Filter cap choice for ultra low ripple

[John Fields]

On Sun, 9 Aug 2015 14:27:50 +0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

Gentlemen,

What is currently the best choice of filter capacitor for the PSU section
of a high-end signal generator the output of which (obviously) really
needs to be as ripple-free as humanly possible?
Just in *GENERAL* terms, please guys; don't want to get bogged down in
specifics and calculations at this stage, please!

tia,
cd

---
In a linear supply, since the ripple will be squashed to nice, clean
DC by the regulator following the reservoir cap(s), the type of
capacitor doesn't matter as much as its value, since that'll determine
the headroom available for the regulator to work under.

John Fields

 

[Clifford Heath]

On 12/08/15 06:18, DecadentLinuxUserNumeroUno wrote:

On Tue, 11 Aug 2015 20:04:19 +0000 (UTC), Cursitor Doom
curd@notformail.com> Gave us:

On Mon, 10 Aug 2015 04:36:39 -0400, DecadentLinuxUserNumeroUno wrote:

You obviously do not know how to read a thread.

Yeah, and YOU obviously don't know how to read AT ALL.

You fucking retard. Phil DID make exactly the suggestion the
oblivious twit missed.

My news server did not propagate Phil's post. I have since been able to
find it on Google Gropes. So f**k you.

 

[John S]

On 8/11/2015 7:22 PM, Clifford Heath wrote:


>So f**k you.

Hey, Cliff. Don't do that. He is diseased both physically and mentally.

 

[DecadentLinuxUserNumeroUn]

On Tue, 11 Aug 2015 16:56:30 -0500, John Fields
<jfields@austininstruments.com> Gave us:

On Sun, 9 Aug 2015 14:27:50 +0000 (UTC), Cursitor Doom
curd@notformail.com> wrote:

Gentlemen,

What is currently the best choice of filter capacitor for the PSU section
of a high-end signal generator the output of which (obviously) really
needs to be as ripple-free as humanly possible?
Just in *GENERAL* terms, please guys; don't want to get bogged down in
specifics and calculations at this stage, please!

tia,
cd

---
In a linear supply, since the ripple will be squashed to nice, clean
DC by the regulator following the reservoir cap(s), the type of
capacitor doesn't matter as much as its value, since that'll determine
the headroom available for the regulator to work under.

John Fields

Yes. One must match the stored energy with the expected loading.

With HV supplies this is especially true. The multiplier caps scale
up with the load demands as does the final storage bank.

 

[DecadentLinuxUserNumeroUn]

On Tue, 11 Aug 2015 21:24:20 +0000 (UTC), Cursitor Doom
<curd@notformail.com> Gave us:

><shrug> Every group has one....

shrug... you're an idiot. Good job, shrug boy.

 

[DecadentLinuxUserNumeroUn]

On Tue, 11 Aug 2015 19:54:17 -0500, John S <Sophi.2@invalid.org> Gave
us:

On 8/11/2015 7:22 PM, Clifford Heath wrote:


So f**k you.

Hey, Cliff. Don't do that. He is diseased both physically and mentally.

Your lack of grasp or knowledge of common colloquialisms pretty much
proves that you are the mentally diseased one on this bus, jackoff.

 

[DecadentLinuxUserNumeroUn]

On Wed, 12 Aug 2015 10:22:38 +1000, Clifford Heath <no.spam@please.net>
Gave us:

On 12/08/15 06:18, DecadentLinuxUserNumeroUno wrote:
On Tue, 11 Aug 2015 20:04:19 +0000 (UTC), Cursitor Doom
curd@notformail.com> Gave us:

On Mon, 10 Aug 2015 04:36:39 -0400, DecadentLinuxUserNumeroUno wrote:

You obviously do not know how to read a thread.

Yeah, and YOU obviously don't know how to read AT ALL.

You fucking retard. Phil DID make exactly the suggestion the
oblivious twit missed.

My news server did not propagate Phil's post. I have since been able to
find it on Google Gropes. So f**k you.

