Searching \ for '[EE]: Crowbar protection' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=crowbar+protection
Search entire site for: 'Crowbar protection'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Crowbar protection'
2002\08\23@183504 by Jinx

face picon face
A couple of days ago I had a 7805 regulator fail and stick
14V up a circuit, taking out an F877 and an LCD. To quote
my old French boss " I em 'ow you say ? urn'appy "

What can I use as a shunt circuit to stop this 'appening
again ? I've seen various semis used (SCR, FET, BJT)
with dividers, zeners etc but know that this list always
comes up trumps. I've tried the rest, now I want to try the
best

Generally I'd want this to protect a circuit that needs 5V @
100mA or less, which is probably not enough to blow a
common fuse, but should be sinkable by a modest semi.
Failure indication by LED would be nice

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spam_OUTlistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@184933 by Olin Lathrop

face picon face
> A couple of days ago I had a 7805 regulator fail and stick
> 14V up a circuit, taking out an F877 and an LCD. To quote
> my old French boss " I em 'ow you say ? urn'appy "
>
> What can I use as a shunt circuit to stop this 'appening
> again ?

You should first ask yourself whether it's really worth it.  This kind of
failure is very rare.  The cost of the protection parts will likely be way
more than the average cost of the damaged circuit.  Of course, there could
be other costs if the circuit fails that can make the protection worth it.
Another justification might be for bench use where failures and abuse are
more likely.

> I've seen various semis used (SCR, FET, BJT)
> with dividers, zeners etc but know that this list always
> comes up trumps. I've tried the rest, now I want to try the
> best

A zener, power transistor, and a resistor work pretty well to make a "high
power" zener equivalent.  The result is less accurate than the original
zener, but just fine for shunt protection.  Of course the whole setup needs
to be able to sink enough current drag the supply low enough or to blow a
fuse.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email .....listservKILLspamspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@184946 by Brendan Moran

flavicon
face
Here's my favorite.  Limited only by the forward voltage of the power
transistor, and the accuracy of zener you can find.


-------+-----------C   E----
      |             B
      +---/\/\/\/---+
      |             |
     ---           ---
     /Z\ Vz = 5.x  \ / SCR
     ---           ---
      |            /|
      |           | |
      +-----------+ |
      |             |
      /             |
      \             |
      /             |
      \             |
      /             |
      |             |
-------+-------------+-----

If I need to explain what any of that is, just tell me.

--Brendan

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservspamKILLspammitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@190355 by Rick C.

flavicon
face
In 30 years I have never had a 78xx go full throttle. It's pretty rare
unless maybe subject to a backfeed voltage to the output. An SCR, zener,
and resistor should suffice on the output. Zener cathode to the output,
anode to resistor to ground and the gate of the SCR to the junction of
the Z and R. Cathode of SCR obviously to ground and anode to the output.
To protect the supply, maybe a 1/4 watt 3R3 or less in series with the
source and kept a distance from any components or circuit board. It will
burn up before anything else does and will let you know it.
Rick

Jinx wrote:

{Quote hidden}

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email EraseMElistservspam_OUTspamTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@191023 by Roman Black
flavicon
face
Jinx wrote:
>
> A couple of days ago I had a 7805 regulator fail and stick
> 14V up a circuit, taking out an F877 and an LCD. To quote
> my old French boss " I em 'ow you say ? urn'appy "

Ouch!

{Quote hidden}

Hi Jinx, try a series resistor before the
7805, drop 5v on the resistor at 100mA, giving
14-5 = 7v for the regulator and dropout.

Try 56 ohms, 1/2 watt as it will get hot over 50mA
or so. This alone gives pretty good protection,
but adding a 1W 5.1v zener across the PIC is the
finishing touch.

The series resistor goes a long way by taking the
heat burden off the 7805, and you probably won't
blow another one. :o)
-Roman

PS. I like to split the cap before/after the
series resistor too, gives much cleaner power to
the 7805.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservspamspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@191421 by Dave Dilatush

picon face
part 1 1362 bytes content-type:text/plain; charset=us-ascii (decoded 7bit)

Jinx wrote...

