Smoothing Capacitor Values

[Michael]

Hi,

I'm making a device for an ROV and need to bring the onboard power of 12V
down to 5V.

After a bit of research I can't seem to find any exact (or near enough)
values for the smoothing capacitors once it's gone through a regulator (I'm
planning on using a L78S05CV
http://www.maplin.co.uk/Module.aspx?ModuleNo=7939&DOY=1m10 ) and just before
(between the 12V and ground).

Is it just a case of bigger is better?

The other thing is some sites say to use film whilst other say I should use
electrolytics and ceramics.....

Can someone please explain?

Thanks,

Michael

 

[Chris]

Hi,

I'm making a device for an ROV and need to bring the onboard power of 12V
down to 5V.

After a bit of research I can't seem to find any exact (or near enough)
values for the smoothing capacitors once it's gone through a regulator (I'm
planning on using a L78S05CV
http://www.maplin.co.uk/Module.aspx?ModuleNo=7939&DOY=1m10 ) and just before
(between the 12V and ground).

Is it just a case of bigger is better?

The other thing is some sites say to use film whilst other say I should use
electrolytics and ceramics.....

Can someone please explain?

Thanks,

Michael

Hi, Michael. The specs say you should have a cap at the input of the
regulator if the power source is more than a couple of inches away,
For being fed from a battery, 10uF should be plenty. For the output, a
small cap helps for high frequency stability, and to keep the noise
voltage within spec. You can either use another 10uF electrolytic, or
an 0.1uF ceramic or film would be fine, too.

When you lay it out, tie the GND pin of the regulator and the - pins of
the electrolytics within an inch of each other if you can.

Good luck
Chris

 

[Michael]

Hi, Michael. The specs say you should have a cap at the input of the
regulator if the power source is more than a couple of inches away,
For being fed from a battery, 10uF should be plenty. For the output, a
small cap helps for high frequency stability, and to keep the noise
voltage within spec. You can either use another 10uF electrolytic, or
an 0.1uF ceramic or film would be fine, too.

When you lay it out, tie the GND pin of the regulator and the - pins of
the electrolytics within an inch of each other if you can.

Good luck
Chris

Thanks again Chris

Michael

 

[Eeyore]

I'm making a device for an ROV and need to bring the onboard power of 12V
down to 5V.

After a bit of research I can't seem to find any exact (or near enough)
values for the smoothing capacitors once it's gone through a regulator (I'm
planning on using a L78S05CV

No smoothing cap is needed because your input supply is DC already.

You will however need some small caps to ensure stability located close to the
regulator. See the manufacturer's data sheet for recommended values.

Graham

 

[Eeyore]

Hi,

I'm making a device for an ROV and need to bring the onboard power of 12V
down to 5V.

After a bit of research I can't seem to find any exact (or near enough)
values for the smoothing capacitors once it's gone through a regulator (I'm
planning on using a L78S05CV

That regulator doesn't have a high enough input voltage spec to survive
automotive supply voltage transients.

Graham

 

[Phil Allison]

"Eeysore"

That regulator doesn't have a high enough input voltage spec to survive
automotive supply voltage transients.

** Errr - so is 35 volts DC max input is not enough ???

What bollocks.





......... Phil

 

[Chris]

That regulator doesn't have a high enough input voltage spec to survive
automotive supply voltage transients.

Graham

If you're still there, Mike, Graham has a good point. An automotive
environment has high voltage transients of over +60VDC, which will
eventually fry your regulator.

For greater reliability, you might want to add a 24V Transzorb (Maplin
N95CA part number 1.5KE24A, £0.46 ea. in stock) and a 1 ohm junkbox
power resistor to your circuit like this (view in fixed font or M$
Notepad):

| _____
| B+ ___ | |
| o----|___|-o-----o----|78S05|---o-------o
| 1R | | |_____| | +
| | +| | +|
| D1| --- | ---
| /-/ C--- | C--- 5VDC
| ^ | | |
| | | | |
| | | | | -
| o----------o-----o-------o------o-------o
| GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

If a spike occurs on B+, the transzorb will act like a zener and clip
the transient to a maximum voltage of around +24V. It's made to absorb
brief, high energy spikes like this, and isn't too expensive for what
you get. A quick scan didn't show any 1 ohm resistors at Maplin, but
you'll probably have one (anything between 0.47 ohms and 2 ohms will
do) in your junkbox. The 1 ohm resistor will limit the current during
the spike, and keep the Transzorb from smoking. Actually, the resistor
will also drop some of the voltage, which might mean you can use a
smaller heat sink, too.

