Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

cascade voltage regulators?

1,296 views
Skip to first unread message

Rainman

unread,
Oct 24, 1998, 3:00:00 AM10/24/98
to
If i have, say.. 15 volts DC, can i use a 7812 to give me a 12 volt
output, and then use a 7805 with the 12 volts as the input, to get 5
volts out? Maximum current draw of 1 Amp from 5 volt supply, and about
22 mA from 12 volt supply.

A yes or no answer will suffice. :)

sincerely,
Ray
rai...@connection.com


Rainman

unread,
Oct 24, 1998, 3:00:00 AM10/24/98
to
Okay, i tried it anyway, and no, it doesn't work. I do get 5.0 V out, but
the 12V regulator o/p drops to 9.6 volts. If i remove the 7805, it goes
back up to 12.0 V. Why does this phenomenon occur? Maybe is there some
sort of intermediate circuitry i could impliment that would correct this.
Any ideas?

CFCF Tech Engineering

unread,
Oct 24, 1998, 3:00:00 AM10/24/98
to
YES you can...!
--

========== Message Seperator ===========

Rainman <rai...@connection.com> wrote in article
<36326AA8...@connection.com>...

CFCF Tech Engineering

unread,
Oct 24, 1998, 3:00:00 AM10/24/98
to
Hi there...

I don't see why this should not work. You have 15 volts feeding into a
12 volt regulator, then the 12 volts feeds a 5 volt regulator... If the
total amperage drawn from the regulators exceeds the rating, or the rating
of the source then you will have the drop you are speaking of.

Make sure that the input of the 12 volt regulator does not go below about
15 volts or so. I believe about 14 volts is the minimum it can work with
on the input... The headroom must be adiquate, and the regulators also
heatsinked properly.


--

========== Message Seperator ===========

Rainman <rai...@connection.com> wrote in article

<36326D13...@connection.com>...


> Okay, i tried it anyway, and no, it doesn't work. I do get 5.0 V out, but
> the 12V regulator o/p drops to 9.6 volts. If i remove the 7805, it goes
> back up to 12.0 V. Why does this phenomenon occur? Maybe is there some
> sort of intermediate circuitry i could impliment that would correct this.
> Any ideas?
>
>
> Rainman wrote:
>

Jack Schidt

unread,
Oct 24, 1998, 3:00:00 AM10/24/98
to
A) Ripple voltage is making the minimum head voltage fall below I/O
differential of 7812 (standard 78XX = 2V MINIMUM @ 1A, A- version = 1.8
MINIMUM @ 1A).

B) Make sure that neither regulator is oscillating due to inadequate
input/output bypassing, or reversed capacitor polarities.

C) If heat were the issue, the 7805 would be the one folding back, as it has
a full 4 + volts drop across it.

Use a scope to confirm.

BTW, are you hooking up the 7805 with no load, and seeing the 7812 output
drop? If so, the 7805 is blown, or hooked up backwards.

This is not a simple yes/no answer, but the best I could do.

Good Luck

JS

Rainman wrote in message <36326D13...@connection.com>...

KW Yap

unread,
Oct 25, 1998, 2:00:00 AM10/25/98
to Rainman
Try direct connect 7805 to 12v,7805 ( max I=1 amp) and put one heat sink
on that,good luck and try an error.
r31088.vcf

Tom MacIntyre

unread,
Oct 25, 1998, 2:00:00 AM10/25/98
to
Rainman <rai...@connection.com> wrote:

>If i have, say.. 15 volts DC, can i use a 7812 to give me a 12 volt
>output, and then use a 7805 with the 12 volts as the input, to get 5
>volts out? Maximum current draw of 1 Amp from 5 volt supply, and about
>22 mA from 12 volt supply.
>
>A yes or no answer will suffice. :)
>
>sincerely,
>Ray
>rai...@connection.com
>

You may be better off to take both the 5V and 12V inputs off the 15V
regulator output, rather than cascading them.

Tom

Pete

unread,
Oct 26, 1998, 3:00:00 AM10/26/98
to
Have you smoothed your 15v supply with a large cap (2000uf+)?