Reminds me of a line the General said in "War Games" to "Mr. Kitrich"
about his computer simulator and taking humans "out of the loop".

"I have concluded that your News service sucks." pretty much covers
it. So fuck you.

 

[DemonicTubes]

On Thursday, August 13, 2015 at 7:43:21 AM UTC-6, DecadentLinuxUserNumeroUno wrote:

On Thu, 13 Aug 2015 08:44:20 -0400, JW <none@dev.null> Gave us:


*yawn*

Says the utter idiot who refers to someone as "AW".

Seems it fits you more so than anyone in your pathetic dreams.

Wiltron 6669A-40 A $50k 40GHz sweep generator way outside your realm.

You lose... again.

You just paid $50k for a 6669A? wow...get your money back, I'll sell you two for half that price X-P

 

[JW]

On Tue, 11 Aug 2015 16:10:06 -0400 DecadentLinuxUserNumeroUno
<DLU1@DecadentLinuxUser.org> wrote in Message id:
<qilksaps7feig9n0qbocabdjr8g47mej6s@4ax.com>:

On Tue, 11 Aug 2015 14:47:45 -0400, JW <none@dev.null> Gave us:

Care to address this particular piece of idiocy, or would you like to
admit you were wrong, AW?

Just bought a 40GHz generator last week. Guess what type of supply it
has.

I don't guess. Why don't you tell me what the model number is?
I'll be happy to admit I was wrong, unlike you.

Oh, that's right... you are too stupid to work with anything in that
frequency range.

Just WTF do you think 'high end' means, chump?

*yawn*

 

[DecadentLinuxUserNumeroUn]

On Thu, 13 Aug 2015 08:44:20 -0400, JW <none@dev.null> Gave us:

*yawn*

Says the utter idiot who refers to someone as "AW".

Seems it fits you more so than anyone in your pathetic dreams.

Wiltron 6669A-40 A $50k 40GHz sweep generator way outside your realm.

You lose... again.

 

[DecadentLinuxUserNumeroUn]

On Thu, 13 Aug 2015 09:36:11 -0700 (PDT), DemonicTubes
<tlackie@gmail.com> Gave us:

On Thursday, August 13, 2015 at 7:43:21 AM UTC-6, DecadentLinuxUserNumeroUno wrote:
On Thu, 13 Aug 2015 08:44:20 -0400, JW <none@dev.null> Gave us:


*yawn*

Says the utter idiot who refers to someone as "AW".

Seems it fits you more so than anyone in your pathetic dreams.

Wiltron 6669A-40 A $50k 40GHz sweep generator way outside your realm.

You lose... again.

You just paid $50k for a 6669A? wow...get your money back, I'll sell you two for half that price X-P

Show me where I said I paid that much for it, you retarded piece of
shit.

Stay the fuck out of conversations you are fucking clueless about and
have no business sticking your shit headedness into, shithead.

Your mother should be in prison for failing to flush you the moment
the retarded slut shat you.

 

[John S]

On 8/13/2015 11:52 AM, DecadentLinuxUserNumeroUno wrote:

You just paid $50k for a 6669A? wow...get your money back, I'll
sell you two for half that price X-P

Show me where I said I paid that much for it, you retarded piece of
shit.

Ok, so maybe you stole it. Or, bought a reject.

 

On Monday, 10 August 2015 11:11:29 UTC+1, upsid...@downunder.com wrote:

On Sun, 09 Aug 2015 11:34:46 -0700, John Larkin
jlarkin@highlandtechnology.com> wrote:

On Sun, 9 Aug 2015 14:27:50 +0000 (UTC), Cursitor Doom
curd@notformail.com> wrote:

Gentlemen,

What is currently the best choice of filter capacitor for the PSU section
of a high-end signal generator the output of which (obviously) really
needs to be as ripple-free as humanly possible?
Just in *GENERAL* terms, please guys; don't want to get bogged down in
specifics and calculations at this stage, please!

tia,
cd

Generally one buys enough filter cap to get the ripple down to maybe 1
volt p-p, then adds a regulator to stabilize the voltage and greatly
reduce ripple. A second regulator or c-multiplier can be used to
further reduce noise and ripple, to power the more sensitive signal
generator stages.