This will make your automotive circuit a lot more reliable.

Cheers
Chris

 

[Rich Webb]

if that was the case, there would be a lot of burnt electronics in
many vehicles.

Google for "automobile electronics load dump".

<http://www.automotivedesignline.com/howto/192500568;jsessionid=FFCP4HWJFBD0WQSNDLRCKH0CJUNN2JVN>
is off the first link. There are many others.

 

[Chris]

if that was the case, there would be a lot of burnt electronics in
many vehicles.

There were a lot of burnt electronics in many vehicles -- in the 1960s,
as the newfangled solid state electronics were being rolled out into
cars. Everyone learned the hard way about load dump transients, which
will frequently happen if the battery connection is intermittent.

Usually some inductance is placed at the power input to automotive
electronics, and there are various methods used to clip the amplitude
of transients, many of them built into the boards in odd places. The
junkbox resistor/Transzorb method is clunky, but it works.

Either that, or if you'd like to use tubes in your car radio, they're
very happy with voltage transients. ;-)

From the Pleistocene, when men were men and women were glad of it

Chris

 

[Michael]

If you're still there, Mike, Graham has a good point. An automotive
environment has high voltage transients of over +60VDC, which will
eventually fry your regulator.

For greater reliability, you might want to add a 24V Transzorb (Maplin
N95CA part number 1.5KE24A, £0.46 ea. in stock) and a 1 ohm junkbox
power resistor to your circuit like this (view in fixed font or M$
Notepad):

| _____
| B+ ___ | |
| o----|___|-o-----o----|78S05|---o-------o
| 1R | | |_____| | +
| | +| | +|
| D1| --- | ---
| /-/ C--- | C--- 5VDC
| ^ | | |
| | | | |
| | | | | -
| o----------o-----o-------o------o-------o
| GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

If a spike occurs on B+, the transzorb will act like a zener and clip
the transient to a maximum voltage of around +24V. It's made to absorb
brief, high energy spikes like this, and isn't too expensive for what
you get. A quick scan didn't show any 1 ohm resistors at Maplin, but
you'll probably have one (anything between 0.47 ohms and 2 ohms will
do) in your junkbox. The 1 ohm resistor will limit the current during
the spike, and keep the Transzorb from smoking. Actually, the resistor
will also drop some of the voltage, which might mean you can use a
smaller heat sink, too.

This will make your automotive circuit a lot more reliable.

Cheers
Chris

Yup I'm still here....havn't quite gathered up the courage to finally submit
my order

The circuit won't be connected up to a car though.....(although it will be
connected to a car battery)

What w rating should the resistor be in that circuit? As big as possible I'm
guessing.....

I guess better safe than sorry is the key phrase here....

Cheers,

Michael

 

[Michael]

Yup I'm still here....havn't quite gathered up the courage to finally
submit my order

The circuit won't be connected up to a car though.....(although it will be
connected to a car battery)

What w rating should the resistor be in that circuit? As big as possible
I'm guessing.....

I guess better safe than sorry is the key phrase here....

Cheers,

Michael

Don't know if it makes a difference......but the ROV is the underwater
type.....

Example: http://www.sut.org.uk/urg_uris/Images/Hercules ROV.gif

Michael

 

[Chris]

Yup I'm still here....havn't quite gathered up the courage to finally submit
my order

The circuit won't be connected up to a car though.....(although it will be
connected to a car battery)

What w rating should the resistor be in that circuit? As big as possible I'm
guessing.....

I guess better safe than sorry is the key phrase here....

Cheers,

Michael

Hi, Mike. Load dump is a problem associated with alternators (and to a
lesser extent, with battery chargers). If you're just running it off
straight battery power (and there's zero chance that it will be powered
when the battery is being charged) you won't need transient protection.


