Searching \ for '[EE]: 7805 bypassing...' 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=7805
Search entire site for: '7805 bypassing...'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: 7805 bypassing...'
2002\04\03@214124 by Barry Gershenfeld

picon face
>But what about the small capacitor(s)?  Anyone have links to a doc
>that explains this?

In my day the popular device was the LM309.  It came in a TO-3
package and the datasheet said "connect no more than 0.1 uF
to the output".

Seems that in the rush to charge up the output capacitor
it would go overcurrent ( > 1A ) and shut down.   After
it shut down there was no current, so it would start
again.  You know what happens next.

I checked Fairchild's data and couldn't find this for the
7805.  But there in the example is the .1 cap.  So
the implication's the same.   Except I couldn't find
a mention of the overcurrent thing.

But the moral is: You can put small caps on the output
to try to get rid of high frequency noise, but let the
regulator do its job on the slower stuff.

Barry

--
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\04\03@232334 by Tom Messenger

flavicon
face
At 06:21 PM 4/3/02 -0800, you wrote:
>>But what about the small capacitor(s)?  Anyone have links to a doc
>>that explains this?

And Barry replied:

>
>In my day the popular device was the LM309.  It came in a TO-3
>package and the datasheet said "connect no more than 0.1 uF
>to the output".

--snip--

>But the moral is: You can put small caps on the output
>to try to get rid of high frequency noise, but let the
>regulator do its job on the slower stuff.

The answer to the original question also addresses Barry's concern.

And the answer is RTFM. Read The Fine Manual. Or in this case, the
datasheet. All linear voltage regulators are NOT created equal. National
Semi produced several that had a gap in acceptable capacitance: up to a
certain amount was fine, more would cause it to oscillate, finally more
still beyond a number in the datasheet would be ok. If this looks spooky to
you, just accept it as good honest info in their data sheet.

Most data sheets for these parts will carefully describe how much
capacitance should be provided. The output cap is often (but not always)
spec'ed to be around 10uf to provide good transient response to load
changes. Some parts in the bad ole days would not tolerate more than .1uf.

What are ya gonna do? Read the data sheet. They are almost always available
on the net and there is no good excuse to design without one. As a matter
of fact, you are not designing if you haven't read the data sheet, rather,
you are sticking parts together and hoping it will work.

Best regards,
Tom M.

--
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\04\03@232546 by Pic Dude

flavicon
face
Wow, that's a new perspective (to me).  I was always under the
impression that when it comes to caps on power supply rails,
the bigger the better to smooth out any ripples.


{Original Message removed}

2002\04\03@234711 by Pic Dude

flavicon
face
Have to agree with you about reading the datasheet, but I haven't
seen the info you are referring to.  I have the National Semi
datasheet (same as actual regulator I'm using), and all I remember
is something about measurements (for elec characteristics) being
taken with a .22uf input cap and .1uf output cap (if I remember
correctly).







{Original Message removed}

2002\04\04@013930 by Tom Messenger

flavicon
face
At 10:53 PM 4/3/02 -0600, you wrote:
>Have to agree with you about reading the datasheet, but I haven't
>seen the info you are referring to.  I have the National Semi
>datasheet (same as actual regulator I'm using), and all I remember
>is something about measurements (for elec characteristics) being
>taken with a .22uf input cap and .1uf output cap (if I remember
>correctly).

I've been assuming the part in questin here is the 7805 (since that's on
the subject line). Searching Nationals site, they say to use the data sheet
for the LM340. Several places therein, it mentions that it is ok to use no
output cap if the load is close to the regulator or a .1uf ceramic. Long
lines to the load could cause your circuit to oscillate. But my guess is
that as others pointed out, your 7805 is suffering from excess power
dissipation.

The output impedance chart shows the results of no cap vs a 10uf tantalum
cap. Finally, it mentions that no protection diode is required for caps up
to 10uf on the output but for larger ones, a diode should be used. The
point of this is that there is plenty of info.

There are also some suggested circuits useful to increase the power
handling abilities. And at the bottom of the page at Nationals site, there
are several apnotes worth looking at, especially AN-103.

Get one of National's cdroms for linear parts; they have most of the data
sheets and apnotes. Then get the cd's from Linear Tech, Maxim, and any
others you can lay your hands on.

You must be starting to have fun now since you are burning your fingers -
the best way to learn!

Best regards,
Tom M.

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


2002\04\04@014629 by Vasile Surducan

flavicon
face
About bypassing, a clever user can bypass a 7805 with a power resistor.
Then you are able to get 1...1,5 A from an 1A device with no ( very big )
problems.
The rule: 1. never let the supply without required load
         2. never change the load value in large steps

Who ever have repaired an old TV knows that...


best Vasile


On Wed, 3 Apr 2002, Pic Dude wrote:

{Quote hidden}

> {Original Message removed}

2002\04\04@110547 by michael brown

flavicon
face
> You must be starting to have fun now since you are burning your fingers -
> the best way to learn!
>
> Best regards,
> Tom M.

;-)

I agree, if you don't "let the smoke out" of something every now and then,
your not really learning.

michael brown

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


2002\04\04@191339 by Pic Dude

flavicon
face
----- Original Message -----
From: "Tom Messenger" <spam_OUTkristTakeThisOuTspamTHEGRID.NET>


> ...
> You must be starting to have fun now since you are burning your fingers -
> the best way to learn!
> ...

"Fun" is the ultimate understatement here.  I've always been fascinated
by electronics, and started playing with it since I was 10.  I was really
into the TTL stuff and 8080, Z80 microprocessors, etc.  However,
information wasn't as easy to come by at that time where I was (in the
Caribbean), so a lot of the non-core circuity was treated as "modules"
-- such as the 5V regulator circuit.  Once I got a working regulator
module, I could reproduce it for any circuit.  However, it was not
optimized -- sometimes it would get hot, and other times it was
unnecessarily too much for a simple circuit.  But it didn't matter since
it was all experimentation, and all just a hobby.  (I remember once
clamping a TO-220 package to a metal desk cause I didn't have a
heatsink).

What's really fulfilling about this info I'm absorbing recently, is that it
fills in the blanks (from my past), even if its 20+ years later.

I've not had the greatest luck with analog circuitry though, and my
guess is that that was due to the significant number of substitutions I
had to make for locally-available parts.  Also, a logic-probe for
digital debugging was easy to get or make.  An oscilloscope was a
bit more unreachable for a pre-teen.

As a side note, I still have a TRS-80, which should still work. 1 Mhz
Z80 was all the rage!  It's hard to believe that I fill my bathtub today
with a 20Mhz PIC.

Thanks for the great information everyone -- it's awesome!

Cheers,
-Neil.

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


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