TI has encrypted Pspice models for many of their parts, but I can't
persuade LT Spice to run the switchers I want to use. Any thoughts on
that?

One of my guys says he can run their Cadence sim, but I'd prefer LT.

Parts are TPS54302, TPS563300, TPS562208. I'd consider paying a
consultant to get them to work in LT Spice.

If sim is too much hassle, we'd solder.

On a sunny day (Wed, 12 Oct 2022 12:11:49 -0700) it happened John Larkin
<jlarkin@highland_atwork_technology.com> wrote in
<su2ekh57ijkql4og4sa35q8083sgv2avnj@4ax.com>:

>
>TI has encrypted Pspice models for many of their parts, but I can't
>persuade LT Spice to run the switchers I want to use. Any thoughts on
>that?
>
>One of my guys says he can run their Cadence sim, but I'd prefer LT.
>
>Parts are TPS54302, TPS563300, TPS562208. I'd consider paying a
>consultant to get them to work in LT Spice.
>
>If sim is too much hassle, we'd solder.

Solder is the only reality, no matter what spice says.
Board layout counts, maybe they have evaluation boards?

On Thu, 13 Oct 2022 03:32:26 GMT, Jan Panteltje
<pNaonStpealmtje@yahoo.com> wrote:

>On a sunny day (Wed, 12 Oct 2022 12:11:49 -0700) it happened John Larkin
><jlarkin@highland_atwork_technology.com> wrote in
><su2ekh57ijkql4og4sa35q8083sgv2avnj@4ax.com>:
>
>>
>>TI has encrypted Pspice models for many of their parts, but I can't
>>persuade LT Spice to run the switchers I want to use. Any thoughts on
>>that?
>>
>>One of my guys says he can run their Cadence sim, but I'd prefer LT.
>>
>>Parts are TPS54302, TPS563300, TPS562208. I'd consider paying a
>>consultant to get them to work in LT Spice.
>>
>>If sim is too much hassle, we'd solder.
>
>Solder is the only reality, no matter what spice says.
>Board layout counts, maybe they have evaluation boards?
>
>

I do have a couple of eval boards for the switchers. We can do that,
but Spice can show currents and such easier.

Everyone should just supply LT Spice models. A few semi makers do.

John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:

>Everyone should just supply LT Spice models. A few semi makers do.

I told that the TI guys every time. I also told them how nice parts I
can buy from Linear/Analog. You can hear teeth grinding in this
moments. :-D

Olaf

On Thu, 13 Oct 2022 06:40:24 +0200, olaf <olaf@criseis.ruhr.de> wrote:

>John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>
>
> >Everyone should just supply LT Spice models. A few semi makers do.
>
>I told that the TI guys every time. I also told them how nice parts I
>can buy from Linear/Analog. You can hear teeth grinding in this
>moments. :-D
>
>Olaf

The LTC parts are expensive, and being able to sim them easily is one
reason why. LT Spice probably added a couple billion dollars to the
price when Analog bought LTC.

I managed to get the TI Pspice model of TPS563300 into LT Spice and
run it. It looks OK until about 400 us into the sim, and hangs there
at 30 fs/s speed.

I guess I'll run the eval board.

We already use a similar part, TPS54302, and it's great. Cute, quiet
little switcher.

On Wed, 12 Oct 2022 22:02:33 -0700, John Larkin
<jlarkin@highlandSNIPMEtechnology.com> wrote:

>On Thu, 13 Oct 2022 06:40:24 +0200, olaf <olaf@criseis.ruhr.de> wrote:
>
>>John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>>
>>
>> >Everyone should just supply LT Spice models. A few semi makers do.
>>
>>I told that the TI guys every time. I also told them how nice parts I
>>can buy from Linear/Analog. You can hear teeth grinding in this
>>moments. :-D
>>
>>Olaf
>
>The LTC parts are expensive, and being able to sim them easily is one
>reason why. LT Spice probably added a couple billion dollars to the
>price when Analog bought LTC.
>
>I managed to get the TI Pspice model of TPS563300 into LT Spice and
>run it. It looks OK until about 400 us into the sim, and hangs there
>at 30 fs/s speed.
>
>I guess I'll run the eval board.
>
>We already use a similar part, TPS54302, and it's great. Cute, quiet
>little switcher.
>

The manual for the eval board is TI document sluuc67a.pdf.