If you decide you still need transient protection for your power
supply, the value of series resistor just depends on the current
requirement. P = (I ^ 2) * R. If you have a wirewound 1 ohm resistor
in your junkbox, it will probably be 5 watts or more, which should be
just fine for any current you might require (the 78S05 is good for 2
amps max, which would mean 4 watts max).

Good luck with your project, and feel free post back if you have any
other questions.

Cheers
Chris

 

[BobG]

Mike... linear regulators are not efficient... Look for a 12V to 5V DC
to DC converter at Jameco or Digikey.. they can be 90% effcient!
Batteries will thank you.

 

[Jamie]

Michael wrote:




That regulator doesn't have a high enough input voltage spec to survive
automotive supply voltage transients.

Graham

if that was the case, there would be a lot of burnt electronics in
many vehicles.

 

[Michael A. Terrell]

"Eeysore"


** Errr - so is 35 volts DC max input is not enough ???

What bollocks.

........ Phil

No, it isn't, because there are spikes in an automotive electrical
system that go well above + 35 volts. If your weren't such a loser, you
would have checked the facts BEFORE you dumped your usual load of crap.
There are standards for automotive electrical and electronics hardware,
including "Load Dump".


Go back to molesting toasters, Phyllis, and leave the real
electronics to the men.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Michael A. Terrell]

There were a lot of burnt electronics in many vehicles -- in the 1960s,
as the newfangled solid state electronics were being rolled out into
cars. Everyone learned the hard way about load dump transients, which
will frequently happen if the battery connection is intermittent.

Usually some inductance is placed at the power input to automotive
electronics, and there are various methods used to clip the amplitude
of transients, many of them built into the boards in odd places. The
junkbox resistor/Transzorb method is clunky, but it works.

Either that, or if you'd like to use tubes in your car radio, they're
very happy with voltage transients. ;-)

Chris

Delco used a "Spark plate" which was a piece of thin double sided PCB
material in their car radios, as well. It was soldered to the case on
one side, and used as a tie point in the DC input and filtering.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Phil Allison]

"Michael A. Terrell
"Eeysore"

No, it isn't, because there are spikes in an automotive electrical
system that go well above + 35 volts.

** Not if the battery is still functional -

FUCKING MORON !


Go back to molesting little kids in Vietnam - like real US soldiers.





........ Phil

 

[Michael A. Terrell]

"Michael A. Terrell
"Eeysore"


** Not if the battery is still functional -

Only in your small mind. There is a lot of stray inductance and
capacitance in the wiring harness. Do you really expect anyone to
believe that the arc when a starter solenoid opens doesn't impress a
number of large spikes on the rest of the electrical system? Your
glaring ignorance is showing, toaster molester.

FUCKING MORON !

About damn time that you admitted it, Phyllis.

Go back to molesting little kids in Vietnam - like real US soldiers.

....... Phil

First of all, i never served in Vietnam. Its you that got caught in
those sexual acts, and real soldiers don't molest anyone, faggot.


--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[Homer J Simpson]

First of all, i never served in Vietnam. Its you that got caught in
those sexual acts, and real soldiers don't molest anyone, faggot.

The photos from Abu Ghraib would seem to indicate otherwise.

 

[Phil Allison]

"Michael A. Terrell"

" Eyesore "

First of all, i never served in Vietnam.

" Actually, I was sent to Alaska INSTEAD of Vietnam. Two months after
I arrived the base I was supposed to go to in Vietnam was overrun, and
the entire staff was killed. "


** What a fucking shame YOU were not killed back then too !!!!

Be one less TOTALLY INSANE Septic Tank **** polluting the earth.

There is a lot of stray inductance and
capacitance in the wiring harness. Do you really expect anyone to
believe that the arc when a starter solenoid opens doesn't impress a
number of large spikes on the rest of the electrical system?

** A 470 uF electro on the input of the voltage reg will suppress such
transients - standard practice to use one.

Most reg ICs have a similar input voltage limit to the one in question
anyhow - so Eyesore has to say what ** HE ** meant by his asinine
remark.


NOT SOME CRIMINAL, LYING PSYCHO **** LIKE YOU !!!


**** OFF - ASSHOLE !!!!!!!!




......... Phil