Since the same charge needs to be transferred during the mains half
cycle, using huge storage capacitors will result a smaller conduction
angle and hence large peak currents. High peak current will cause many
problems, including strong stray magnetic fields that can be coupled
into various circuitry.

When dealing with 50/60Hz transformers, tf resistance limits reservoir charge current. Then huge caps just give reduced ripple.

If low noise and not maximum efficiency is the goal, use moderate size
storage capacitors with possibly additional series resistance and the
peak currents remain small. Of course, this requires higher secondary
voltages, but the increased peak rectified voltage can be limited with
any conventional regulator.

NT

 

[DecadentLinuxUserNumeroUn]

On Thu, 13 Aug 2015 12:24:16 -0500, John S <Sophi.2@invalid.org> Gave
us:

On 8/13/2015 11:52 AM, DecadentLinuxUserNumeroUno wrote:

You just paid $50k for a 6669A? wow...get your money back, I'll
sell you two for half that price X-P

Show me where I said I paid that much for it, you retarded piece of
shit.

Ok, so maybe you stole it. Or, bought a reject.

If you actually had any functioning neurons, retard boy, you would
note that the device is a very old one and as such is not likely to be
on the market at its originally sold price. I don't expect you or a
retard like you to understand.

 

[krw]

On Thu, 13 Aug 2015 12:24:16 -0500, John S <Sophi.2@invalid.org>
wrote:

On 8/13/2015 11:52 AM, DecadentLinuxUserNumeroUno wrote:

You just paid $50k for a 6669A? wow...get your money back, I'll
sell you two for half that price X-P

Show me where I said I paid that much for it, you retarded piece of
shit.

Ok, so maybe you stole it. Or, bought a reject.

DimBulb bought himself?

 

On Thu, 13 Aug 2015 15:44:29 -0700 (PDT), tabbypurr@gmail.com wrote:

On Monday, 10 August 2015 11:11:29 UTC+1, upsid...@downunder.com wrote:
On Sun, 09 Aug 2015 11:34:46 -0700, John Larkin
jlarkin@highlandtechnology.com> wrote:

On Sun, 9 Aug 2015 14:27:50 +0000 (UTC), Cursitor Doom
curd@notformail.com> wrote:

Gentlemen,

What is currently the best choice of filter capacitor for the PSU section
of a high-end signal generator the output of which (obviously) really
needs to be as ripple-free as humanly possible?
Just in *GENERAL* terms, please guys; don't want to get bogged down in
specifics and calculations at this stage, please!

tia,
cd

Generally one buys enough filter cap to get the ripple down to maybe 1
volt p-p, then adds a regulator to stabilize the voltage and greatly
reduce ripple. A second regulator or c-multiplier can be used to
further reduce noise and ripple, to power the more sensitive signal
generator stages.

Since the same charge needs to be transferred during the mains half
cycle, using huge storage capacitors will result a smaller conduction
angle and hence large peak currents. High peak current will cause many
problems, including strong stray magnetic fields that can be coupled
into various circuitry.

When dealing with 50/60Hz transformers, tf resistance limits reservoir charge current. Then huge caps just give reduced ripple.

This is true for small transformers, but with any larger transformers
(e.g. those requiring soft starter to protect house fuses) will see
the feeding network impedance.

In addition to a large current distortion, there will also be some
voltage distortion due to flattening of the voltage waveform.

 

[JW]

On Thu, 13 Aug 2015 09:42:44 -0400 DecadentLinuxUserNumeroUno
<DLU1@DecadentLinuxUser.org> wrote in Message id:
<gm7psa9e6gcjae0edul6r6tpbfolu03m1j@4ax.com>:

On Thu, 13 Aug 2015 08:44:20 -0400, JW <none@dev.null> Gave us:


*yawn*

Says the utter idiot who refers to someone as "AW".