>A couple of days ago I had a 7805 regulator fail and stick
>14V up a circuit, taking out an F877 and an LCD. To quote
>my old French boss " I em 'ow you say ? urn'appy "
>
>What can I use as a shunt circuit to stop this 'appening
>again ? I've seen various semis used (SCR, FET, BJT)
>with dividers, zeners etc but know that this list always
>comes up trumps. I've tried the rest, now I want to try the
>best
>
>Generally I'd want this to protect a circuit that needs 5V @
>100mA or less, which is probably not enough to blow a
>common fuse, but should be sinkable by a modest semi.
>Failure indication by LED would be nice

Since you've got plenty of voltage up front of your regulator,
you might consider a series current limiter instead of an
overvoltage crowbar circuit.  The attached example is of a
"precision current limiter" circuit from National Semiconductor's
LM117 datasheet.

The advantage of a current limiter, in my view, is that it
protects against a wider variety of mishaps than a crowbar, which
can only protect against overvoltage.

I concur with Olin; the type of failure you experienced isn't
very common.  Far more common is when some fumblebumpkin (like
me) accidently shorts something out with a screwdriver or probe.

DD



part 2 2546 bytes content-type:image/gif; name=limiter.gif (decode)


part 3 136 bytes
--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email @spam@listservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@192038 by Jim

flavicon
face
14 V eh?

Sounds like automotive mains.

A nasty environ - I lost LM555's when powered
directly off automotive mains YEARS ago - on
one of my first projects!

Solution was to add a real - sorry, an "RF" term
there - add a series resistor along with a shunt
electrolytic cap - any 'spikes' are met by (dropped
across) the R (chosen based on some maths) against
the influence of the electrolytic cap ...

RF Jim



   "Our ability to manufacture fraud has exceeded
    our  ability to detect it."

    - Al Pacino as Viktor Taransky in the movie 'Simone'

{Original Message removed}

2002\08\23@192439 by Jinx

face picon face
> You should first ask yourself whether it's really worth it.

Unfortunately the customer was in possession when it happened
so there's some credibility to be regained. I've rarely seen a reg
fail short circuit, but that's of little comfort to the customer. No idea
why that one chose to go postal but it did, and it was just dumb
luck it blew when it wasn't here

> This kind of failure is very rare.  The cost of the protection parts
> will likely be way more than the average cost of the damaged circuit.

In this case it was $30 for the LCD and $12 for the PIC, and they are
the bulk of the circuit cost. I think it's blown the crystal too. Bottom
line
is that for a dollar or two I'd like to show the customer that I'm taking it
seriously (even if it never happens again, which it probably won't)

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email KILLspamlistservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@195428 by Jinx

face picon face
> 14 V eh?
>
> Sounds like automotive mains.

Close - gel cell

> A nasty environ - I lost LM555's when powered
> directly off automotive mains YEARS ago - on
> one of my first projects!
>
> Solution was to add a real - sorry, an "RF" term
> there - add a series resistor along with a shunt
> electrolytic cap - any 'spikes' are met by (dropped
> across) the R (chosen based on some maths) against
> the influence of the electrolytic cap ...

No problem with spikes or anything else untoward AFAICT, this
particular regulator just decided to give up its day job

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@200321 by Brendan Moran

flavicon
face
> No problem with spikes or anything else untoward AFAICT, this
> particular regulator just decided to give up its day job

Decided to pursue its dream of being a heating element, did it?

--Brendan

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spamBeGonelistservspamBeGonespammitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@202016 by Jim

flavicon
face
Too bad we can't pop open/disolve the package and
take a look at the die and *see* what the failure
mode was: punch-through, metal migtration, over
temp, manufacturing defect, bond wire ???

We used to do that on our own RF power transistors
when I worked at Heathkit in Benton Harbor
(actually St. Joseph!). We did it VERY carefully
as the packages were BeO (Berillium Oxide is
toxic, we didn't *break* these open however,
there *is* a trick to doing it).

RF Jim

   "Our ability to manufacture fraud has exceeded
    our  ability to detect it."

    - Al Pacino as Viktor Taransky in the movie 'Simone'

{Original Message removed}

2002\08\23@202634 by Olin Lathrop

face picon face
> No problem with spikes or anything else untoward AFAICT, this
> particular regulator just decided to give up its day job

Well, maybe, but my experience with customers who have sudden unusual field
problems is that there's usually a very different explaination.  In other
words they're not going to tell you "We accidentally touched your gizmo to
the arc welder".  You're going to hear "It just went.  Nobody has been near
it for 3 years, and the whole time it was alone on a table on an anti-static
mat in a climate controlled chamber surrounded by 3 inches of lead
shielding".