Why do they do that?!

On a sunny day (Wed, 12 Oct 2022 22:09:13 -0700) it happened John Larkin
<jlarkin@highlandSNIPMEtechnology.com> wrote in
<g97fkht9tcoi8gtdmmoqapkc57boaugatj@4ax.com>:

>On Wed, 12 Oct 2022 22:02:33 -0700, John Larkin
><jlarkin@highlandSNIPMEtechnology.com> wrote:
>
>>On Thu, 13 Oct 2022 06:40:24 +0200, olaf <olaf@criseis.ruhr.de> wrote:
>>
>>>John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>>>
>>>
>>> >Everyone should just supply LT Spice models. A few semi makers do.
>>>
>>>I told that the TI guys every time. I also told them how nice parts I
>>>can buy from Linear/Analog. You can hear teeth grinding in this
>>>moments. :-D
>>>
>>>Olaf
>>
>>The LTC parts are expensive, and being able to sim them easily is one
>>reason why. LT Spice probably added a couple billion dollars to the
>>price when Analog bought LTC.
>>
>>I managed to get the TI Pspice model of TPS563300 into LT Spice and
>>run it. It looks OK until about 400 us into the sim, and hangs there
>>at 30 fs/s speed.
>>
>>I guess I'll run the eval board.
>>
>>We already use a similar part, TPS54302, and it's great. Cute, quiet
>>little switcher.
>>
>
>The manual for the eval board is TI document sluuc67a.pdf.
>
>Why do they do that?!

I dunno, google found it in one second, downloaded it and looking at it.
I do rename all those funny name pfds to what it really is on my system.
Most scientific papers also have a doi. ?? strange code, as do pictures
of space objects from NASA etc..

raspberrypi: ~/Downloads # mv sluuc67a.pdf TPS562202_evaluation_board__sluuc67a.pd
I leave the original 'code' in, preceded by a double __

Looks almost if they use a random password generator for the filenames....

BTW simple board, some positive spikes in Vout ...

On 13/10/2022 05:40, olaf wrote:
> John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>
>
> >Everyone should just supply LT Spice models. A few semi makers do.
>
> I told that the TI guys every time. I also told them how nice parts I
> can buy from Linear/Analog. You can hear teeth grinding in this
> moments. :-D

If the salesmen take that message of lost sales back home eventually the
suits that want to keep them all encrypted will get the message.

AES 256 bit encryption won't yield to brute force attacks.

--
Regards,
Martin Brown

On 13/10/2022 06:48, Jan Panteltje wrote:
> On a sunny day (Wed, 12 Oct 2022 22:09:13 -0700) it happened John Larkin
> <jlarkin@highlandSNIPMEtechnology.com> wrote in
> <g97fkht9tcoi8gtdmmoqapkc57boaugatj@4ax.com>:
>
>> On Wed, 12 Oct 2022 22:02:33 -0700, John Larkin
>> <jlarkin@highlandSNIPMEtechnology.com> wrote:
>>
>>> On Thu, 13 Oct 2022 06:40:24 +0200, olaf <olaf@criseis.ruhr.de> wrote:
>>>
>>>> John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>>>>
>>>>
>>>>> Everyone should just supply LT Spice models. A few semi makers do.
>>>>
>>>> I told that the TI guys every time. I also told them how nice parts I
>>>> can buy from Linear/Analog. You can hear teeth grinding in this
>>>> moments. :-D
>>>>
>>>> Olaf
>>>
>>> The LTC parts are expensive, and being able to sim them easily is one
>>> reason why. LT Spice probably added a couple billion dollars to the
>>> price when Analog bought LTC.
>>>
>>> I managed to get the TI Pspice model of TPS563300 into LT Spice and
>>> run it. It looks OK until about 400 us into the sim, and hangs there
>>> at 30 fs/s speed.
>>>
>>> I guess I'll run the eval board.
>>>
>>> We already use a similar part, TPS54302, and it's great. Cute, quiet
>>> little switcher.
>>>
>>
>> The manual for the eval board is TI document sluuc67a.pdf.
>>
>> Why do they do that?!
>
> I dunno, google found it in one second, downloaded it and looking at it.
> I do rename all those funny name pfds to what it really is on my system.
> Most scientific papers also have a doi. ?? strange code, as do pictures
> of space objects from NASA etc..
>
> raspberrypi: ~/Downloads # mv sluuc67a.pdf TPS562202_evaluation_board__sluuc67a.pd
> I leave the original 'code' in, preceded by a double __
>
> Looks almost if they use a random password generator for the filenames....

