contract Design
Website: http://seventransistorlabs.com

Jim Thompson

unread,

May 11, 2015, 7:03:06 PM5/11/15

to

On Mon, 11 May 2015 17:12:03 -0500, "Tim Williams"
<tiw...@seventransistorlabs.com> wrote:

>"Jim Thompson" <To-Email-Use-Th...@On-My-Web-Site.com> wrote
>in message news:dp32lat5ilmiqir9k...@4ax.com...
>> To conceal my techniques. Did you read... 8 lines versus 34 ??
>
>8 lines? Looks more like 20 to me.
>
>Tim

That's LTspice's encryption scheme that adds all those extra
characters.

Tim Williams

unread,

May 11, 2015, 7:57:27 PM5/11/15

to

"Jim Thompson" <To-Email-Use-Th...@On-My-Web-Site.com> wrote
in message news:06d2lat69uque78p0...@4ax.com...

>>8 lines? Looks more like 20 to me.
>>
>>Tim
>
> That's LTspice's encryption scheme that adds all those extra
> characters.

Well either way, your claim is absurd, because the file size (not that 2k
files are any concern these days) certainly isn't 8 lines, and without
cracked encryption (anyone know if LTSpice / PSpice / HSpice encryption
has been cracked? anyone working on it?), impossible to verify.

There's also the possibility that fewer lines contain more math, resulting
in worse execution time or convergence anyway.

But really, I'm guessing it's your tanh fit stuff, which you've posted
about publicly before. You also claim it's based on publically published
curves. So hiding it is all the more absurd...

Jim Thompson

unread,

May 11, 2015, 8:02:10 PM5/11/15

to

On Mon, 11 May 2015 19:01:40 -0500, "Tim Williams"
<tiw...@seventransistorlabs.com> wrote:

>"Jim Thompson" <To-Email-Use-Th...@On-My-Web-Site.com> wrote
>in message news:06d2lat69uque78p0...@4ax.com...
>>>8 lines? Looks more like 20 to me.
>>>
>>>Tim
>>
>> That's LTspice's encryption scheme that adds all those extra
>> characters.
>
>Well either way, your claim is absurd, because the file size (not that 2k
>files are any concern these days) certainly isn't 8 lines, and without
>cracked encryption (anyone know if LTSpice / PSpice / HSpice encryption
>has been cracked? anyone working on it?), impossible to verify.
>
>There's also the possibility that fewer lines contain more math, resulting
>in worse execution time or convergence anyway.
>
>But really, I'm guessing it's your tanh fit stuff, which you've posted
>about publicly before. You also claim it's based on publically published
>curves. So hiding it is all the more absurd...
>
>Tim

New technique. It definitely works. Try it or not, I don't care. It
was originally posted to the LTspice List where the question was
broached initially.

Those with interest in circuit work should subscribe to the LTspice
List... you'll learn some new techniques... and not be annoyed by
cockroach invasions into technical threads.

Phil Hobbs

unread,

May 11, 2015, 8:58:43 PM5/11/15

to

On 5/11/2015 4:23 PM, Jim Thompson wrote:
> On Mon, 11 May 2015 10:34:26 -0700, John Larkin
> <jla...@highlandtechnology.com> wrote:
>
>> On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
>> <To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>>
>>> Varistor/MOV Spice Modeling has been updated.
>>>
>>> See VaristorMOV.zip on the Device Models & Subcircuits Page of
>>> my website. ...Jim Thompson
>>
>> Encrypted!
>
> To conceal my techniques.

Seems pretty pointless at your age, Jim. Are you ever going to pass on
all those pearls of wisdom, or are you going to take them to the grave
like King Tut?

Soon would be good, if you're ever going to.

Cheers

Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

Jim Thompson

unread,

May 11, 2015, 9:04:12 PM5/11/15

to

On Mon, 11 May 2015 20:58:40 -0400, Phil Hobbs
<pcdhSpamM...@electrooptical.net> wrote:

>On 5/11/2015 4:23 PM, Jim Thompson wrote:
>> On Mon, 11 May 2015 10:34:26 -0700, John Larkin
>> <jla...@highlandtechnology.com> wrote:
>>
>>> On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
>>> <To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>>>
>>>> Varistor/MOV Spice Modeling has been updated.
>>>>
>>>> See VaristorMOV.zip on the Device Models & Subcircuits Page of
>>>> my website. ...Jim Thompson
>>>
>>> Encrypted!
>>
>> To conceal my techniques.
>
>Seems pretty pointless at your age, Jim. Are you ever going to pass on
>all those pearls of wisdom, or are you going to take them to the grave
>like King Tut?
>
>Soon would be good, if you're ever going to.
>
>Cheers
>
>Phil Hobbs

Right now I'm making money at it. Why should I give away the store?
Does Larkin post a _complete_ schematic of his instruments and
manufacturing techniques here?