Many years ago I worked for a company where the main (and only) computer was
a Prime 750 time sharing system.  Although it wasn't my real job, I was
usually the guy that would have to deal with computer failures.  They were
much more common in those days.  Average up time for the whole system
without incidence was probably just a month or two.

Early one morning I came in to find the system dead.  After some poking
around I isolated the problem to the serial lines board.  I pulled the board
out and saw a scortch mark the size of a quarter at a particular chip.  I
traced it back to figure out what serial line it was, then went to find the
other end.  This turned out to be 100 feet away in the production area.  I
found the end draped over a piece of supporting metal of a CRT monitor that
had the cover removed.  The end was dangling a few mm from the CRT's socket,
so it was pretty obvious what had happened.  Later when the production folks
showed up, I asked them about it.  Would you believe that despite the
smoking gun evidence that nobody had been near that setup for days?  Funny
how some of them knew to protest no knowledge before I had a chance to ask
them what had happened.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email TakeThisOuTlistservEraseMEspamspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@214050 by Paul Hutchinson

flavicon
face
What a bummer, Jinx :-(

Other replies have given you great crowbar circuits but, I do have one idea
about how it may have failed.

IIRC, this type of failure can be caused if the input is shorted and you
have a lot of capacitance (>5uF) on the output of the 7805.

An input short failure can be prevented by adding a diode from the input to
the output of the 7805. IIRC, the National Semi LM340/LM7805 data sheet
mentions this type of failure and the diode solution.

Paul

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservspamTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body


2002\08\23@214634 by David Duffy

flavicon
face
Jinx wrote:
>No problem with spikes or anything else untoward AFAICT, this
>particular regulator just decided to give up its day job

Any chance that it started oscillating? What decoupling caps do
you have on the input and output of the regulator? About the only
regulator I can recall being dud was one that was out of spec at
power-up. Something like 13V instead of 12V. I don't remember
the brand now - was some time ago. If your decoupling is all ok I
reckon you were just unlucky. Doubly so for it failing right in front
of the customer!  Murphy must have been at his place, not yours.
Regards...

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservEraseMEspam.....mitvma.mit.edu with SET PICList DIGEST in the body


2002\08\24@013614 by Jinx

face picon face
> > No problem with spikes or anything else untoward AFAICT, this
> > particular regulator just decided to give up its day job
>
> Well, maybe, but my experience with customers who have sudden
> unusual field problems is that there's usually a very different
> explaination. In other words they're not going to tell you "We
> accidentally touched your gizmo to the arc welder".  You're going
> to hear "It just went.Nobody has been near it for 3 years, and the
> whole time it was alone on a table on an anti-static mat in a climate
> controlled chamber surrounded by 3 inches of lead shielding"

This very afternoon I've had returned to me a prototype that someone's
been testing for a couple of weeks. As far as I was aware it was working
perfectly well to the specs as arranged. The work of art (IMHO ;-) ) that
was taken away is virtually unrecognisable. I'm livid. Having changed
their mind after delivery, it's been hacked something awful. Relays are
hanging out of the box, the LCD doesn't work, probably the PIC neither,
an LED is missing and who knows what else is OS. Apart from the
obvious physical changes, they've tried to bypass s/w controlled relays
with toggle switches (incompetently, as they admitted......so why do it
????), with the result that many pins of the PIC have had 12V on them.
Which explains why the bloody thing is dead in the water and "doesn't
work". It's not so much the repairs, (which they will pay for), it's the
total lack of consideration and being already busy it's not much fun
doing the same work twice

I shall, however, maintain a dignified composure and mutter just a
gritted-toothed "flippin' 'eck"

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspammitvma.mit.edu


2002\08\24@035752 by Russell McMahon

face
flavicon
face
If the supply was in an automotive application (although it sounds like it
wasn't) then the LM293x series are worth looking at. These are designed to
be tolerant of various abuses beyond that expected in more peaceful
environments.


