EDAboard.com | EDAboard.eu | EDAboard.de | EDAboard.co.uk | RTV forum PL | NewsGroups PL

"Use MOSFET body didoes"

Ask a question - edaboard.com

elektroda.net NewsGroups Forum Index - Electronics Design - "Use MOSFET body didoes"

Goto page Previous  1, 2, 3  Next


Guest

Thu Jan 05, 2017 7:47 am   



On Thursday, January 5, 2017 at 3:21:49 PM UTC+11, John Larkin wrote:
Quote:
On 4 Jan 2017 19:32:31 -0800, Winfield Hill <hill_at_rowland.harvard.edu
wrote:

John Larkin wrote...

We tried that, but it only helped some, not enough
to make the chip usable.

I suppose most users simply lived with the noise.

The spikes were making huge offsets in opamps all over the board.


Somebody wasn't being careful about where the spikes went. The sort of goof who dumps that kind of spike into a capacitor whose other end is buried in a board-wide ground plane deserves everything they get.

You do have to worry about routing switching spikes around the shortest possible circuit (minimal included area), and that does mean paying attention to every element around the loop.

<snip>

--
Bill Sloman, Sydney


Guest

Thu Jan 05, 2017 8:29 am   



On Wed, 04 Jan 2017 10:47:56 -0800, John Larkin
<jjlarkinxyxy_at_highlandtechnology.com> wrote:

Quote:
On Wed, 4 Jan 2017 12:48:08 -0500, bitrex
bitrex_at_de.lete.earthlink.net> wrote:

On 01/04/2017 12:10 PM, John Larkin wrote:
On Wed, 4 Jan 2017 11:58:44 -0500, bitrex
bitrex_at_de.lete.earthlink.net> wrote:

On 01/04/2017 11:52 AM, John Larkin wrote:
On 4 Jan 2017 08:14:07 -0800, Winfield Hill <hill_at_rowland.harvard.edu
wrote:

bill.sloman_at_ieee.org wrote...
John Larkin wrote:
Tim Wescott wrote:

Old hat. To be used with caution, because not all body diodes are
created equal -- for some fast switching regulators you really want
to parallel the FETs with Schottky diodes, as they recover quicker.
A FET with a body diode that has slow recovery can become smoke
pretty quickly.

Some mosfet body diodes make nice step-recovery diodes.

So they can spray really high-frequency noise around the circuit?
Just what you want when you have to meet EMI standards.

All MOSFETs do that, to one degree or another. Some have
added soft-recovery diodes in parallel with the body diode,
which helps, but designers of half-bridge stages, etc.,
are always struggling to prevent body-diode conduction,
or limit its current*time, or slow switching, or snub any
voltage spike from the sudden high dI/dt at snap off.
The problem is worse at maximum switched current. And
most off-line designs can't use added Schottky diodes,
since high-voltage versions have too high an ON voltage.


This is a really classic SRD waveform:

https://dl.dropboxusercontent.com/u/53724080/Parts/LM3102_SwitcherRise.JPG

This switcher is basically unusable. A paralleled external schottky
only helped some.


Didn't someone post a "design idea" here once where the cutoff ringing
from the leakage L of a flyback was recovered into the output rail, somehow?

My pic was of a synchronous buck regulator, so there was no leakage
inductance. The big-slam risetime may well have been scope+probe
limited. The spike messed up opamps all over the board.


Would adding some R in series with the high-side bootstrap cap make a
difference?

I doubt it. The big spike sure looks like step recovery in the
substrate diode. The diode takes over from the low-side mosfet for
about 20 ns, then when the high-side fet turns on, the diode fights it
for a few ns, then snaps off.


How about a Schottky in parallel with the low-side FET/Diode?

John Larkin
Guest

Thu Jan 05, 2017 8:30 am   



On 4 Jan 2017 19:32:31 -0800, Winfield Hill <hill_at_rowland.harvard.edu>
wrote:

Quote:
John Larkin wrote...

We tried that, but it only helped some, not enough
to make the chip usable.

I suppose most users simply lived with the noise.


The spikes were making huge offsets in opamps all over the board.

Quote:

A common approach is to slow the switcher's dV/dt
with an aggressive R+C snubber on the switched node.
Yes, it can dramatically reduce the efficiency, but
it quiets the noise. Did you try that? The other
approach is to give up on synchronous converters.


That LM3102 was especially awful. We use other synchronous switchers
without problems.

