Path: ...!news.mixmin.net!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net>
Newsgroups: sci.electronics.design
Subject: Re: Ever smaller DC-DC converter modules by TI
Date: Mon, 2 Sep 2024 16:35:15 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 109
Message-ID: <vb4pg3$2trd3$1@dont-email.me>
References: <vb1ouh$210pa$1@solani.org>
 <vb1qnl$1gmdp$1@dont-email.me>
 <0a6c0562-cea1-4fdd-b283-04a8c57e762a@ademu.nl>
 <nnd$2f75283c$0bf25191@b9829d5a22e7a1d0>
 <vb374g$20con$1@solani.org>
 <vb3v93$3ogml$1@dont-email.me>
 <0fobdjdni427sa8daucko0lkipimq5038n@4ax.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Date: Mon, 02 Sep 2024 18:35:16 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="70374c240c7f865ef1f351c4b8905c96";
	logging-data="3075491"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/a1fg3gtkNMkBc9NxRCrBw"
User-Agent: NewsTap/5.5 (iPhone/iPod Touch)
Cancel-Lock: sha1:ILFulGJueKTR5SEA7wF1J1J7UwQ=
	sha1:egcy0o39ekjk5zRGSabAP9jttTY=
Bytes: 5193

john larkin <jlarkin_highland_tech> wrote:
> On Mon, 2 Sep 2024 09:07:48 -0000 (UTC), Phil Hobbs
> <pcdhSpamMeSenseless@electrooptical.net> wrote:
> 
>> Jan Panteltje <alien@comet.invalid> wrote:
>>> On a sunny day (Sun, 1 Sep 2024 17:43:26 +0200) it happened Arie de Muijnck
>>> <noreply@ademu.nl> wrote in <nnd$2f75283c$0bf25191@b9829d5a22e7a1d0>:
>>> 
>>>> On 2024-09-01 17:29, Arie de Muijnck wrote:
>>>>> On 2024-09-01 16:35, john larkin wrote:
>>>>>> Jan Panteltje <alien@comet.invalid> wrote:
>>>>>>> Ever smaller DC-DC converter modules by TI:
>>>>>>> ?á
>>>>>>> https://www.electronicdesign.com/technologies/power/article/55134121/electronic-design-tis-magnetic-materials-and-packaging-magic-help-slash-dc-dc-module-size-even-further
>>>>>>> This innovation is based largely, but not entirely,
>>>>>>> on advances in magnetic materials,
>>>>>>> and it didn?ÇÖt come quickly.
>>>>>>> Noted Jeff Morroni, director of power management research and development
>>>>>>> at TI's Kilby Labs,
>>>>>>> 
>>>>>>> For example:
>>>>>>> 1.8 to 5.5V in to 5.5V out at 5.5A size 2.5 x 2.6 mm boost module
>>>>>>> More examples in link.
>>>>>>> 
>>>>>> 
>>>>>> 1W isolated 5 -> 3.3V, 4x5x1 mm, SS, *76 MHz* switching frequency, for
>>>>>> $1.50 @ 1ku?
>>>>>> 
>>>>>> Potentially pretty cool!
>>>>>> 
>>>>>> Cheers
>>>>>> 
>>>>>> Phil Hobbs
>>>>>> 
>>>>> 
>>>>> I noticed in the datasheet to avoid disabling it under full load.
>>>>> To me that reads the shrapnel would reach the ceiling...
>>>>> 
>>>>> Arie
>>>> 
>>>> After better reading the datasheet of the boost converter, TPSM81033,
>>>> most interesting for me right now.
>>>> And not disabling but removing the input voltage. Like with the good old
>>>> LM78xx regulators.
>>>> 
>>>> "
>>>> Device Limitations (intended to be removed for final material):
>>>> 1. Inductor current limit function is not optimized. TI don't recommend
>>>> to use TPSM81033 at the
>>>> condition that the input current is higher than 5A. Final material is
>>>> expected to improve the current
>>>> limit function.
>>>> 2. TPSM81033 doesn't support to remove input voltage at heavy load. TI
>>>> recommend to remove
>>>> input voltage at no load, or connect EN to GND first with load and then
>>>> remove input voltage. Final
>>>> material is expected to support remove input voltage at heavy load.
>>> "
>>> 
>>> That sure is a big limitation!
>>> 
>> 
>> I’ll certainly be reading the datasheet carefully before using the part,
>> but I expect that it’s not a big worry in real life. 
>> 
>> You’re going to have some sort of reservoir cap on the input, and with a 76
>> MHz switching frequency, its discharge will be nice and smooth.  
>> 
>> It would take work to cut off the input supply fast enough to cause
>> inductive problems. 
>> 
>> Cheers 
>> 
>> Phil Hobbs 
> 
> I'm always suspious when I see a note like that on a data sheet.
> Oftimes, they are hiding a latchup or something.
> 
> If it does latch up, a brief brownout, or some spike, might fry the
> thing. I'd get an eval board and play with one before designing it
> into anything.
> 
No fear. I don’t use any switcher chips that I haven’t tried out, good and
hard. 

My soul was seared by a TI part that the datasheet swore would keep on
PWMing right down to zero load current. They made a big deal about it. 

Aha! I says, says I, “Perfect for my fancy isolated triple-output flybuck!”

After I’d designed it into the guts of a biggish board, it turned out to be
all lies—with a light load on the main output, it went into burp mode and
the isolated outputs collapsed. 

Barstids.  Never again. 

Cheers 

Phil Hobbs 

> Sometimes the current ratings are optimistic too.
> 
> 



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