It is a stupid design of content management system that gives each new
entity a more or less random but unique filename from some rule. To
avoid typo collisions they populate its namespace sparsely.
>
> BTW simple board, some positive spikes in Vout ...

--
Regards,
Martin Brown

Am 13.10.22 um 10:12 schrieb Martin Brown:
> On 13/10/2022 05:40, olaf wrote:
>> John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>>
>>
>>   >Everyone should just supply LT Spice models. A few semi makers do.
>>
>> I told that the TI guys every time. I also told them how nice parts I
>> can buy from Linear/Analog. You can hear teeth grinding in this
>> moments. :-D
>
> If the salesmen take that message of lost sales back home eventually the
> suits that want to keep them all encrypted will get the message.

The TI op amp models work better in LT spice than those from AD,
as far as I have used them.
ADA4898 - THE HORROR!

The first model got about everything wrong. Phase,
noise densities in Volts...

Years later came a 2nd edition, it still has convergence
problems when you insert a second ADA4898 into your circuit,
completely unrelated.
I would not know how to create that behaviour if I wanted it.

That design went TI.

Cheers,
Gerhard

On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin Brown
<'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:

>It is a stupid design of content management system that gives each new
>entity a more or less random but unique filename from some rule. To
>avoid typo collisions they populate its namespace sparsely.

This image impressed me today:
https://www.space.com/solar-orbiter-5th-sun-flyby-video
Of course I downloaded the animated gif, it has a similar cryptic name...

On 13/10/2022 11:24, Jan Panteltje wrote:
> On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin Brown
> <'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:
>
>> It is a stupid design of content management system that gives each new
>> entity a more or less random but unique filename from some rule. To
>> avoid typo collisions they populate its namespace sparsely.
>
> This image impressed me today:
> https://www.space.com/solar-orbiter-5th-sun-flyby-video
> Of course I downloaded the animated gif, it has a similar cryptic name...

Of the recent JWST images these are particularly cool

star

https://webbtelescope.org/contents/media/images/01GEJB2906TM9VR2FSJ4TFMNQM

planet Neptune & rings

https://webbtelescope.org/contents/media/images/2022/046/01GCVNBC11Z11KDTJ5CFGTMX72

Again note random content management strings at work :(

--
Regards,
Martin Brown

On a sunny day (Thu, 13 Oct 2022 11:40:11 +0100) it happened Martin Brown
<'''newspam'''@nonad.co.uk> wrote in <ti8puc$1255$1@gioia.aioe.org>:

>On 13/10/2022 11:24, Jan Panteltje wrote:
>> On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin Brown
>> <'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:
>>
>>> It is a stupid design of content management system that gives each new
>>> entity a more or less random but unique filename from some rule. To
>>> avoid typo collisions they populate its namespace sparsely.
>>
>> This image impressed me today:
>> https://www.space.com/solar-orbiter-5th-sun-flyby-video
>> Of course I downloaded the animated gif, it has a similar cryptic name...
>
>Of the recent JWST images these are particularly cool
>
>star
>
>https://webbtelescope.org/contents/media/images/01GEJB2906TM9VR2FSJ4TFMNQM
>
>planet Neptune & rings
>
>https://webbtelescope.org/contents/media/images/2022/046/01GCVNBC11Z11KDTJ5CFGTMX72
>
>Again note random content management strings at work :(

Lots to be discovered in the near infrared!
Maybe with so many images of the same object around they _have_ to resort to some extra code in the filenames,
I mean how many images of Neptune are around, thousands if not more?