It's easier than typing AlwaysWrong, AW.

Seems it fits you more so than anyone in your pathetic dreams.

Wiltron 6669A-40 A $50k 40GHz sweep generator way outside your realm.

In 1991 it may have been.
http://www.ebay.com/itm/Wiltron-6669A-40-40Ghz-Sweeper-Option-003-/111733810468?hash=item1a03daa524&nma=true&si=dvqofCNLX4lPgItmd0XSxR9XUaQ%253D&orig_cvip=true&rt=nc&_trksid=p2047675.l2557

Sold for less than $1K. Pocket change. I spend 5 times that on Ebay every
month. Must be a reason why they're going so cheap, maybe they're a POS?

Anyway, I've never worked on one.

I usually work on HP/Agilent generators, although I repaired a Marconi
fairly recently. I've never seen one with a linear power supply.

> You lose... again.

Take a look in the service manual that covers the 66XX series. Module A13
is referenced as a switching power supply.

 

[JW]

On Thu, 13 Aug 2015 12:24:16 -0500 John S <Sophi.2@invalid.org> wrote in
Message id: <mqijoo$hlu$1@dont-email.me>:

On 8/13/2015 11:52 AM, DecadentLinuxUserNumeroUno wrote:

You just paid $50k for a 6669A? wow...get your money back, I'll
sell you two for half that price X-P

Show me where I said I paid that much for it, you retarded piece of
shit.

Ok, so maybe you stole it. Or, bought a reject.

For some reason, they're pretty cheap on Ebay. Even AW could probably
afford one.

 

[DecadentLinuxUserNumeroUn]

On Fri, 14 Aug 2015 07:41:49 -0400, JW <none@dev.null> Gave us:

On Thu, 13 Aug 2015 12:24:16 -0500 John S <Sophi.2@invalid.org> wrote in
Message id: <mqijoo$hlu$1@dont-email.me>:

On 8/13/2015 11:52 AM, DecadentLinuxUserNumeroUno wrote:

You just paid $50k for a 6669A? wow...get your money back, I'll
sell you two for half that price X-P

Show me where I said I paid that much for it, you retarded piece of
shit.

Ok, so maybe you stole it. Or, bought a reject.

For some reason, they're pretty cheap on Ebay.

Those are the 20GHz low end models. Find the "option 3" units which
do 40GHz, and watch what they go for, idiot.

 

[John Larkin]

On Fri, 14 Aug 2015 07:39:32 -0400, JW <none@dev.null> wrote:

On Thu, 13 Aug 2015 09:42:44 -0400 DecadentLinuxUserNumeroUno
DLU1@DecadentLinuxUser.org> wrote in Message id:
gm7psa9e6gcjae0edul6r6tpbfolu03m1j@4ax.com>:

On Thu, 13 Aug 2015 08:44:20 -0400, JW <none@dev.null> Gave us:


*yawn*

Says the utter idiot who refers to someone as "AW".

It's easier than typing AlwaysWrong, AW.

Seems it fits you more so than anyone in your pathetic dreams.

Wiltron 6669A-40 A $50k 40GHz sweep generator way outside your realm.

In 1991 it may have been.
http://www.ebay.com/itm/Wiltron-6669A-40-40Ghz-Sweeper-Option-003-/111733810468?hash=item1a03daa524&nma=true&si=dvqofCNLX4lPgItmd0XSxR9XUaQ%253D&orig_cvip=true&rt=nc&_trksid=p2047675.l2557

Sold for less than $1K. Pocket change. I spend 5 times that on Ebay every
month. Must be a reason why they're going so cheap, maybe they're a POS?

Anyway, I've never worked on one.

I usually work on HP/Agilent generators, although I repaired a Marconi
fairly recently. I've never seen one with a linear power supply.

You lose... again.

Take a look in the service manual that covers the 66XX series. Module A13
is referenced as a switching power supply.

When AW says "you lose" it means that he is wrong yet again.


--

John Larkin Highland Technology, Inc
lunatic fringe electronics

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