Searching \ for '[EE] tapping power from PC supply' 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/power.htm?key=power
Search entire site for: 'tapping power from PC supply'.

Exact match. Not showing close matches.
PICList Thread
'[EE] tapping power from PC supply'
2004\09\24@151352 by Peter Johansson

flavicon
face

Appologies in advance for a very basic EE question...

What issues should I be aware of if I want to tap power directly from
the my PC's power supply to drive my breadboard(s)?  The PC provides a
very convenient source of regulated +5v and +12v, and is easily
obtained through one of the 4-pin drive connectors.  A cable could be
very easily made from sacrificed fan taps and 120v line power cable
(16-18 ga.)

I imagine the distance (approx 5 feet) from the regulator might have
some issues, but are they serious, and can they be overcome with a
large decoupling capacitor for the entire breadboard?  If so what
values would be apropriate?

-p.

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@154337 by Bob Blick

face picon face

> What issues should I be aware of if I want to tap power directly from
> the my PC's power supply to drive my breadboard(s)?  The PC provides a
> very convenient source of regulated +5v and +12v, and is easily
> obtained through one of the 4-pin drive connectors.  A cable could be
> very easily made from sacrificed fan taps and 120v line power cable
> (16-18 ga.)

The biggest problem will likely be when you accidentally short circuit it,
and it reboots yout computer.

Much better to get a separate power supply for prototyping.

Cheerful regards,

Bob


_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@154901 by Rob Young

picon face
>
> Appologies in advance for a very basic EE question...
>
> What issues should I be aware of if I want to tap power directly from
> the my PC's power supply to drive my breadboard(s)?  The PC provides a
> very convenient source of regulated +5v and +12v, and is easily
> obtained through one of the 4-pin drive connectors.  A cable could be
> very easily made from sacrificed fan taps and 120v line power cable
> (16-18 ga.)
>
> I imagine the distance (approx 5 feet) from the regulator might have
> some issues, but are they serious, and can they be overcome with a
> large decoupling capacitor for the entire breadboard?  If so what
> values would be apropriate?
>
> -p.

Many PC switching supplies require a minimum load on +5V, +12V or sometimes
both for the switcher to start "switching".  The information is sometimes
found on a sticker on the supply or if you are lucky and it is a brand-name
supply you can find a datasheet with Google.

The simple way to make the minimum load is with large power resistors across
the appropriate rails.  Just remember P=I*I*R or P=(V*V)/R and size the
resistors appropriately.  Also, don't touch them either because they will be
hot.

Also, most PC supplies aren't that well regulated.  +/-5% to +/-10% and
possibly worse.  You will also get a lot of switching noise and its
amplitude and possibly its frequency will depend on how you load the supply.

I would recommend that you keep your eyes open for a good deal on a nice
linear bench supply (+5V @ 1 to 5A, +15V @ 1A adjustable and -15V @ 1A
adjustable is nice, plus some kind of current limiting control is wonderful)
but in the short term you can use a PC supply.  Just remember it won't be
the greatest and you may be able to toast a bagel on your dummy load
resistors...

Rob Young
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@165755 by Wouter van Ooijen

face picon face
> > What issues should I be aware of if I want to tap power
> directly from
> > the my PC's power supply to drive my breadboard(s)?

*if* you want to do this take the 12V and run it trough an 78(L)05. Or
take the USB.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu



_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@182840 by Peter Johansson

flavicon
face

I wrote:

> >What issues should I be aware of if I want to tap power directly from
> >the my PC's power supply to drive my breadboard(s)?

Rob Young replies:

> Many PC switching supplies require a minimum load on +5V, +12V or sometimes
> both for the switcher to start "switching".  The information is sometimes
> found on a sticker on the supply or if you are lucky and it is a brand-name
> supply you can find a datasheet with Google.

I'm aware of this, but it shouldn't be an issue since I'm already
driving the rest of the computer with the same supply.