On 13/10/2022 12:02, Jan Panteltje wrote:
> On a sunny day (Thu, 13 Oct 2022 11:40:11 +0100) it happened Martin Brown
> <'''newspam'''@nonad.co.uk> wrote in <ti8puc$1255$1@gioia.aioe.org>:
>
>> On 13/10/2022 11:24, Jan Panteltje wrote:
>>> On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin Brown
>>> <'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:
>>>
>>>> It is a stupid design of content management system that gives each new
>>>> entity a more or less random but unique filename from some rule. To
>>>> avoid typo collisions they populate its namespace sparsely.
>>>
>>> This image impressed me today:
>>> https://www.space.com/solar-orbiter-5th-sun-flyby-video
>>> Of course I downloaded the animated gif, it has a similar cryptic name...
>>
>> Of the recent JWST images these are particularly cool
>>
>> star
>>
>> https://webbtelescope.org/contents/media/images/01GEJB2906TM9VR2FSJ4TFMNQM
>>
>> planet Neptune & rings
>>
>> https://webbtelescope.org/contents/media/images/2022/046/01GCVNBC11Z11KDTJ5CFGTMX72
>>
>> Again note random content management strings at work :(
>
> Lots to be discovered in the near infrared!

It could be a real boon for cosmology. They can almost see back to the
very first stars by using foreground Abell galaxy clusters as magnifying
lenses (and accepting a lot of distortion). The sparkler galaxy is one
such where the globular clusters around it are visible too.

https://iopscience.iop.org/article/10.3847/2041-8213/ac90ca

> Maybe with so many images of the same object around they _have_ to resort to some extra code in the filenames,
> I mean how many images of Neptune are around, thousands if not more?

I'd still prefer them named something like:

<catalogue_name>_<YYYYMMDDHHMMSS>_<wavelength>_<lead observer>
>

--
Regards,
Martin Brown

On 10/13/2022 14:20, Martin Brown wrote:
> On 13/10/2022 12:02, Jan Panteltje wrote:
>> On a sunny day (Thu, 13 Oct 2022 11:40:11 +0100) it happened Martin Brown
>> <'''newspam'''@nonad.co.uk> wrote in <ti8puc$1255$1@gioia.aioe.org>:
>>
>>> On 13/10/2022 11:24, Jan Panteltje wrote:
>>>> On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin
>>>> Brown
>>>> <'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:
>>>>
>>>>> It is a stupid design of content management system that gives each new
>>>>> entity a more or less random but unique filename from some rule. To
>>>>> avoid typo collisions they populate its namespace sparsely.
>>>>
>>>> This image impressed me today:
>>>>    https://www.space.com/solar-orbiter-5th-sun-flyby-video
>>>>     Of course I downloaded the animated gif, it has a similar
>>>> cryptic name...
>>>
>>> Of the recent JWST images these are particularly cool
>>>
>>> star
>>>
>>> https://webbtelescope.org/contents/media/images/01GEJB2906TM9VR2FSJ4TFMNQM
>>>
>>>
>>> planet Neptune & rings
>>>
>>> https://webbtelescope.org/contents/media/images/2022/046/01GCVNBC11Z11KDTJ5CFGTMX72
>>>
>>>
>>> Again note random content management strings at work :(
>>
>> Lots to be discovered in the near infrared!
>
> It could be a real boon for cosmology. They can almost see back to the
> very first stars by using foreground Abell galaxy clusters as magnifying
> lenses (and accepting a lot of distortion). The sparkler galaxy is one
> such where the globular clusters around it are visible too.
>
> https://iopscience.iop.org/article/10.3847/2041-8213/ac90ca
>
>
>> Maybe with so many images of the same object around they _have_ to
>> resort to some extra code in the filenames,
>> I mean how many images of Neptune are around, thousands if not more?
>
> I'd still prefer them named something like:
>
> <catalogue_name>_<YYYYMMDDHHMMSS>_<wavelength>_<lead observer>
>>
>

The naming mess is typically the product of data intended for machine
processing being served for human consumption. Like say the file names
in unix world, they are case sensitive - which is OK as long as they are
not meant for humans, which of course they are.

