How high a frequency does a bridge rectifier handle ?

[Daku]

Could some electronics guru shed some light on
the following ? Normally, when we think of a bridge
rectifier, input signal frequencies of 50/60 Hz come
to mind. How high could the input frequencies rise?
Is there some limit ? Thanks in advance for your
help.

[P E Schoen]

"Daku" wrote in message
news:df5f21ab-ecbe-4623...@i10g2000pbl.googlegroups.com...

I'm using an ordinary bridge for 8 kHz, but it seems to be rather
inefficient. So I searched for Schottky bridges, and found some that should
be a lot better:
http://www.mouser.com/ProductDetail/Central-Semiconductor/CBRHDSH2-100-TR13/?qs=sGAEpiMZZMutncetXCRkfsXyqqpaPA3Y

No specs on frequency or recovery time.

If 28V and 200mA are OK, here is a little one SOT-143:
http://www.newark.com/infineon/bas4002a-rpp/schottky-diode-array/dp/68R0434

But also no frequency specs.

A comparable product from TI shows 15-30 nSec recovery. That's pretty
fast...

Paul

[Phil Allison]

"Daku"

> Could some electronics guru shed some light on
> the following ? Normally, when we think of a bridge
> rectifier, input signal frequencies of 50/60 Hz come
> to mind. How high could the input frequencies rise?

** Ordinary power diodes get hotter as the frequency rises above about 1kHz.

> Is there some limit ?

** Package temps over 100C are not wise.

At frequencies of 10kHz or more, folk generally go for bridges made from
high speed diodes.



... Phil

[Daku]

On Jan 18, 3:33 am, "P E Schoen" <p...@peschoen.com> wrote:
> "Daku"  wrote in message
>
> news:df5f21ab-ecbe-4623...@i10g2000pbl.googlegroups.com...
>
> > Could some electronics guru shed some light
> > on the following ? Normally, when we think of a
> > bridge rectifier, input signal frequencies of
> > 50/60 Hz come to mind. How high could the
> > input frequencies rise?
> > Is there some limit ? Thanks in advance
> > for your help.
>
> I'm using an ordinary bridge for 8 kHz, but it seems to be rather
> inefficient. So I searched for Schottky bridges, and found some that should

> be a lot better:http://www.mouser.com/ProductDetail/Central-Semiconductor/CBRHDSH2-10...

>
> No specs on frequency or recovery time.
>

> If 28V and 200mA are OK, here is a little one SOT-143:http://www.newark.com/infineon/bas4002a-rpp/schottky-diode-array/dp/6...

>
> But also no frequency specs.
>
> A comparable product from TI shows 15-30 nSec recovery. That's pretty
> fast...
>
> Paul

Actually, Vishay claims on its Web site that it has < 10 nanosecond
reverse recovery time diodes.

[mike]

A bridge rectifier is typically a set of four diodes of
arbitrary specification. So, no, there ain't no limit
on the bridge...just the diodes...read the spec.
You can make one hell of a fast diode bridge out of 1N21 diodes.
And that's 60 or more years old technology.

If you mean, bridge rectifier as a packaged diode quad
optimized for 60 Hz., then, yes, there are issues that
depend on the individual diodes in the package...'cause
they're optimized for cost when used at 60 Hz....read the spec.


You're working the problem backwards. Decide what frequencies
you need, voltages and currents, how much loss you can tolerate, how
much heat you
can get rid of, how much EMI you can tolerate, whether your input
is sine wave or not. Then figger out the specs of the bridge you need.
Then RTFM to determine whether you can afford the diodes that meet your
spec. Junction capacitance and reverse recovery time are two relevant
specs.

No, it's not a simple question.

[George Herold]

Double balanced mixers use diodes. (in a ring)
I've got a low frequncy one from mini-circits that's good to 200MHz.
I'm sure they go higher. (But maybe stop using diodes?)

George H.

[John Larkin]

On Wed, 18 Jan 2012 00:03:55 -0800 (PST), Daku <daku...@gmail.com>
wrote:

This is running at about 90 KHz. Works fine.

ftp://jjlarkin.lmi.net/ESM_power.pdf

D2 and D3 are little surface-mount schottky bridge rectifiers.

There's no fundamental limit on bridge frequency. Schottky diodes work
into the GHz.

John

(Note that this is a real, working schematic, with parts values.)

[John Larkin]

Some sampling scopes used 4-diode bridge samplers. The Tek 7S14 did,
and got about 2 GHz bandwidth using a potted bridge rectifier. Weird
design.