Phil Hobbs

unread,

May 11, 2015, 9:17:46 PM5/11/15

to

On 5/11/2015 9:04 PM, Jim Thompson wrote:
> On Mon, 11 May 2015 20:58:40 -0400, Phil Hobbs
> <pcdhSpamM...@electrooptical.net> wrote:
>
>> On 5/11/2015 4:23 PM, Jim Thompson wrote:
>>> On Mon, 11 May 2015 10:34:26 -0700, John Larkin
>>> <jla...@highlandtechnology.com> wrote:
>>>
>>>> On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
>>>> <To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>>>>
>>>>> Varistor/MOV Spice Modeling has been updated.
>>>>>
>>>>> See VaristorMOV.zip on the Device Models & Subcircuits Page of
>>>>> my website. ...Jim Thompson
>>>>
>>>> Encrypted!
>>>
>>> To conceal my techniques.
>>
>> Seems pretty pointless at your age, Jim. Are you ever going to pass on
>> all those pearls of wisdom, or are you going to take them to the grave
>> like King Tut?
>>
>> Soon would be good, if you're ever going to.
>

> Right now I'm making money at it. Why should I give away the store?
> Does Larkin post a _complete_ schematic of his instruments and
> manufacturing techniques here?

Because you're in your mid 70s. I know your dad lived into his early
90s, and was pretty sharp till the end, but there's no guarantee that
you'll do the same. What do you want your legacy to be? It's pretty
late not to be thinking about that, especially given the amount of time
and energy it will take to pass on the best of what you know in a form
accessible to young people.

John L.'s company will probably be manufacturing stuff long after he's
retired, and he isn't the sole designer, so that isn't the same case at all.

Also you vastly overestimate the opportunity cost of publishing
stuff--it's usually a net win.

My book gives away the majority of what I know about what I do, and
besides helping a lot of young folks (which was the point of the
exercise), it has been the foundation of my consulting practice. The
first edition was published when I was 40.

John Larkin

unread,

May 11, 2015, 9:43:02 PM5/11/15

to

On Mon, 11 May 2015 18:04:05 -0700, Jim Thompson

<To-Email-Use-Th...@On-My-Web-Site.com> wrote:

>On Mon, 11 May 2015 20:58:40 -0400, Phil Hobbs
><pcdhSpamM...@electrooptical.net> wrote:
>
>>On 5/11/2015 4:23 PM, Jim Thompson wrote:
>>> On Mon, 11 May 2015 10:34:26 -0700, John Larkin
>>> <jla...@highlandtechnology.com> wrote:
>>>
>>>> On Sun, 10 May 2015 08:46:07 -0700, Jim Thompson
>>>> <To-Email-Use-Th...@On-My-Web-Site.com> wrote:
>>>>
>>>>> Varistor/MOV Spice Modeling has been updated.
>>>>>
>>>>> See VaristorMOV.zip on the Device Models & Subcircuits Page of
>>>>> my website. ...Jim Thompson
>>>>
>>>> Encrypted!
>>>
>>> To conceal my techniques.
>>
>>Seems pretty pointless at your age, Jim. Are you ever going to pass on
>>all those pearls of wisdom, or are you going to take them to the grave
>>like King Tut?
>>
>>Soon would be good, if you're ever going to.
>>
>>Cheers
>>
>>Phil Hobbs
>
>Right now I'm making money at it. Why should I give away the store?

An MOV model is "the store"? That's sort of sad.

>Does Larkin post a _complete_ schematic of his instruments and
>manufacturing techniques here?
>
> ...Jim Thompson

Sometimes I post actual product schematics, usually sheets relevant to
the topic, and photos of real boxes and boards. Sometimes I post
(non-encrypted!) Spice files. Sometimes I post ideas and scribbles
that I just imagined.

I don't post manufacturing techniques, because my production people do
that, and it's fairly standard stuff. I have posted a lot of
breadboarding techniques, stuff that I do myself.

I don't post encrypted Spice circuits and then tell people that I'm
smart, just because I say so.

I still don't know why your thing would be any better than a couple of
zeners and resistors, and you won't say why.

--

John Larkin Highland Technology, Inc

picosecond timing laser drivers and controllers

Jim Thompson

unread,

May 11, 2015, 9:51:10 PM5/11/15

to

OK. I'll divulge all my "secrets":

(1) All derivatives must exist and be continuous
(2) All functions must be BOUNDED smoothly, per (1) above.