       RM

> Unfortunately the customer was in possession when it happened
> so there's some credibility to be regained. I've rarely seen a reg
> fail short circuit, but that's of little comfort to the customer. No idea
> why that one chose to go postal but it did, and it was just dumb
> luck it blew when it wasn't here

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestEraseMEspamEraseMEmitvma.mit.edu


2002\08\24@060922 by Peter L. Peres

picon face
On Fri, 23 Aug 2002, Brendan Moran wrote:

>Here's my favorite.  Limited only by the forward voltage of the power
>transistor, and the accuracy of zener you can find.

That circuit will destroy the zener diode and/or the thyristor gate when
an event occurs and it will have 0.6V voltage drop through the transistor,
dissipating power. It is not self resetting (it could be with that parts
count). Just an opinion.

Peter

{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestTakeThisOuTspamspammitvma.mit.edu


2002\08\24@070419 by Peter L. Peres

picon face
On Sat, 24 Aug 2002, Jinx wrote:

>A couple of days ago I had a 7805 regulator fail and stick
>14V up a circuit, taking out an F877 and an LCD. To quote
>my old French boss " I em 'ow you say ? urn'appy "
>
>What can I use as a shunt circuit to stop this 'appening
>again ? I've seen various semis used (SCR, FET, BJT)
>with dividers, zeners etc but know that this list always
>comes up trumps. I've tried the rest, now I want to try the
>best

There is no 'best'. Use a thyristor with a divider (can contain a zener to
increase trip point precision), and a fuse. There are fast crowbars that
use mosfets to short the power lines. Wire the led in parallel with the
fuse, with a suitable resistor. TL431s can be used to drive the thyristor
with greater precision.

Peter

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspamspamspamBeGonemitvma.mit.edu


2002\08\24@143722 by Brendan Moran

flavicon
face
>That circuit will destroy the zener diode and/or the thyristor gate when
>an event occurs and it will have 0.6V voltage drop through the transistor,
>dissipating power. It is not self resetting (it could be with that parts
>count). Just an opinion.
Is that due to the junction between the zener Anode and the gate of the
SCR? If so, just add a resistor there. No biggie. Everything else is
protected by resistors.

> >
> >-------+-----------C E----
> > | B
> > +---/\/\/\/---+
> > | |
> > --- ---
> > /Z\ Vz = 5.x \ / SCR
> > --- ---
> > | /|
> > | | |
> > +-----------+ |
> > | |
> > / |
> > \ |
> > / |
> > \ |
> > / |
> > | |
> >-------+-------------+-----

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestKILLspamspammitvma.mit.edu


2002\08\25@064351 by Howard Winter

face
flavicon
picon face
Hi - I'm new here and I may be about to talk absolute rubbish, but I may as well start as I mean to go on!
:-)
On Sat, 24 Aug 2002 11:23:30 +1200, Jinx wrote:

<Snip>
> In this case it was $30 for the LCD and $12 for the PIC, and they are
> the bulk of the circuit cost.

Ouch!

> I think it's blown the crystal too.

Err - would this happen?  I've never tried putting 14VDC across a crystal, but I'd be surprised if it was
fatal, unless it physically overstressed the crystal itself and broke it.

> Bottom lineis that for a dollar or two I'd like to show the customer that I'm taking it
> seriously (even if it never happens again, which it probably won't)

I'm with you, in that a solution to a problem that *has* happened (however rare) does much more for customer
relations than a shrug and "well that's never happened before - so it probably won't happen again"!

Since the failure in shorted mode is so rare, my detective instincts lead me to suspect something else -
disconnecting the common connection to a 78xx will work, of course, but are you sure it wasn't an incoming
spike of some sort?  If it's mains powered, could it have been interference from something noisy?  If it's
automotive, has it had any arc- or spot-welding done?  What about thunderstorms, or nuclear-weapons testing
nearby?  (Any suspicious French fleets thereabouts? :-)

In terms of circuitry, I'd go for overkill:  A fuse on input, followed by a "dropper" resistor that will
result in about 8V being presented to the 7805, and one of the crowbars already suggested downstream of the
7805.  This means that you are giving the 7805 an easier life, the crowbar will prevent damage if it does fail
or have its common disconnected, and the fuse will close things down before the fire!