> Also, most PC supplies aren't that well regulated.  +/-5% to +/-10% and
> possibly worse.  You will also get a lot of switching noise and its
> amplitude and possibly its frequency will depend on how you load the supply.

I measured the +5v line and I'm getting 5.09V so in this particular
case I think I'll be ok.  I would think that even a relatively small
capacitor should be able to filter out any ripple.

Bob Blick hit an issue that I completely missed: a short on my
breadboard will likely take the whole PC down.  That's probably enough
reason not to do this right there without some current-limiting facility.

> I would recommend that you keep your eyes open for a good deal on a nice
> linear bench supply (+5V @ 1 to 5A, +15V @ 1A adjustable and -15V @ 1A
> adjustable is nice, plus some kind of current limiting control is
> wonderful) but in the short term you can use a PC supply.  Just remember it
> won't be the greatest and you may be able to toast a bagel on your dummy
> load resistors...

Yes, a benchtop power supply with multiple fixed and variable
voltages, adjustable current limiting, and current/voltage readout on
every channel sure would be nice, but something like that is *way* out
of my price range... or would it?

The biggest cost here would be the meters, but I'd be willing to bet
that I could use a PIC for all the voltage and current monitoring and
then send the output via serial to a PC.  Since the PIC is monitoring
current, it could probably be used as a cheap and easy current
limiter.  (Would this be fast enough?)  Variable output voltages might
even be dialed in right from the PC.  Hmmm... ponder, ponder...

If I were to build such a device, would there be any downside to using
a switching supply as the unregulated input?  That is, use the 12V
line from the PC supply to generate a clean, nicely regulated supply
4-11V, and the 5V line for voltages below 4V.  It would be silly to
purchase a PC power supply for such a purpose, but then again it would
be equally silly to purchase a big transformer, bridge, and filter
caps if I don't *really* need them.  I'd probably use a 6V halogen
bulb (or a pair of 6V and 12V bulbs, if necessary) as my dummy load,
mounted on a gooseneck as a worklight.

Thoughts?  Comments?  Suggestions?

-p.
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@192505 by Rob Young

picon face
-- snip ---

> Yes, a benchtop power supply with multiple fixed and variable
> voltages, adjustable current limiting, and current/voltage readout on
> every channel sure would be nice, but something like that is *way* out
> of my price range... or would it?

-- snip --

>
> Thoughts?  Comments?  Suggestions?
>

There was an issue of Circuit Cellar with plans for a multi-channel current
limited supply.  A few years back.  Google may help or maybe the Circuit
Cellar web site has a search feature.

As to "price range", consider how much you would be using it.  Good quality
tools are ALWAYS worth what you pay for them.  You could buy a pretty nice
B&K brand supply or some other non-name brand new for a reasonable amount.
Used HP & Kepco supplies appear on Ebay regularly.


Rob Young
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@194151 by Mike Hord

picon face
> There was an issue of Circuit Cellar with plans for a multi-channel current
> limited supply.  A few years back.  Google may help or maybe the Circuit
> Cellar web site has a search feature.

February 2002, Issue 139

"A Tracking Lab Power Supply" by R. Lacoste.

A good place to start, actually.  He starts from 220 instead of 110, but that's
easily remedied by a different transformer.

Two tracking 0-20V 3A sections and 1 0-20V 2A section.

Of course, the heretic didn't put a PIC in anywhere.  You'll have to add that,
as well as the Wifi and IEEE.1394.  ;-)

Mike H.
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\24@235619 by William Chops Westfield

face picon face
On Sep 24, 2004, at 3:28 PM, Peter Johansson wrote:

> Yes, a benchtop power supply with multiple fixed and variable
> voltages, adjustable current limiting, and current/voltage readout on
> every channel sure would be nice, but something like that is *way* out
> of my price range... or would it?
>
> The biggest cost here would be the meters, but I'd be willing to bet
> that I could use a PIC for all the voltage and current monitoring and
> then send the output via serial to a PC.

Or a cheap surplus LCD.  I've been contemplating a project like this;
but it's unlikely that it'll creep up on my to-do list - I already
have  a lab power supply.