Rainman wrote:
>
> Okay, i tried it anyway, and no, it doesn't work. I do get 5.0 V out, but
> the 12V regulator o/p drops to 9.6 volts. If i remove the 7805, it goes

> back up to 12.0 V. Why does this phenomenon occur? Maybe is there some
> sort of intermediate circuitry i could impliment that would correct this.
> Any ideas?
>

Paul Grohe

unread,
Oct 26, 1998, 3:00:00 AM10/26/98
to
On Sat, 24 Oct 1998 20:02:48 -0400,
in the newsgroup sci.electronics.basics,
Rainman <rai...@connection.com>
from thoughtfully posted:

> If i have, say.. 15 volts DC, can i use a 7812 to give me a 12 volt
> output, and then use a 7805 with the 12 volts as the input, to get 5
> volts out? Maximum current draw of 1 Amp from 5 volt supply, and about
> 22 mA from 12 volt supply.
>
> A yes or no answer will suffice. :)

Well....Yes.....BUT

You may run into two problems:

#1. "Dropout Voltage" problems

#2. 78xx series is current limited to around 1A.

#1 is related to the fact that regulators must have a certain
voltage difference between the input and output to keep them
"alive". This voltage difference, called "dropout voltage", gets
bigger as more current is drawn.

Depending on the manufacturer, and even the particular lot the
part came from, this voltage may be 1.5 to 2.5V for a 78xx series
under full load. That means, under load, the input voltage should
not drop below (12V + 2.5V) or 14.5V (always plan for worst case
scenarios).

If your 5V regulator is pulling 1A out of the 12V regulator, then
you have to make sure the input voltage to the 12V regulator
never falls below 14.5V, or you loose regulation. And if you
decide to draw more current from the 12V line, then you push the
12V regulator that much closer to current limiting.

Why not just feed each regulator from the same raw supply? No
"cascading". Only one regulator would require heatsinking.

The total power dissipated ("wasted") will be the same. The 5V
regulator will be "wasting" all 10W by itself. The 12V would only
be supplying the 22mA, which is just a few dozen mW, so it
probably would not need a heatsink.

If both regulators are next to each other - and on the same
heatsink - then feed them both from the same 15V supply.

Of course, if this is "unavoidable" because the regulator is
"downstream" somewhere else on the board - then "cascading" is
okay to do. In fact, it is quite common - but you had *better*
provide *plenty* of heatsinking for both regulators - especially
the 5V one.

Cheers,
Paul Grohe


--
---------------------------------------------------------------
Paul Grohe National Semiconductor Corp.
Assoc. Analog Apps Engineer 2900 Semiconductor Drive
Standard Analog Apps Group Mail Stop C2693
mailto:gr...@galaxy.nsc.com Santa Clara, CA. 95052-8090 USA
Usenet Disclaimer: Any opinions expressed are mine, not NSC's
----------------------------------------------------------------

Jim H

unread,
Oct 26, 1998, 3:00:00 AM10/26/98
to
On Sat, 24 Oct 1998 20:13:08 -0400, Rainman <rai...@connection.com>
wrote:

>Okay, i tried it anyway, and no, it doesn't work. I do get 5.0 V out, but
>the 12V regulator o/p drops to 9.6 volts. If i remove the 7805, it goes
>back up to 12.0 V. Why does this phenomenon occur? Maybe is there some
>sort of intermediate circuitry i could impliment that would correct this.
>Any ideas?
>

There's no real reason for you to pre-regulate the input to the 5 volt
regulator with the 12 volt regulator. Try running them both directly
from the 15 volt supply.

Also, you better check on the input to your 12 volt regulator. Look
at this point with an oscilloscope while the circuit is under load.
You may find that this 15 volt supply is pulling down under load and
you're not getting enough "head" voltage for your 12 V regulator. You
should try to have 3 volts of "head" voltage to get the full current
out of it. If your 15 volt supply is just the unloaded voltage from
an unregulated supply, then it will very likely pull down too much and
you will need a better "coarse" supply for your whole system.

Jim H.

jlpr...@gmail.com

unread,
May 14, 2015, 9:44:10 AM5/14/15
to
El sábado, 24 de octubre de 1998, 2:00:00 (UTC-5), Rainman escribió:
> If i have, say.. 15 volts DC, can i use a 7812 to give me a 12 volt
> output, and then use a 7805 with the 12 volts as the input, to get 5
> volts out? Maximum current draw of 1 Amp from 5 volt supply, and about
> 22 mA from 12 volt supply.
>
> A yes or no answer will suffice. :)
>
> sincerely,
> Ray
> rai...@connection.com

YES
0 new messages