Cheers,

Howard Winter
St.Albans, England

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\08\25@082806 by Russell McMahon

face
flavicon
face
> Since the failure in shorted mode is so rare, my detective instincts lead
me to suspect something else -

Someone else noted the need for reverse voltage protection diodes. These are
covered in the data sheet. The need arises when the input voltage falls much
more rapidly than the output and the output cap discharges "backwards" into
the regulator. This can occur if the input is shorted suddenly OR if the
output load is small relative to the output cap provided while the input can
discharge rapidly when turned off - either because there is little or n
input cap or because some other load rapidly discharges the input cap.

In these cases and others the LM293x family are specifically designed to
survive. Their forte is automotive situations replete with alternators, car
batteries, inductive loads and numerous somewhat silly people wielding
jumper leads and reverse connected batteries. I strongly recommend this
regulator family. (Standard disclaimer, no shares in NatSemi or other
financial etc interest yada yada).

You said 7805 - hopefully NOT a 78L05 - the TO92 miniature version. These do
die for reasons that I have never bothered to determine - not using them is
easier. The TO220 version costs less usually and the TO220 dissipates more
power when unheatsunk than the TO92 pkg.


       RM

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\08\25@090609 by Roman Black

flavicon
face
Russell McMahon wrote:

> Someone else noted the need for reverse voltage protection diodes. These are
> covered in the data sheet. The need arises when the input voltage falls much
> more rapidly than the output and the output cap discharges "backwards" into
> the regulator. This can occur if the input is shorted suddenly OR if the
> output load is small relative to the output cap provided while the input can
> discharge rapidly when turned off - either because there is little or n
> input cap or because some other load rapidly discharges the input cap.


In the case I mentioned with a large series resistor
(like 56 ohms etc), you can sometimes dispense with
the diodes as the resistor will protect the 7805
if the input cap is shorted. It is only a good solution
where you need <100mA from the 7805.

I first saw this technique in some "student proofed"
logic trainers which used a 100 ohm resistor in front
of each 7805 reg, and the regs sagged at currents
over 60mA. More recently i've seen this all the time
in TV sets etc where they use a series resistor before
the regs as the resistor takes the bulk of the
dissipation and saves the cost of a heatsink etc.
I always use a series resistor now any time the 7805
has 12v or higher input voltage.
-Roman

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\08\26@104000 by llile

flavicon
face
I had some fun with a 78M05 (the TO220 kind) last week.  My circuit was
designed to provide about 35V to the input of a 78M05.  That is about the
limit of the input voltage as well.

The chip works fine up to about 38 volts input, but near 40V input (I was
testing what happened at higher input voltages) the 5V end of it rapidly
loses it's ability to supply current.  I was trying to figure out whether
the chip was shutting itself down because it was outside of it's SOA (I
think these things have a thermal shutdown ) or just failing.

Now, the problem would occur just milliseconds after the power was turned
on, so I decided it couldn't be a thermal shutdown issue.  As soon as I
turned on a few LEDs and sucked up some current the power supply would
dive to 4V and reboot my PIC.  Also, the problem seemed to get worse as
the overvoltage was applied more times, so I decided it was a failure mode
instead of a self-protection mode.  After a few hundred cycles on an
automatic test machine the unit would not work at all.   I replaced the
7805 with a fresh one and added a 33V zener to clip those peaks.  This
design also has a resistor in front of the 7805 to reduce power
dissipation.

This morning my automatic test machine shows 1100 cycles without a hitch.

I haven't had to use a pre-regulator before my regulator in anything but a
precision benchtop power supply.  I wonder if there is a cheap automotive
5V regulator that will handle 48V DC input voltage without complaining?

-- Lawrence Lile





Russell McMahon <apptechSTOPspamspamspam_OUTPARADISE.NET.NZ>
Sent by: pic microcontroller discussion list <spamBeGonePICLISTSTOPspamspamEraseMEMITVMA.MIT.EDU>
08/25/02 07:28 AM
Please respond to pic microcontroller discussion list


       To:     KILLspamPICLISTspamBeGonespamMITVMA.MIT.EDU
       cc:
       Subject:        Re: [EE]: Crowbar protection


> Since the failure in shorted mode is so rare, my detective instincts
lead
me to suspect something else -

Someone else noted the need for reverse voltage protection diodes. These
are
covered in the data sheet. The need arises when the input voltage falls
much
more rapidly than the output and the output cap discharges "backwards"
into
the regulator. This can occur if the input is shorted suddenly OR if the
output load is small relative to the output cap provided while the input
can
discharge rapidly when turned off - either because there is little or n
input cap or because some other load rapidly discharges the input cap.