That's all there is to it other than knowing circuit design at the
device level which you can pick up easily with 50 or so years or
experience >:-}

Seriously.

More seriously, leukemia testing begins tomorrow.

I'll consider your advice... maybe it is time to do a load dump.

Phil Hobbs

unread,

May 11, 2015, 9:56:36 PM5/11/15

to

I'm praying for good test results for you.

Jim Thompson

unread,

May 11, 2015, 11:48:25 PM5/11/15

to

On Mon, 11 May 2015 18:42:16 -0700, John Larkin

GFYU

John Larkin

unread,

May 12, 2015, 12:01:17 AM5/12/15

to

On Mon, 11 May 2015 20:48:20 -0700, Jim Thompson

After you get well, you're welcome to try.

Tim Williams

unread,

May 12, 2015, 5:55:31 AM5/12/15

to

"Phil Hobbs" <pcdhSpamM...@electrooptical.net> wrote in message
news:55515040...@electrooptical.net...

> Seems pretty pointless at your age, Jim. Are you ever going to pass on
> all those pearls of wisdom, or are you going to take them to the grave
> like King Tut?

On a related note then, if anyone has something they'd like to model, I
could take a crack at it.

I suspect Jim isn't doing anything I don't know, and if he sees some
examples, maybe he'd even admit what it is, too ;-)

Ralph Barone

unread,

May 12, 2015, 9:03:19 AM5/12/15

to

"Tim Williams" <tiw...@seventransistorlabs.com> wrote:
> "Phil Hobbs" <pcdhSpamM...@electrooptical.net> wrote in message
> news:55515040...@electrooptical.net...
>> Seems pretty pointless at your age, Jim. Are you ever going to pass on
>> all those pearls of wisdom, or are you going to take them to the grave
>> like King Tut?
>
> On a related note then, if anyone has something they'd like to model, I
> could take a crack at it.
>
> I suspect Jim isn't doing anything I don't know, and if he sees some
> examples, maybe he'd even admit what it is, too ;-)
>
> Tim

Perhaps not the right place in the thread to say this, but what problems
would come up if you modelled a varistor as a behavioural current source
with I = k* V^alpha in series with a bulk resistance? I've had pretty good
luck with that model (just not in SPICE).

Jim Thompson

unread,

May 12, 2015, 10:12:53 AM5/12/15

to

On Tue, 12 May 2015 04:59:44 -0500, "Tim Williams"
<tiw...@seventransistorlabs.com> wrote:

>"Phil Hobbs" <pcdhSpamM...@electrooptical.net> wrote in message
>news:55515040...@electrooptical.net...
>> Seems pretty pointless at your age, Jim. Are you ever going to pass on
>> all those pearls of wisdom, or are you going to take them to the grave
>> like King Tut?
>
>On a related note then, if anyone has something they'd like to model, I
>could take a crack at it.
>
>I suspect Jim isn't doing anything I don't know,

Bwahahahahahaha! I needed a laugh right now.

>and if he sees some
>examples, maybe he'd even admit what it is, too ;-)
>
>Tim

How about an NCP1072? That was requested on the LTspice List >:-}

John Larkin

unread,

May 12, 2015, 10:19:57 AM5/12/15

to

How about a resistor, diode, voltage source?

A couple in parallel if you want some curvature.

Bridge rectifier if you need bipolar.

John Larkin

unread,

May 12, 2015, 3:55:45 PM5/12/15

to

On Tue, 12 May 2015 13:03:16 GMT, Ralph Barone
<addre...@invalid.invalid> wrote:

The LT Spice "g" componant is a voltage dependant current source. The
relationship could be your equation, or a polynomial, or anything.
That would be *one* line of text. Add capacitance or ESL as needed.

--

John Larkin Highland Technology, Inc

picosecond timing precision measurement

Jim Thompson

unread,

May 12, 2015, 4:03:09 PM5/12/15

to

On Tue, 12 May 2015 12:55:37 -0700, John Larkin
<jla...@highlandtechnology.com> wrote:

>On Tue, 12 May 2015 13:03:16 GMT, Ralph Barone
><addre...@invalid.invalid> wrote:
>
>>"Tim Williams" <tiw...@seventransistorlabs.com> wrote:
>>> "Phil Hobbs" <pcdhSpamM...@electrooptical.net> wrote in message
>>> news:55515040...@electrooptical.net...
>>>> Seems pretty pointless at your age, Jim. Are you ever going to pass on
>>>> all those pearls of wisdom, or are you going to take them to the grave
>>>> like King Tut?
>>>
>>> On a related note then, if anyone has something they'd like to model, I
>>> could take a crack at it.
>>>
>>> I suspect Jim isn't doing anything I don't know, and if he sees some
>>> examples, maybe he'd even admit what it is, too ;-)
>>>
>>> Tim
>>
>>Perhaps not the right place in the thread to say this, but what problems
>>would come up if you modelled a varistor as a behavioural current source
>>with I = k* V^alpha in series with a bulk resistance? I've had pretty good
>>luck with that model (just not in SPICE).
>
>The LT Spice "g" componant is a voltage dependant current source. The
>relationship could be your equation, or a polynomial, or anything.
>That would be *one* line of text. Add capacitance or ESL as needed.

Yep. The MOV equation is one line of GVALUE... with coefficients set
by VTEST, ITEST, VCLAMP, ICLAMP, Rseries, Lseries and Cpar, right from
the data sheet.

But apparently too difficult for Larkin the psychopathic asshole to
master plugging in the numbers.

John Larkin

unread,

May 12, 2015, 4:40:56 PM5/12/15

to

One line? What line? This one maybe?

1e0b22458c9e013c316e1a17d2fb21f94017955ce88f51ef5c03aab498e88eb5e71f71e1875841719b8947fe3d6f920cb7af1fedfab169c9c17d1a3c4a3fa116

Jim Thompson

unread,

May 12, 2015, 4:44:52 PM5/12/15

to

On Tue, 12 May 2015 13:40:48 -0700, John Larkin

You will NEVER see the equation, as simple as it is, NEVER >:-}

John Larkin

unread,

May 12, 2015, 5:09:24 PM5/12/15

to

On Tue, 12 May 2015 13:44:46 -0700, Jim Thompson

Oh no, no, I'm heartbroken.

Jim Thompson

unread,

May 12, 2015, 5:41:32 PM5/12/15

to

On Tue, 12 May 2015 14:09:16 -0700, John Larkin

Good >:-}

Tim Williams

unread,

May 12, 2015, 7:49:44 PM5/12/15

to

"Jim Thompson" <To-Email-Use-Th...@On-My-Web-Site.com> wrote
in message news:pg24lati560f9obr5...@4ax.com...

>>and if he sees some
>>examples, maybe he'd even admit what it is, too ;-)
>>
>>Tim
>
> How about an NCP1072? That was requested on the LTspice List >:-}

http://www.onsemi.com/PowerSolutions/supportDoc.do?type=models&rpn=NCP1072

"Document ID" says it's an "ISpice Model", but extension says it's a
.DWG.... WTF?

Of course, an IC is something quite a bit more involved than a passive
two-terminal. It would take time, especially if no good resources are
available. I wouldn't do that for free (nor would I guess you, either!).

Jim Thompson

unread,

May 12, 2015, 7:57:06 PM5/12/15

to

On Tue, 12 May 2015 18:54:48 -0500, "Tim Williams"
<tiw...@seventransistorlabs.com> wrote:

>"Jim Thompson" <To-Email-Use-Th...@On-My-Web-Site.com> wrote
>in message news:pg24lati560f9obr5...@4ax.com...
>>>and if he sees some
>>>examples, maybe he'd even admit what it is, too ;-)
>>>
>>>Tim
>>
>> How about an NCP1072? That was requested on the LTspice List >:-}
>
>http://www.onsemi.com/PowerSolutions/supportDoc.do?type=models&rpn=NCP1072
>
>"Document ID" says it's an "ISpice Model", but extension says it's a
>.DWG.... WTF?

That link is certainly an AutoCAD Drawing (.DWG)

>
>Of course, an IC is something quite a bit more involved than a passive
>two-terminal. It would take time, especially if no good resources are
>available. I wouldn't do that for free (nor would I guess you, either!).

A typical I/C (or hybrid, as I've been doing lately) usually takes me
3-6 weeks.

>
>Tim

There _is_ an ISpice model of the NCP1072 rummaging around, but it
apparently won't run on anything but ISpice. I haven't tried to
decipher its syntax into Berkeley-compliant... yet.

Tim Williams

unread,

May 12, 2015, 8:12:12 PM5/12/15

to

"Ralph Barone" <addre...@invalid.invalid> wrote in message
news:338871053453128354.323302address_is-invalid.invalid@shawnews...

> Perhaps not the right place in the thread to say this, but what problems
> would come up if you modelled a varistor as a behavioural current source
> with I = k* V^alpha in series with a bulk resistance? I've had pretty
> good
> luck with that model (just not in SPICE).