What IS your budget?  Cheap VOMs are well under $10, so a wall-wart
(free)
a couple 317s (cheap) and two dedicated meters would run you well under
$50.  But then, commercial units start around $60

    http://www.mpja.com/productview.asp?product=14600+PS

I like your idea of using the PC as a display.  I can see converting an
old (ie free or near free) PC into a whole electronics experimentor's
dream.  Power supply  (0-12V) mounted and accessible from a drive bay,
sound card used as a scope and signal generator, parallel port based
logic analyzer, and of course device programmers "everywhere", plus
assorted IDEs and layout SW.  Most of this already exists; putting
it together into an extensible package wouldn't be TOO hard (sorta
the "winavr" of hardware?)  Hmm...

BillW

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\25@013425 by Peter Johansson

flavicon
face
Mike Hord writes:

> > There was an issue of Circuit Cellar with plans for a multi-channel current
> > limited supply.  A few years back.  Google may help or maybe the Circuit
> > Cellar web site has a search feature.
>
> February 2002, Issue 139
>
> "A Tracking Lab Power Supply" by R. Lacoste.
>
> A good place to start, actually.  He starts from 220 instead of 110, but that's
> easily remedied by a different transformer.
>
> Two tracking 0-20V 3A sections and 1 0-20V 2A section.
>
> Of course, the heretic didn't put a PIC in anywhere.  You'll have to add that,
> as well as the Wifi and IEEE.1394.  ;-)

Then you should look at the follow-on to that article which appeared
in the December issue of that year.  Although it uses an Amtel MCU and
not a PIC.  ;-)

http://www.circuitcellar.com/library/print/1202/millier149/

This circuit is really much more what I'm looking for: multiple
outputs all at relatively low current.  While in this case the author
chose to replace multiple digital displays and multi-turn pots of
Lacoste's design with a rotary switch, MCU and single LCD, my idea
would be to take it a step further removing all display from the front
panel and moving it to a window on my PC.  It saves money, but that's
really only a small part.  The supply could then then can be small an
unobtrusive, or turned into a little piece of desk art, as I think I
might do.  I purged a lot of junk a while back, and one of the things
that went was some old stereo gear with some absolutely gorgeous heat
sinks.  I was *so* tempted to save them, but alas they went.  But I
bet it wouldn't be too hard to find more.  Anyways, polish them to a
mirror finish and mount fins-up on a small black box.  Get some retro
cloth-insulated wire for supply leads (they still make the stuff for
lamp restoration) for the finishing touch.  But if it had to have a
display, it would definitely have nixie tubes.  ;-)

Ok, pardon my mind wandering there... Where was I?  I'm not sure how
useful having all those displays active all the time really is.  It is
a good interface that allows us to easyly notice when there is a
problem, however we must poll it for information.  With software on a
PC, it would be trivial to add a little slider to set min and max safe
values on every parameter, and generate an audible alarm whenever one
of those conditions is exceeded.

Excessive in it's own sense, perhaps, but it's an interesting design
exploration, and a very good project to learn power supply design and
MCU interfacing all in one go.

But in the meantime, I'm happily tapping into my PCs power, regulating
off the 12v line with an old 7805...

-p.

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\25@084925 by Howard Winter

face
flavicon
picon face
On Fri, 24 Sep 2004 18:28:39 -0400, Peter Johansson wrote:

> Bob Blick hit an issue that I completely missed: a short on my
> breadboard will likely take the whole PC down.  That's probably enough
> reason not to do this right there without some current-limiting facility.

Indeed, because even a fairly modest PC PSU will be rated for about 30A on the +5V line, so your accidental
short could be dissipating 150W without upsetting the power supply!  That's pretty bad news for your long
wires, PCB tracks, and the little cloud of metal vapour that used to be the wire that caused the short!

Better to go for one of Wouter's ideas:  USB will supply up to 500mA at 5V, or use the +12V and regulate it
yourself with a 7805.  The latter has overload protection so will stop any embarassing smoke-releasing...