On 2022-10-13 12:40, Martin Brown wrote:
> On 13/10/2022 11:24, Jan Panteltje wrote:
>> On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin Brown
>> <'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:
>>
>>> It is a stupid design of content management system that gives each new
>>> entity a more or less random but unique filename from some rule. To
>>> avoid typo collisions they populate its namespace sparsely.
>>
>> This image impressed me today:
>> https://www.space.com/solar-orbiter-5th-sun-flyby-video
>> Of course I downloaded the animated gif, it has a similar cryptic name...
>
> Of the recent JWST images these are particularly cool
>
> star

A very cool picture indeed. The Wikipedia article claims that the
periodic bands are shells of matter expulsed by periodic He fusion
oscillations at an 8 year period, whereas the description with the
JWST picture says it's a binary star with that period.

I suppose the periodic bands would be a flat spiral then instead of
a series of concentric shells. It's hard to be sure, but it looks
like the bands are concentric.

Jeroen Belleman

Gerhard Hoffmann wrote:
> Am 13.10.22 um 10:12 schrieb Martin Brown:
>> On 13/10/2022 05:40, olaf wrote:
>>> John Larkin <jlarkin@highlandsnipmetechnology.com> wrote:
>>>
>>>
>>>   >Everyone should just supply LT Spice models. A few semi makers do.
>>>
>>> I told that the TI guys every time. I also told them how nice parts I
>>> can buy from Linear/Analog. You can hear teeth grinding in this
>>> moments. :-D
>>
>> If the salesmen take that message of lost sales back home eventually
>> the suits that want to keep them all encrypted will get the message.
>
> The TI op amp models work better in LT spice than those from AD,
> as far as I have used them.
> ADA4898 - THE HORROR!
>
> The first model got about everything wrong. Phase,
> noise densities in Volts...
>
> Years later came a 2nd edition, it still has convergence
> problems when you insert a second ADA4898 into your circuit,
> completely unrelated.
> I would not know how to create that behaviour if I wanted it.
>
> That design went TI.

The push to get native models for AD parts seems to have stalled since
Mike Engelhardt left. A pity.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

On 13/10/2022 14:53, Jeroen Belleman wrote:
> On 2022-10-13 12:40, Martin Brown wrote:
>> On 13/10/2022 11:24, Jan Panteltje wrote:
>>> On a sunny day (Thu, 13 Oct 2022 09:48:31 +0100) it happened Martin
>>> Brown
>>> <'''newspam'''@nonad.co.uk> wrote in <ti8jcv$1pnk$2@gioia.aioe.org>:
>>>
>>>> It is a stupid design of content management system that gives each new
>>>> entity a more or less random but unique filename from some rule. To
>>>> avoid typo collisions they populate its namespace sparsely.
>>>
>>> This image impressed me today:
>>>   https://www.space.com/solar-orbiter-5th-sun-flyby-video
>>>    Of course I downloaded the animated gif, it has a similar cryptic
>>> name...
>>
>> Of the recent JWST images these are particularly cool
>>
>> star
>
> A very cool picture indeed. The Wikipedia article claims that the
> periodic bands are shells of matter expulsed by periodic He fusion
> oscillations at an 8 year period, whereas the description with the
> JWST picture says it's a binary star with that period.

I think W-R stars are pretty much already down to burning helium in
their cores so that sounds implausible to me (but its not my area).

I think the latter explanation is much more plausible and fits with
observations in the X-ray here:

https://academic.oup.com/pasj/article/67/6/121/2470041
>
> I suppose the periodic bands would be a flat spiral then instead of
> a series of concentric shells. It's hard to be sure, but it looks
> like the bands are concentric.
Not necessarily. When the two stars are at their closest their strong
solar winds tear across their surfaces and the weaker one loses lots of
material. Weaker probably meaning the one with lowest surface gravity

FWIW I see concentric shells of material too. Slightly odd shape though
- I'd expected an oval symmetry like with planetary nebulae like M57.

--
Regards,
Martin Brown