TPS54302 is a great little part. It switches fast but doesn't have a
gigantic current spike.

https://dl.dropboxusercontent.com/u/53724080/Parts/VREGS/TPS54302/TPS54302_rise.JPG

https://dl.dropboxusercontent.com/u/53724080/Parts/VREGS/TPS54302/DSC06828.JPG

It does spread-spectrum, too, which reduces official EMI.

https://dl.dropboxusercontent.com/u/53724080/Parts/VREGS/TPS54302/DSC06826.JPG



--

John Larkin Highland Technology, Inc

lunatic fringe electronics

John Larkin
Guest

Thu Jan 05, 2017 8:30 am   



On Wed, 04 Jan 2017 20:29:43 -0500, krw_at_notreal.com wrote:

Quote:
On Wed, 04 Jan 2017 10:47:56 -0800, John Larkin
jjlarkinxyxy_at_highlandtechnology.com> wrote:

On Wed, 4 Jan 2017 12:48:08 -0500, bitrex
bitrex_at_de.lete.earthlink.net> wrote:

On 01/04/2017 12:10 PM, John Larkin wrote:
On Wed, 4 Jan 2017 11:58:44 -0500, bitrex
bitrex_at_de.lete.earthlink.net> wrote:

On 01/04/2017 11:52 AM, John Larkin wrote:
On 4 Jan 2017 08:14:07 -0800, Winfield Hill <hill_at_rowland.harvard.edu
wrote:

bill.sloman_at_ieee.org wrote...
John Larkin wrote:
Tim Wescott wrote:

Old hat. To be used with caution, because not all body diodes are
created equal -- for some fast switching regulators you really want
to parallel the FETs with Schottky diodes, as they recover quicker.
A FET with a body diode that has slow recovery can become smoke
pretty quickly.

Some mosfet body diodes make nice step-recovery diodes.

So they can spray really high-frequency noise around the circuit?
Just what you want when you have to meet EMI standards.

All MOSFETs do that, to one degree or another. Some have
added soft-recovery diodes in parallel with the body diode,
which helps, but designers of half-bridge stages, etc.,
are always struggling to prevent body-diode conduction,
or limit its current*time, or slow switching, or snub any
voltage spike from the sudden high dI/dt at snap off.
The problem is worse at maximum switched current. And
most off-line designs can't use added Schottky diodes,
since high-voltage versions have too high an ON voltage.


This is a really classic SRD waveform:

https://dl.dropboxusercontent.com/u/53724080/Parts/LM3102_SwitcherRise.JPG

This switcher is basically unusable. A paralleled external schottky
only helped some.


Didn't someone post a "design idea" here once where the cutoff ringing
from the leakage L of a flyback was recovered into the output rail, somehow?

My pic was of a synchronous buck regulator, so there was no leakage
inductance. The big-slam risetime may well have been scope+probe
limited. The spike messed up opamps all over the board.


Would adding some R in series with the high-side bootstrap cap make a
difference?

I doubt it. The big spike sure looks like step recovery in the
substrate diode. The diode takes over from the low-side mosfet for
about 20 ns, then when the high-side fet turns on, the diode fights it
for a few ns, then snaps off.

How about a Schottky in parallel with the low-side FET/Diode?


We tried that, but it only helped some, not enough to make the chip
usable.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

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

Winfield Hill
Guest

Thu Jan 05, 2017 8:30 am   



John Larkin wrote...
Quote:

We tried that, but it only helped some, not enough
to make the chip usable.


I suppose most users simply lived with the noise.

A common approach is to slow the switcher's dV/dt
with an aggressive R+C snubber on the switched node.
Yes, it can dramatically reduce the efficiency, but
it quiets the noise. Did you try that? The other
approach is to give up on synchronous converters.


--
Thanks,
- Win


Guest

Thu Jan 05, 2017 8:30 am   



On Wed, 04 Jan 2017 17:46:51 -0800, John Larkin
<jjlarkinxyxy_at_highlandtechnology.com> wrote:

Quote:
On Wed, 04 Jan 2017 20:29:43 -0500, krw_at_notreal.com wrote:

On Wed, 04 Jan 2017 10:47:56 -0800, John Larkin
jjlarkinxyxy_at_highlandtechnology.com> wrote:

On Wed, 4 Jan 2017 12:48:08 -0500, bitrex
bitrex_at_de.lete.earthlink.net> wrote:

On 01/04/2017 12:10 PM, John Larkin wrote:
On Wed, 4 Jan 2017 11:58:44 -0500, bitrex
bitrex_at_de.lete.earthlink.net> wrote:

On 01/04/2017 11:52 AM, John Larkin wrote:
On 4 Jan 2017 08:14:07 -0800, Winfield Hill <hill_at_rowland.harvard.edu
wrote:

bill.sloman_at_ieee.org wrote...
John Larkin wrote:
Tim Wescott wrote:

Old hat. To be used with caution, because not all body diodes are
created equal -- for some fast switching regulators you really want
to parallel the FETs with Schottky diodes, as they recover quicker.
A FET with a body diode that has slow recovery can become smoke
pretty quickly.

Some mosfet body diodes make nice step-recovery diodes.

So they can spray really high-frequency noise around the circuit?
Just what you want when you have to meet EMI standards.

All MOSFETs do that, to one degree or another. Some have
added soft-recovery diodes in parallel with the body diode,
which helps, but designers of half-bridge stages, etc.,
are always struggling to prevent body-diode conduction,
or limit its current*time, or slow switching, or snub any
voltage spike from the sudden high dI/dt at snap off.
The problem is worse at maximum switched current. And
most off-line designs can't use added Schottky diodes,
since high-voltage versions have too high an ON voltage.


This is a really classic SRD waveform:

https://dl.dropboxusercontent.com/u/53724080/Parts/LM3102_SwitcherRise.JPG

This switcher is basically unusable. A paralleled external schottky
only helped some.


Didn't someone post a "design idea" here once where the cutoff ringing
from the leakage L of a flyback was recovered into the output rail, somehow?

My pic was of a synchronous buck regulator, so there was no leakage
inductance. The big-slam risetime may well have been scope+probe
limited. The spike messed up opamps all over the board.


Would adding some R in series with the high-side bootstrap cap make a
difference?

I doubt it. The big spike sure looks like step recovery in the
substrate diode. The diode takes over from the low-side mosfet for
about 20 ns, then when the high-side fet turns on, the diode fights it
for a few ns, then snaps off.

How about a Schottky in parallel with the low-side FET/Diode?

We tried that, but it only helped some, not enough to make the chip
usable.


I'd like to have seen your layout, too.

John Larkin
Guest

Fri Jan 06, 2017 4:39 am   



On 4 Jan 2017 19:32:31 -0800, Winfield Hill <hill_at_rowland.harvard.edu>
wrote:

Quote:
John Larkin wrote...

We tried that, but it only helped some, not enough
to make the chip usable.

I suppose most users simply lived with the noise.

A common approach is to slow the switcher's dV/dt
with an aggressive R+C snubber on the switched node.
Yes, it can dramatically reduce the efficiency, but
it quiets the noise. Did you try that? The other
approach is to give up on synchronous converters.


I have seen the same SRD effect in discrete mosfets, in a half-bridge
config where the lower fet substrate diode is conducting and gets hard
reversed when the upper fet turns on. In that case, the spike blew out
the gate on the lower fet.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

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

Winfield Hill
Guest

Fri Jan 06, 2017 5:13 am   



John Larkin wrote...
Quote:

That LM3102 was especially awful. We use other synchronous
switchers without problems.


Too much deadtime, almost 20ns. Unforgivable in an IC
that includes the MOSFET switches with well-understood
delays. Also, sometime a little overlap is good.

Quote:
TPS54302 is a great little part. It switches fast
but doesn't have a gigantic current spike.


Nice.


--
Thanks,
- Win

John Larkin
Guest

Fri Jan 06, 2017 6:45 am   



On 5 Jan 2017 14:13:26 -0800, Winfield Hill <hill_at_rowland.harvard.edu>
wrote:

Quote:
John Larkin wrote...

That LM3102 was especially awful. We use other synchronous
switchers without problems.

Too much deadtime, almost 20ns. Unforgivable in an IC
that includes the MOSFET switches with well-understood
delays. Also, sometime a little overlap is good.

TPS54302 is a great little part. It switches fast
but doesn't have a gigantic current spike.

Nice.


The long dead time is maybe pumping carriers into the substrate diode,
and encouraging it to snap.

The original SRD effect was discovered by Boff accidentally. Some PN
diode just happened to have the right (or maybe wrong) diffusion
profile to make it a step-recovery thing.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

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

Tim Williams
Guest

Fri Jan 06, 2017 3:47 pm   



"John Larkin" <jjlarkinxyxy_at_highlandtechnology.com> wrote in message
news:hqkq6cd2otdv4o05pbercnf39iql5clg6i_at_4ax.com...
Quote:
The current spikes must be huge. There is probably no reasonable way
to fix this one.


