🚀 go-pugleaf

streeter@theory.lcs.mit.edu (Kenneth B. Streeter) writes:


>Some of you may recall a previous query concerning building an
>electronic odometer for my car.  (The application is TSD rallying.)
>Well, I've pursued a number of ideas, and still haven't really reached
>a decision.  However, I am having power supply difficulties with one
>of my test circuits.

>In order to supply the 5V needed to drive the circuits, I am using a
>pair of 7805 regulators, using the car's 12V power as a supply.  I
>have each regulator driving about half a dozen CMOS chips (uv erasable
>PALs, mainly) and five or six hexadecimal LEDS.  (The regulators
>aren't in parallel, they're each driving separate sets of chips.)
>However, I seem to be having power supply difficulties.  The
>regulators can just barely handle the current requirements, it seems.
>They get REAL hot (even with heat sinks) and can barely muster 4
>volts.  This makes the circuit VERY flakey.

>Driving the entire circuit from an external 5-volt supply (bypassing
>the regulators) indicates 1.26 amps of current being drawn.

Your symptoms sure sound a lot like one or more regulators are oscillating.
The pair should be able to handle 1.26 amps with room to spare if
the load is partitioned well and the heat sinks are adequate.  Have
you hung a wideband scope on the output and looked for oscillation?
I've seen these babies oscillate well up into VHF.  You MUST have 
low ESR bypassing as near to the pins as possible.

If there is no oscillation, then I agree with another poster that
the problem is likely thermal foldback.  You need more heatsink and/or
to distribute the heat load.  One trick I use is to hook several
power diodes up in series with the raw power input to the regulator.
Each diode will drop between 0.7 and 0.8 volts which at your
current draw is about a watt.  Move those watts off the regulator 
and onto the diodes.  One thing you have to be careful with using
this technique is not to drop your input margin too much or the
voltage will sag during low input voltage conditions such as when
you are cranking the car.  I'd keep at least 7 volts on the input 
pin of the regulators under all conditions.


John

-- 
John De Armond, WD4OQC  | We can no more blame our loss of freedom on congress
Radiation Systems, Inc. | than we can prostitution on pimps.  Both simply
Atlanta, Ga             | provide broker services for their customers.
{emory,uunet}!rsiatl!jgd|  - Dr. W Williams |                **I am the NRA**