John

[cassiope]

On Jan 18, 8:24 am, John Larkin

<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Wed, 18 Jan 2012 06:46:35 -0800 (PST), George Herold
>

Of course, this "potted bridge rectifier" was made with some
relatively exotic
(I think Tek-made) Schottky diodes... not an off-the-shelf part.

[Paul Hovnanian P.E.]

Like others have posted, the bridge architecture has no inherent upper
frequency limits. The freq. limit would be based on the parasitics of the
individual bridge diodes, plus component layout at higher frequencies.

Some of those big high current, high reverse voltage rectifiers might only
be good for 10s or 100s of kHz.

--
Paul Hovnanian mailto:Pa...@Hovnanian.com
------------------------------------------------------------------
What if no one ever asked a hypothetical question?

[Dave Platt]

>A bridge rectifier is typically a set of four diodes of
>arbitrary specification. So, no, there ain't no limit
>on the bridge...just the diodes...read the spec.
>You can make one hell of a fast diode bridge out of 1N21 diodes.
>And that's 60 or more years old technology.

Yup. A full-wave diode rectifying bridge is a close topological
cousin to the diode ring used in double-balanced mixers, and these are
frequently used at switching (local-oscillator) frequencies up in to
the high MHz range.

It really all depends on the diodes.

>If you mean, bridge rectifier as a packaged diode quad
>optimized for 60 Hz., then, yes, there are issues that
>depend on the individual diodes in the package...'cause
>they're optimized for cost when used at 60 Hz....read the spec.

A lot of the common power-rectifier diodes (e.g. 1N400x family) have a
PIN structure, and are relatively slow to shut off in the face of
reverse bias... they've got a long carrier lifetime. I suspect that
many common power-rectification bridges use diodes of a similar
nature.

You can buy power-rectification bridges made out of HEXFRED or similar
fast-recovery diodes (see http://www.partsconnexion.com/PDF/l422.pdf
for example) or build your own from diodes of this sort.

--
Dave Platt <dpl...@radagast.org> AE6EO
Friends of Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!

[P E Schoen]

"Paul Hovnanian P.E." wrote in message
news:h5KdnWKtaPqOiIrS...@posted.isomediainc...

> Like others have posted, the bridge architecture has no inherent upper
> frequency limits. The freq. limit would be based on the parasitics of the
> individual bridge diodes, plus component layout at higher frequencies.

> Some of those big high current, high reverse voltage rectifiers might
> only be good for 10s or 100s of kHz.

Here is a link showing how the recovery time can be measured. In particular,
it shows that a common 1N4004 diode has a t(rr) of 1.2 uSec. So it would be
unusable at 1 MHz, perhaps have 10% switching losses at 100 kHz, but should
be fine for most audio purposes up to 20-30 kHz or so. Larger bridge
rectifiers over 1 amp are likely to be worse, but it is difficult to find
the specs since they are usually optimized for 50/60/400 Hz where switching
time is largely irrelevant.
http://www.avtechpulse.com/appnote/techbrief9/

I was surprised to find Schottky bridges that handle 2 amps and 100V for
less than a buck:
http://www.mouser.com/ProductDetail/Central-Semiconductor/CBRHDSH2-100-TR13/?qs=sGAEpiMZZMutncetXCRkfsXyqqpaPA3Y
http://www.centralsemi.com/PDFs/products/CBRHDSH2-100.PDF
Capacitance is 250 pF.

Switching time is not specified, but is likely to be about as good as the
1N5818. This is also not specified, because the usefulness is more a factor
of junction capacitance, which in this case (20-200pF) limits use to 1 or 2
MHz:
http://www.onsemi.com/pub_link/Collateral/1N5817-D.PDF

Paul

[o...@grrr.id.au]

Sadly from an ftp site I cannot access :( Not my end 'cos
I downloaded >2GB from mirror.aarnet.edu.au last week using
wget for windows on Win7.

But there's a workaround :o) Search for the file in google,
select 'quick view', then file, Download original, if you
like to view in local .pdf viewer.

But I didn't see the 555! ;)

Grant.

[Jim Yanik]

o...@grrr.id.au wrote in news:uuoeh71ekjio6t9fv...@4ax.com:

I suspect junction capacitance of the schottky diode may have some effect
on operation.

Power diodes are different than small-signal diodes.

--
Jim Yanik
jyanik
at
localnet
dot com

[John Larkin]

On Wed, 18 Jan 2012 20:00:10 -0600, Jim Yanik <jya...@abuse.gov>
wrote:

These are cute, only 100 pF or so.

61.222.192.61/mccsemi/up_pdf/MB12S-MB110S(MBS-1).pdf

John