"Current spikes"
"fix"

Well, the trivial solution would be supply inductance. With a bit of
damping, or a diode clamp, to absorb the reaction on the other half of the
waveform.

Y'know, snubbers? Those old things? Yeah.

It's too bad that exactly zero percent of monolithic switching regulators
are designed to do that. It would be great to have the switching transistor
power separate from the logic power. But no.

I would imagine most regulators don't have enough PSRR to withstand a
snubber spike* while operating normally.

*The spike is downward first (shoot-through pulls down on the supply), then
upward (whatever was holding the node low (recovery, switching). It's a lot
of high frequency voltage content.

With supply dI/dt snubbing, it's even reasonable -- recommended, even -- to
push deadtime into the negative numbers (i.e., intentional shoot-through).
Real efficiency gains to be had, in suitable situations. Nobody designs
chips with adjustable deadtime though. Not a range like that.

Tim

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

Winfield Hill
Guest

Fri Jan 06, 2017 9:27 pm   



John Larkin wrote...
Quote:

On 5 Jan 2017 14:13:26 -0800, Winfield Hill <hill_at_rowland.harvard.edu
wrote:

John Larkin wrote...

That LM3102 was especially awful. We use other synchronous
switchers without problems.

Too much deadtime, almost 20ns. Unforgivable in an IC
that includes the MOSFET switches with well-understood
delays. Also, sometime a little overlap is good.

TPS54302 is a great little part. It switches fast
but doesn't have a gigantic current spike.

Nice.

The long dead time is maybe pumping carriers into the substrate diode,
and encouraging it to snap.


No maybe about it!

Quote:
The original SRD effect was discovered by Boff accidentally. Some PN
diode just happened to have the right (or maybe wrong) diffusion
profile to make it a step-recovery thing.


--
Thanks,
- Win

bitrex
Guest

Fri Jan 06, 2017 10:19 pm   



On 01/06/2017 03:47 AM, Tim Williams wrote:
Quote:
"John Larkin" <jjlarkinxyxy_at_highlandtechnology.com> wrote in message
news:hqkq6cd2otdv4o05pbercnf39iql5clg6i_at_4ax.com...
The current spikes must be huge. There is probably no reasonable way
to fix this one.

"Current spikes"
"fix"

Well, the trivial solution would be supply inductance. With a bit of
damping, or a diode clamp, to absorb the reaction on the other half of
the waveform.

Y'know, snubbers? Those old things? Yeah.

It's too bad that exactly zero percent of monolithic switching
regulators are designed to do that. It would be great to have the
switching transistor power separate from the logic power. But no.

I would imagine most regulators don't have enough PSRR to withstand a
snubber spike* while operating normally.

*The spike is downward first (shoot-through pulls down on the supply),
then upward (whatever was holding the node low (recovery, switching).
It's a lot of high frequency voltage content.

With supply dI/dt snubbing, it's even reasonable -- recommended, even --
to push deadtime into the negative numbers (i.e., intentional
shoot-through). Real efficiency gains to be had, in suitable
situations. Nobody designs chips with adjustable deadtime though. Not
a range like that.

Tim


The '3102's data sheet brags that it is an "Automotive Grade Product!",
so I wonder if part of the issue was that they were mostly concerned
with ensuring it could supply the (de-rated) max output current at 125 C
rather than...other things...

John Larkin
Guest

Fri Jan 06, 2017 10:52 pm   



On Fri, 6 Jan 2017 02:47:16 -0600, "Tim Williams"
<tiwill_at_seventransistorlabs.com> wrote:

Quote:
"John Larkin" <jjlarkinxyxy_at_highlandtechnology.com> wrote in message
news:hqkq6cd2otdv4o05pbercnf39iql5clg6i_at_4ax.com...
The current spikes must be huge. There is probably no reasonable way
to fix this one.

"Current spikes"
"fix"

Well, the trivial solution would be supply inductance. With a bit of
damping, or a diode clamp, to absorb the reaction on the other half of the
waveform.


The current shoot-through is from the top of the external bypass cap,
into the chip (upper nfet turning on, lower nfet doing the
charge-storage-snap thing) then out the ground pin of the chip back
into the cap. After the snap, there is also the very fast voltage rise
and ring at the output pin. So there are insane dI/dT and dV/dT at
snap time.