I've seen that advertised before in appnotes, but it doesn't make sense.
There's no physical reason to have a power law in a semiconducting device,
and it suggests way more leakage current than there should be (the
breakdown region might be softer than an exponential, but there still has
to be an exponential leakage tail in there). Perhaps it's just a crude
fit to the random breakdown-ESR pattern. In much the same way as 1/f
noise (another odd order power law) can be modeled as a stacked chain of
noise sources of random bandwidth.

SPICE won't appreciate it, because a negative number to a random power is
likely to result in some random complex number. You'd at least need
abs(V) to start, then put the sign back on the current later. Most of the
derivatives all disappear at V=0, which doesn't help.

A symmetrical exponential function, like tanh, would probably do a good
job, though being a bit too sharp. As John says, connecting some
resistors in series with that, then cleaning it up with a few more of
different threshold voltages and ESRs, would do; but some may balk at this
solution using "too many lines". :)

A single-line rendering of that isn't actually possible, because a
"resistor in series with an exponential" is a transcendental equation, and
has to be solved iteratively by the SPICE engine. Your alternative would
be to build a "dulled" tanh function (say, toning down the exponential
asymptotes by taking the sqrt or something), but that is also impossible,
because the only thing that's "dull" enough to tame an exponential is a
log (any polynomial or power law just becomes a constant factor to the
exponent). But that simply undoes the exponent entirely, giving flat
asymptotes; and doesn't work for negative values (see
http://www.wolframalpha.com/input/?i=ln%28tanh%28x%29%29 ).

Probably, best would be to sit down with a spreadsheet and plug in curves
until it works. There's always boring old polynomials, which are probably
quite a good idea in this case -- with the right combination of (complex)
poles and zeroes, the function can be odd (= gives opposite current for
negative argument) and the asymptotes can be linear or quadratic (linear
would make sense in that it's the minimum ESR when all semiconducting
grains are conducting). With some tweaking, perhaps a non-geometric
polynomial could be built that exhibits realistic leakage current, and
approximates the V^alpha asymptote.

One can also make polynomials from other polynomials (Chebyshev and other
named orthogonal polynomial series are typically better for building
curve-fits than just throwing coefficients at a geometric series), or from
other functions (e.g., the periodic polynomials in cos^n(phi) and such,
useful for harmonic analysis).

Ralph Barone

unread,

May 12, 2015, 10:20:23 PM5/12/15

to

"Tim Williams" <tiw...@seventransistorlabs.com> wrote:
> "Ralph Barone" <addre...@invalid.invalid> wrote in message
> news:338871053453128354.323302address_is-invalid.invalid@shawnews...
>> Perhaps not the right place in the thread to say this, but what problems
>> would come up if you modelled a varistor as a behavioural current source
>> with I = k* V^alpha in series with a bulk resistance? I've had pretty
>> good
>> luck with that model (just not in SPICE).
>
> I've seen that advertised before in appnotes, but it doesn't make sense.
> There's no physical reason to have a power law in a semiconducting device,
> and it suggests way more leakage current than there should be (the
> breakdown region might be softer than an exponential, but there still has
> to be an exponential leakage tail in there). Perhaps it's just a crude
> fit to the random breakdown-ESR pattern. In much the same way as 1/f
> noise (another odd order power law) can be modeled as a stacked chain of
> noise sources of random bandwidth.

I can't say whether it makes physical sense, but in the limited research
I've done into the characteristics of metal oxide varistors, the I = k*
V^alpha equation keeps popping up. In the curve matching that I have done
(in Excel, mind you), a bit of series resistance kept the curve from
flattening out too much at high applied voltages, and also made sense in
terms of the bulk resistance of the material (disregarding the grain
boundaries where all the non-linear magic occurs).

> SPICE won't appreciate it, because a negative number to a random power is
> likely to result in some random complex number. You'd at least need
> abs(V) to start, then put the sign back on the current later. Most of the
> derivatives all disappear at V=0, which doesn't help.

Definitely, it would need a bit of a wrapper around it (sign() and abs()
functions) to make it symmetrical. I tend to post from my phone, so
tenseness is rewarded (at least while I'm typing).

hemanth...@gmail.com

unread,

Jul 25, 2018, 3:07:37 AM7/25/18

to

On Sunday, May 10, 2015 at 9:16:12 PM UTC+5:30, Jim Thompson wrote:
> Varistor/MOV Spice Modeling has been updated.
>
> See VaristorMOV.zip on the Device Models & Subcircuits Page of my
> website.
>