Cheers,


Howard Winter
St.Albans, England


_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\25@091243 by Peter L. Peres

picon face

On Fri, 24 Sep 2004, Peter Johansson wrote:

> Appologies in advance for a very basic EE question...
>
> What issues should I be aware of if I want to tap power directly from
> the my PC's power supply to drive my breadboard(s)?  The PC provides a
> very convenient source of regulated +5v and +12v, and is easily
> obtained through one of the 4-pin drive connectors.  A cable could be
> very easily made from sacrificed fan taps and 120v line power cable
> (16-18 ga.)
>
> I imagine the distance (approx 5 feet) from the regulator might have
> some issues, but are they serious, and can they be overcome with a
> large decoupling capacitor for the entire breadboard?  If so what
> values would be apropriate?

Be very careful of wires and connectors and never leave it run unattended.
The PC psu gives enough amps to be able to set cables on fire in minutes
under the right conditions, although it is short circuit and overload
proof.

Peter
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\25@101829 by Gerhard Fiedler

picon face
> Bob Blick hit an issue that I completely missed: a short on my
> breadboard will likely take the whole PC down.  That's probably enough
> reason not to do this right there without some current-limiting facility.

Those PC power supplies go for a few bucks, and even cheaper if you take
them out of old decommissioned PCs. There's really no need to tap the
supply of the computer you're working with, right?

> Yes, a benchtop power supply with multiple fixed and variable
> voltages, adjustable current limiting, and current/voltage readout on
> every channel sure would be nice, but something like that is *way* out
> of my price range... or would it?

I once bought a bench top PS in a surplus store for IIRC $30. Not multiple,
but one good output (adjustable voltage and current limit and meters for
both) and a good place to start.

Gerhard
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\25@114102 by Robert Monsen

picon face
Ebay has bench PSUs on sale all the time. You can get a nice current and
voltage adjustable one with a display for under $50 US

I bought one from this place, and its fine for what I need:

http://www.mpja.com/listitems.asp?dept=2&main=1

You almost never need more than an amp for electronics/uC work. The one
I got has 18V/3A, both adjustable. The only complaint I have is that the
pots they used don't have very good adjustability. A fine-tune-knob
adjustment for both voltage and current would be useful. Also, dual +-
output is occasionally useful, but can be arranged from a single supply
pretty easily.

Regards,
Bob Monsen

Gerhard Fiedler wrote:

{Quote hidden}

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\25@145434 by Gus S.Calabrese

face
flavicon
face
PC supplies can be had very cheaply.
Tiger Direct has them for $15.
I have a box of them at my home.
I use them for all sorts of things.
Getting rid of noise can be accomplished
as mentioned earlier by dropping 12V to
5V.  You can easily put front ends on the
output to track or current limit or whatever.
Harbor Freight has sales where DMMs
are sold for $3-4.  Those could be used for
for voltage or current monitoring.

All controlled by a PIC of course.

http://www.tigerdirect.com/applications/SearchTools/item-details.asp?
EdpNo=935375&CatId=1076
http://harborfreight.com/

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

2004\09\27@002849 by Bill Mc Donald

flavicon
face
Verrrrry interesting Idea.

Worthy of far more thought



----- Original Message -----
From: "William Chops Westfield" <spam_OUTwestfwTakeThisOuTspammac.com>
To: "Microcontroller discussion list - Public." <.....piclistKILLspamspam@spam@mit.edu>
Sent: Saturday, September 25, 2004 3:56 PM
Subject: Re: [EE] tapping power from PC supply


{Quote hidden}

_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist


'[EE] tapping power from PC supply'
2004\10\13@094655 by alan smith
picon face
Do it everyday...well....using a PC switcher for
general 12V, 3.3V, 5V....

Need to short across the enable pin and ground so the
supply will turn on.


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

More... (looser matching)
- Last day of these posts
- In 2004 , 2005 only
- Today
- New search...