In these cases and others the LM293x family are specifically designed to
survive. Their forte is automotive situations replete with alternators,
car
batteries, inductive loads and numerous somewhat silly people wielding
jumper leads and reverse connected batteries. I strongly recommend this
regulator family. (Standard disclaimer, no shares in NatSemi or other
financial etc interest yada yada).

You said 7805 - hopefully NOT a 78L05 - the TO92 miniature version. These
do
die for reasons that I have never bothered to determine - not using them
is
easier. The TO220 version costs less usually and the TO220 dissipates more
power when unheatsunk than the TO92 pkg.


       RM

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\08\26@141257 by Peter L. Peres

picon face
> 7805

Put the zener in series with the 7805 input ? A zener costs about as much
as a medium power resistor I think. 1W zeners are inexpensive. There is
the problem of zener failure (they fail shorted usually).

Peter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\08\26@163017 by Olin Lathrop

face picon face
> I haven't had to use a pre-regulator before my regulator in anything but a
> precision benchtop power supply.  I wonder if there is a cheap automotive
> 5V regulator that will handle 48V DC input voltage without complaining?

That's a pretty large drop proportionally.  At 48V in 5V out efficiency is
about 10%.  How about a simple buck converter?


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\08\26@164721 by D. Schouten

flavicon
face
> I haven't had to use a pre-regulator before my regulator in anything but a
> precision benchtop power supply.  I wonder if there is a cheap automotive
> 5V regulator that will handle 48V DC input voltage without complaining?

You could use a TL783 from TI. This regulator handles an input/output
differential voltage of up to 125VDC. However, it would cost you an
additional two resistors too since it is an adjustable regulator. But still,
stepping linearly down from 48V to 5V would at least require a small output
current to keep dissipation reasonable. I use switchers for all my 48V to 5V
or 12V converters for years now. Works like a dream.

Daniel...

{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\08\26@170352 by llile

flavicon
face
>  I use switchers for all my 48V to 5V
or 12V converters for years now. Works like a dream.

I have been afraid of switchers because I thought they would cost a lot
more than good ol' 78xx linear regulators.  What is the price premium for
a switcher including all those inductors over a plain linear regulator?

Obviously the switcher is more efficient, and some people goota have that
efficiency.  My stuff is all going in 1500 watt ovens, so efficiency is
not an issue, just cost and reliability.

--Lawrence










"D. Schouten" <.....d.schoutenspam_OUTspamQUICKNET.NL>
Sent by: pic microcontroller discussion list <TakeThisOuTPICLIST.....spamTakeThisOuTMITVMA.MIT.EDU>
08/26/02 03:47 PM
Please respond to pic microcontroller discussion list


       To:     TakeThisOuTPICLISTKILLspamspamspamMITVMA.MIT.EDU
       cc:
       Subject:        Re: [EE]: Crowbar protection


> I haven't had to use a pre-regulator before my regulator in anything but
a
> precision benchtop power supply.  I wonder if there is a cheap
automotive
> 5V regulator that will handle 48V DC input voltage without complaining?

You could use a TL783 from TI. This regulator handles an input/output
differential voltage of up to 125VDC. However, it would cost you an
additional two resistors too since it is an adjustable regulator. But
still,
stepping linearly down from 48V to 5V would at least require a small
output
current to keep dissipation reasonable. I use switchers for all my 48V to
5V
or 12V converters for years now. Works like a dream.

Daniel...

{Quote hidden}

These
> do
> die for reasons that I have never bothered to determine - not using them
> is
> easier. The TO220 version costs less usually and the TO220 dissipates
more
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\08\26@171610 by Spehro Pefhany

picon face
At 07:19 PM 8/23/02 -0500, you wrote:
>Too bad we can't pop open/disolve the package and
>take a look at the die and *see* what the failure
>mode was: punch-through, metal migtration, over
>temp, manufacturing defect, bond wire ???

