No current sense resistor required for MOSFET switch protection
in [Design]
|
Prev: I will renew all of my works books. I am Isaac Newton, have you already drawn a comet on your eggs on Easter?
Next: I hope this is a heartening economic sign...
From: John Larkin on 7 Apr 2010 18:00 On Wed, 07 Apr 2010 17:42:45 -0400, Phil Hobbs <pcdhSpamMeSenseless(a)electrooptical.net> wrote: >On 4/6/2010 12:31 PM, John Larkin wrote: >> On Mon, 5 Apr 2010 17:23:01 -0700 (PDT), "oparr(a)hotmail.com" >> <oparr(a)hotmail.com> wrote: >> >>> No relatively expensive high side current monitor, comparator or latch >>> ICs either. You're wasting your time and money using these items if >>> all you need is a fast electronic "fuse". A single BJT, in conjunction >>> with the MOSFET, is all that is required for UVLO, overcurrent and >>> short circuit protection. Throw in a voltage reference, like the >>> TL431, if the switch has to cover a wide voltage range. >> >> I sort of doubt it. >> >> John >> > >He can do most of that (if he's prepared to be crappy enough) by using >one of those beta-dependent BJT circuits of yours--stick a BJT in series >with the source of the MOSFET, and give it just enough base current to >almost saturate at the highest current you want to allow. That would work at the microamp sort of level. Not very useful. The thing I like about beta limiting, in addition to its electrical virtues, is that everybody has been taught many, many times to never, never do it. I can't pass up a challenge like that. People who will cheerfully use a jfet with 10:1 Idss ratings won't consider beta design using a BCX71K with a less than 2:1 HFE spread. I really like the Supertex depletion fets as signal-level protectors. John
From: Jim Thompson on 7 Apr 2010 18:07 On Wed, 07 Apr 2010 17:42:45 -0400, Phil Hobbs <pcdhSpamMeSenseless(a)electrooptical.net> wrote: >On 4/6/2010 12:31 PM, John Larkin wrote: >> On Mon, 5 Apr 2010 17:23:01 -0700 (PDT), "oparr(a)hotmail.com" >> <oparr(a)hotmail.com> wrote: >> >>> No relatively expensive high side current monitor, comparator or latch >>> ICs either. You're wasting your time and money using these items if >>> all you need is a fast electronic "fuse". A single BJT, in conjunction >>> with the MOSFET, is all that is required for UVLO, overcurrent and >>> short circuit protection. Throw in a voltage reference, like the >>> TL431, if the switch has to cover a wide voltage range. >> >> I sort of doubt it. >> >> John >> > >He can do most of that (if he's prepared to be crappy enough) by using >one of those beta-dependent BJT circuits of yours--stick a BJT in series >with the source of the MOSFET, and give it just enough base current to >almost saturate at the highest current you want to allow. > >NOT how you'd want to do it in real life, but it'd limit the current and >so on. > >Cheers > >Phil Hobbs Yep. Larkin's magic bag of tricks can certainly give one heartburn ;-) ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | 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 | The only thing bipartisan in this country is hypocrisy
From: Jim Thompson on 7 Apr 2010 18:10 On Wed, 07 Apr 2010 15:00:34 -0700, John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote: >On Wed, 07 Apr 2010 17:42:45 -0400, Phil Hobbs ><pcdhSpamMeSenseless(a)electrooptical.net> wrote: > >>On 4/6/2010 12:31 PM, John Larkin wrote: >>> On Mon, 5 Apr 2010 17:23:01 -0700 (PDT), "oparr(a)hotmail.com" >>> <oparr(a)hotmail.com> wrote: >>> >>>> No relatively expensive high side current monitor, comparator or latch >>>> ICs either. You're wasting your time and money using these items if >>>> all you need is a fast electronic "fuse". A single BJT, in conjunction >>>> with the MOSFET, is all that is required for UVLO, overcurrent and >>>> short circuit protection. Throw in a voltage reference, like the >>>> TL431, if the switch has to cover a wide voltage range. >>> >>> I sort of doubt it. >>> >>> John >>> >> >>He can do most of that (if he's prepared to be crappy enough) by using >>one of those beta-dependent BJT circuits of yours--stick a BJT in series >>with the source of the MOSFET, and give it just enough base current to >>almost saturate at the highest current you want to allow. > >That would work at the microamp sort of level. Not very useful. > >The thing I like about beta limiting, in addition to its electrical >virtues, is that everybody has been taught many, many times to never, >never do it. I can't pass up a challenge like that. People who will >cheerfully use a jfet with 10:1 Idss ratings won't consider beta >design using a BCX71K with a less than 2:1 HFE spread. > >I really like the Supertex depletion fets as signal-level protectors. > >John > > > Hell! I regularly design around PNP's with a worst-case beta of 1 (yep, one!) I love chip processes that have depletion-mode FET's... just great for Bandgap starters. ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | 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 | The only thing bipartisan in this country is hypocrisy
From: Winfield Hill on 7 Apr 2010 19:37 Jim Thompson wrote... > > I love chip processes that have depletion-mode FET's... > just great for Bandgap starters. Don't all fabs have JFETs? Maybe not defined and 'scoped out, but aren't they at least able to accommodate them if you roll your own? -- Thanks, - Win
From: Jim Thompson on 7 Apr 2010 19:42
On 7 Apr 2010 16:37:57 -0700, Winfield Hill <Winfield_member(a)newsguy.com> wrote: >Jim Thompson wrote... >> >> I love chip processes that have depletion-mode FET's... >> just great for Bandgap starters. > > Don't all fabs have JFETs? Maybe not defined and 'scoped > out, but aren't they at least able to accommodate them if > you roll your own? Some fabs have an N-well resistor that is sort of a JFET. The problem is you want one that cuts off sharply enough that start-up goes away completely. I've only seen a few useable for low voltage circuits, such as in RFID tags. ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | 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 | The only thing bipartisan in this country is hypocrisy |