Smart voltage regulator (NOT) = flat battery

G

Gilbert Smith

I frequently end up with a flat battery in my Forester05, despite the
garage test rig reporting that the alternator and battery are in
perfect condition.

What am I doing wrong ? I am driving without the heater fan or
sidelights on. If you run the battery low by parking with the lights
on (say), there is no way the battery will ever recover unless you
drive with the fan on. (I am not joking). Short journeys or long.

Charging lead-acid batteries is not rocket science, all you do is put
14.6v across the terminals until the charge current shows it to be
fully charged, then maintain a 14.2v trickle charge. Try plugging a
voltmeter into your cigar lighter socket and watch what happens on
your lovely Subaru, then compare it to any other make.
 
Gilbert said:
I frequently end up with a flat battery in my Forester05, despite the
garage test rig reporting that the alternator and battery are in
perfect condition.

What am I doing wrong ? I am driving without the heater fan or
sidelights on. If you run the battery low by parking with the lights
on (say), there is no way the battery will ever recover unless you
drive with the fan on. (I am not joking). Short journeys or long.

Charging lead-acid batteries is not rocket science, all you do is put
14.6v across the terminals until the charge current shows it to be
fully charged, then maintain a 14.2v trickle charge. Try plugging a
voltmeter into your cigar lighter socket and watch what happens on
your lovely Subaru, then compare it to any other make.
Hmm,
Sounds like you have bad battery. Checked each cell with hydrometer?
Probably one cell is sick.
 
Gilbert said:
Charging lead-acid batteries is not rocket science, all you do is put
14.6v across the terminals until the charge current shows it to be

Hi,

Have you been charging the battery externally when it's gone flat, or
simply relying on the alternator to do it?

If you're suffering from a weak or sick battery (see Tony's note on a
bad cell), you're putting extra strain on the alternator. I've seen more
than one alternator "give up" on a bad battery: the owner puts in a new
battery and hardly any time later, it's flat cuz his alternator's gone
bad as well. It's sorta "chicken and egg" diagnosing the "real" culprit!

I've also had good results when buying new batteries if I put a trickle
charger on overnight the first night the battery's in the vehicle.
Again, the idea is to minimize the strain on the alternator in case the
battery has been sitting on the shelf a while and has self-drained a
small amount.

Rick
 
Rick Courtright said:
Hi,

Have you been charging the battery externally when it's gone flat, or
simply relying on the alternator to do it?

If you're suffering from a weak or sick battery (see Tony's note on a
bad cell), you're putting extra strain on the alternator. I've seen more
than one alternator "give up" on a bad battery: the owner puts in a new
battery and hardly any time later, it's flat cuz his alternator's gone
bad as well. It's sorta "chicken and egg" diagnosing the "real" culprit!

I've also had good results when buying new batteries if I put a trickle
charger on overnight the first night the battery's in the vehicle.
Again, the idea is to minimize the strain on the alternator in case the
battery has been sitting on the shelf a while and has self-drained a
small amount.

Rick

As I said in my first paragraph, the Subaru test rig reports both
battery and alternator in perfect condition. I have also tried a new
battery because I initially thought it was the obvious solution.

The problem is simple, the system does not hold the battery teminals
above 14.1v (as per battery manufacturer's recomendation) if you do
not keep the blower or the lights on.

The Impreza08 car they lent me performs in a very similar fashion, but
differs in that it keeps returning to charging (14.6v) every two
minutes, whereas the Forester05 gives up after a couple of bursts.

The real question is why does the voltage remain permanently in the
14.2 to 14.7v range when I do have the fan switched on ?? It is
nothing to do with the extra load requirement, the facts are simple -
over 14.1v = current into the battery, below this = no charging taking
place, despite the battery being only 50% charged.
 
Rick Courtright said:
Hi,

Have you been charging the battery externally when it's gone flat, or
simply relying on the alternator to do it?

If you're suffering from a weak or sick battery (see Tony's note on a
bad cell), you're putting extra strain on the alternator. I've seen more
than one alternator "give up" on a bad battery: the owner puts in a new
battery and hardly any time later, it's flat cuz his alternator's gone
bad as well. It's sorta "chicken and egg" diagnosing the "real" culprit!

I've also had good results when buying new batteries if I put a trickle
charger on overnight the first night the battery's in the vehicle.
Again, the idea is to minimize the strain on the alternator in case the
battery has been sitting on the shelf a while and has self-drained a
small amount.