Or the middle lead just got itself opened (bad
wire bond, bad soldering to PCB). That the only
time I've seen a 78xx do that.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
TakeThisOuTspeffspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
9/11 United we Stand

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2002\08\27@045327 by Russell McMahon

face
flavicon
face
> At 07:19 PM 8/23/02 -0500, you wrote:
> >Too bad we can't pop open/disolve the package and
> >take a look at the die and *see* what the failure
> >mode was: punch-through, metal migtration, over
> >temp, manufacturing defect, bond wire ???

Boiling in concentrated sulphuric acid is an approved method to do just
that, I've been told.
Don't try that at home :-) !

       RM

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\08\27@143243 by Peter L. Peres

picon face
On Mon, 26 Aug 2002 llileEraseMEspamSALTONUSA.COM wrote:

>>  I use switchers for all my 48V to 5V
>or 12V converters for years now. Works like a dream.
>
>I have been afraid of switchers because I thought they would cost a lot
>more than good ol' 78xx linear regulators.  What is the price premium for
>a switcher including all those inductors over a plain linear regulator?
>
>Obviously the switcher is more efficient, and some people goota have that
>efficiency.  My stuff is all going in 1500 watt ovens, so efficiency is
>not an issue, just cost and reliability.

Can't you steal power from a tap on the heater ? This is very common in
certain categories of appliances (no necessarily for home use). Also a
small (<5W) power transformer with thermal fuse should be less than $1 in
quantity (the kind used in cheap multivoltage wallwarts f.ex. - the whole
device is sold retail in ones for $3 here - this is not a wholesale
bargain, it's the normal price of the things). Switching the display to
LCD should drop the power requirements quite a lot. Then you can use a EL
or FL backlight driven from mains directly.

I think that it is hard to compete with the price of 3-5W transformers
with a smpsu.

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\08\27@215726 by Roman Black

flavicon
face
llile@SALTONUSA.COM wrote:
>
> >  I use switchers for all my 48V to 5V
> or 12V converters for years now. Works like a dream.
>
> I have been afraid of switchers because I thought they would cost a lot
> more than good ol' 78xx linear regulators.  What is the price premium for
> a switcher including all those inductors over a plain linear regulator?
>
> Obviously the switcher is more efficient, and some people goota have that
> efficiency.  My stuff is all going in 1500 watt ovens, so efficiency is
> not an issue, just cost and reliability.


Hi Lawrence, there is a design for 2-transistor
switcher regulator here:
http://www.romanblack.com/smps.htm

This is a REALLY cheap regulator, and can use a
cheaper inductor than the ones usually specified
for "proper" switcher chips.
Total costing I did for 500 Q was about $0.35 USD
for the entire regulator, transistors, inductor,
caps, everything.

But, it doesn't regulate very tightly, with voltage
drooping 0.3v as current goes from 20mA to 50mA.
For many PIC designs that is no problem, but if you
want better regulation you can simply tack a 5.1v
1W zener across the output and get a nice regulated
output for a fraction less efficiency. You can also
reduce the value of R1 to improve regulation again if
you don't mind losing a few % more efficiency.

With your 40+ input volts it will still be about 5
to 6 times more efficient than a 7805 regulator,
and possibly cheaper. :o)
-Roman

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\08\28@101509 by llile

flavicon
face
Dewd!  You are Awesome!  Thanks!

-- Lawrence Lile






Roman Black <RemoveMEfastvidEraseMEspamspam_OUTEZY.NET.AU>
Sent by: pic microcontroller discussion list <@spam@PICLISTRemoveMEspamEraseMEMITVMA.MIT.EDU>
08/27/02 08:52 PM
Please respond to pic microcontroller discussion list


       To:     EraseMEPICLISTspam@spam@MITVMA.MIT.EDU
       cc:
       Subject:        Re: [EE]: Crowbar protection


Hi Lawrence, there is a design for 2-transistor
switcher regulator here:
http://www.romanblack.com/smps.htm

This is a REALLY cheap regulator, and can use a
cheaper inductor than the ones usually specified
for "proper" switcher chips.
Total costing I did for 500 Q was about $0.35 USD
for the entire regulator, transistors, inductor,
caps, everything.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email @spam@listservspam_OUTspam.....mitvma.mit.edu with SET PICList DIGEST in the body


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...