Quote:

Y'know, snubbers? Those old things? Yeah.


How would you snub the current spike?

Quote:

It's too bad that exactly zero percent of monolithic switching regulators
are designed to do that. It would be great to have the switching transistor
power separate from the logic power. But no.


I have personally seen one synchronous switcher, and one
discrete-mosfet switcher, that had the step-recovery effect. Most
don't. I think it has to do with the diffusion profile in the
substrate diode.

Snarkiness is not usually a successful design methodology.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

John Larkin
Guest

Sat Jan 07, 2017 2:56 am   



On 6 Jan 2017 06:27:40 -0800, Winfield Hill <hill_at_rowland.harvard.edu>
wrote:

Quote:
John Larkin wrote...

On 5 Jan 2017 14:13:26 -0800, Winfield Hill <hill_at_rowland.harvard.edu
wrote:

John Larkin wrote...

That LM3102 was especially awful. We use other synchronous
switchers without problems.

Too much deadtime, almost 20ns. Unforgivable in an IC
that includes the MOSFET switches with well-understood
delays. Also, sometime a little overlap is good.

TPS54302 is a great little part. It switches fast
but doesn't have a gigantic current spike.

Nice.

The long dead time is maybe pumping carriers into the substrate diode,
and encouraging it to snap.

No maybe about it!

The original SRD effect was discovered by Boff accidentally. Some PN
diode just happened to have the right (or maybe wrong) diffusion
profile to make it a step-recovery thing.


We once built some 2KV, nanosecond-width, 100 KHz, water-cooled pulse
generators, using drift step-recovery diodes. Nobody that I know of
fabricates DSRDs on purpose, so one tests other devices hoping to find
one with, accidentally, the right doping profile. Our favorite was the
c-b junction of a high-voltage NPN power transistor.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

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


Guest

Sat Jan 07, 2017 8:30 am   



On Fri, 6 Jan 2017 10:19:13 -0500, bitrex
<bitrex_at_de.lete.earthlink.net> wrote:

Quote:
On 01/06/2017 03:47 AM, Tim Williams wrote:
"John Larkin" <jjlarkinxyxy_at_highlandtechnology.com> wrote in message
news:hqkq6cd2otdv4o05pbercnf39iql5clg6i_at_4ax.com...
The current spikes must be huge. There is probably no reasonable way
to fix this one.

"Current spikes"
"fix"

Well, the trivial solution would be supply inductance. With a bit of
damping, or a diode clamp, to absorb the reaction on the other half of
the waveform.

Y'know, snubbers? Those old things? Yeah.

It's too bad that exactly zero percent of monolithic switching
regulators are designed to do that. It would be great to have the
switching transistor power separate from the logic power. But no.

I would imagine most regulators don't have enough PSRR to withstand a
snubber spike* while operating normally.

*The spike is downward first (shoot-through pulls down on the supply),
then upward (whatever was holding the node low (recovery, switching).
It's a lot of high frequency voltage content.

With supply dI/dt snubbing, it's even reasonable -- recommended, even --
to push deadtime into the negative numbers (i.e., intentional
shoot-through). Real efficiency gains to be had, in suitable
situations. Nobody designs chips with adjustable deadtime though. Not
a range like that.

Tim


The '3102's data sheet brags that it is an "Automotive Grade Product!",
so I wonder if part of the issue was that they were mostly concerned
with ensuring it could supply the (de-rated) max output current at 125 C
rather than...other things...


Generally a 150C junction but that's not the problem. TI makes a
*lot* of automotive rated parts.

Goto page Previous  1, 2, 3  Next

elektroda.net NewsGroups Forum Index - Electronics Design - "Use MOSFET body didoes"

Ask a question - edaboard.com

Arabic versionBulgarian versionCatalan versionCzech versionDanish versionGerman versionGreek versionEnglish versionSpanish versionFinnish versionFrench versionHindi versionCroatian versionIndonesian versionItalian versionHebrew versionJapanese versionKorean versionLithuanian versionLatvian versionDutch versionNorwegian versionPolish versionPortuguese versionRomanian versionRussian versionSlovak versionSlovenian versionSerbian versionSwedish versionTagalog versionUkrainian versionVietnamese versionChinese version
RTV map EDAboard.com map News map EDAboard.eu map EDAboard.de map EDAboard.co.uk map