AVR-Chat

Do you Yahoo!?
Yahoo! Mail Address AutoComplete - You start. We finish.

At 10:39 AM 9/23/2004, jay marante wrote:

>i have my project on its final stage, the problem is that the 7805 voltage regulator heats up the heat sink too much. is this ok? i have inputs between 12V - 18V. it is connected to a 6V SPDT reed relay switch, which in turn, facilitates the switching of power source (main and batt).
> 
>also, the relay coil is powered by only 5V (from the regulator) instrad of 6V, and sometimes it doesn't toggle the internal switch. is this because of the coil supply of 5V instead of 6V? but sometimes it does work properly upon switching. and sometimes, i need to tap the relay for the internal switch to toggle.
> 
>thanks for the help.

Power the relay from the 12V side, and add a series resistor if needed. 
You obviously need to shed some load from the 5V supply.

If you can lower the input voltage by any amount, that will help. 
If nothing else, several 1n4001 diodes in series will help spread out the heating.

--- In AVR-Chat@yahoogroups.com, jay marante <jaythesis@y...> wrote:

> ... the 7805 voltage regulator
> heats up the heat sink too much.

You have the information you need to solve your problem in other 
posts. But, I would like to ask what you mean by "too much". Are you 
worried that the 7805 will be damaged or is the temperature a problem 
for some other reason?

Regulators are typically OK with junction temperatures up to 125 
degrees celsius. A TO-220 package typically has a junction to case 
thermal resistance of three degrees celsius per watt. Let's guess 
that your heat sink has a thermal resistance to ambient of 15 degrees 
celsius per watt and ambient temperature is 30 degrees. A quick bit 
of math, just like solving a voltage divider of two resistors, and we 
get the maximum heat sink temperature of 109 degrees. Thus, the 
regulator will be OK even if the heatsink is hotter than a boiling 
pan of water. Maybe you're worried for nothing! Mind you, if this 
heatsink is somewhere you might touch it, other factors come into 
play, which is exactly what I'm getting at.

Graham.

Jay -

Some basic principles are in order here.

The (approximate) amount of power dissipated by the
regulator is the load current times the voltage difference
across the regulator. 

Temperature rise is directly proportional to power.

Thus, you have two choices with a given regulator and heat
sink: (1) reduce the load current and (2) reduce the
input-output voltage difference.

Moving the relay to the input side would help, but it would
be hard to adapt with such a large input voltage range
using a simple resistor. 

You might consider a second (6V?) regulator just for the
relay. 

The minimum input voltage of a 7805 is about 7.8V. Thus,
you can affort to "waste" 4V of your input. This is
governed by the minimum input of the regulator and the
minimum available source voltage. Adding 5 1N400x diodes in
series would probably help.

Can you use a relay with lower coil current? That would
help.

Your relay problems are PROBABLY the result of trying to
run a 6V relay at 5V. That is very typical of the behavior.
Can you replace it with a real 5V relay? Or, why not a 12V
relay. That would many are probably rated for voltages up
to 18V.

Jim
---------------------------------------------------------------
The Think Different Store
http://www.thinkdifferentstore.com/
For All Your Mac Gear
---------------------------------------------------------------

>
>Your relay problems are PROBABLY the result of trying to
>run a 6V relay at 5V. That is very typical of the behavior.
>Can you replace it with a real 5V relay? Or, why not a 12V
>relay. That would many are probably rated for voltages up
>to 18V.

Pull-in and drop-out voltages are very handy to know!

At 08:39 AM 9/23/04 -0700, you wrote:
>i have my project on its final stage, the problem is that the 7805 voltage 
>regulator heats up the heat sink too much. is this ok? i have inputs 
>between 12V - 18V. it is connected to a 6V SPDT reed relay switch, which 
>in turn, facilitates the switching of power source (main and batt).
>
>also, the relay coil is powered by only 5V (from the regulator) instrad of 
>6V, and sometimes it doesn't toggle the internal switch. is this because 
>of the coil supply of 5V instead of 6V? but sometimes it does work 
>properly upon switching. and sometimes, i need to tap the relay for the 
>internal switch to toggle.

I agree with the suggestions of looking for a 12V relay.  Try to find an 
automotive relay they have to withstand/work in a large voltage range so 
they may be able to operate over the input range.  In any case check the 
pull-in drop-out voltage specs first.

  On your dropout problem I have my doubts that the issue is driving the 
relay at 5V, the pull-in voltage on the relays I've
worked with usually have a larger range.  Another possibility is that your 
power supply impedance is too high and the voltage to the coil is dropping 
below your supply voltage.

You need to know how much current your circuit is drawing (and how stable 
the output supply is) before you can determine what your heat source is 
(and how reasonable it is).  A simple shunt measurement will help here.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III