Rick
I have charged it on the bench on each occasion it went flat.
 
Gilbert said:
I have charged it on the bench on each occasion it went flat.

Although I don't hang out at Forester specific forums, I'd expect this
to be a widely reported problem if ALL Foresters exhibit it. So, I'd
have to say the shop has not done proper/enough troubleshooting.

Did you buy the car used? Is the V belt in good condition/tight?

I agree that charging batteries is not rocket science, but there are a
LOT of odd ways that system can fail in a car. My daughter's 95 maxima
has an oil leak that very gradually destroyed her alternator. While
doing so, we had many very strange problems with intermittent 'flat'
batteries. And batteries (for cars) can not tolerate many cycles to 0
volts before their longevity is also compromised. That created another
level of confusion. Add to that some parts of the country are so hot
average battery lifespan is well under 3 years....

if you frequently need to idle for long periods of time with accessories
on, investigate using a smaller alternator pulley.

I dunno

Carl
 
I read this one yesterday and thought this guy is full of it cause things
don't add up.
I have a built in volt meter on my car and it reads just fine.
Your cig lighter may be corroded inside or something like that.
Also, it sounds like you may have a bad connection somewhere in your battery
terminals.
Ever take them off and scrap the old lead off the contact areas and replace
them? I put a few drops of motor oil on top of each one and the car rattles
it all over to protect them from corrosion after cleaning.

I say this, because it sounds like your battery is not getting charged,
that's why it drains so fast.

Oh, and all I can think of is the fan is causing enough current to pass the
bad connection to cause it to also let it charge, so that can make some
sense too.
 
Gilbert said:
The problem is simple, the system does not hold the battery teminals
above 14.1v (as per battery manufacturer's recomendation) if you do
not keep the blower or the lights on.

Hi,

I had an early water-cooled VW that wouldn't charge above about 14-14.1
volts, w/ attendant short battery life. Turns out VW sourced their
alternators that year from Motorola, and the following year from Bosch
cuz of recurring similar problems w/ the Motorola units. I replaced the
alternator w/ a Bosch, and the problem went away. I'm gonna wonder out
loud if Subaru may have gotten a "bad" batch of alternators? The dealer
told me the same thing you were told: "It all checks out ok." Well, then
why do I have flat batteries? "Uhh..."

Also, bad grounds were once problematic on a lot of German (and English)
cars--they'd have the ground wire screwed down onto a painted surface
instead of having bare metal to bare metal contact. Have you checked the
connection where your battery ground wire cable meets the body/chassis?
I'm also gonna wonder out loud if your fan's providing enough ground to
allow proper voltage output? I'm not sure exactly how various
alternators are wired up, but it seems from observation some are more
tolerant of the voltage range they'll work w/ than others.

Otherwise, it's still a mystery to me.

Rick
 
Intermittent loose/cracked battery cell?
It's happened to me and had everyone stumped
for a while.
Worth checking.
-C-
 
Carl 1 Lucky Texan said:
Although I don't hang out at Forester specific forums, I'd expect this
to be a widely reported problem if ALL Foresters exhibit it. So, I'd
have to say the shop has not done proper/enough troubleshooting.

Did you buy the car used? Is the V belt in good condition/tight?

I agree that charging batteries is not rocket science, but there are a
LOT of odd ways that system can fail in a car. My daughter's 95 maxima
has an oil leak that very gradually destroyed her alternator. While
doing so, we had many very strange problems with intermittent 'flat'
batteries. And batteries (for cars) can not tolerate many cycles to 0
volts before their longevity is also compromised. That created another
level of confusion. Add to that some parts of the country are so hot
average battery lifespan is well under 3 years....

if you frequently need to idle for long periods of time with accessories
on, investigate using a smaller alternator pulley.

I dunno

Carl

Thanks for all the suggestions in this thread, everything is pristine
in the engine compartment, earth lead tight, etc, and the fault is
identical with a new battery (I tried that first, but have now swapped
back).

The crux of the matter is that the charging system behaves perfectly
if I have either the fan or the side lights on (eg: voltage above 14.1
at all times).

Secondly, an Impreza08 they lent me behaves in a very similar manner
wrt lights/fan on or off. The only difference was that without the fan
or lights it did go down to 12.8v quite regularly but kept trying
again, whereas mine stays off charge after the first five minutes.

The alternator is controlled by the ECU, hence the "smart", but I
reckon they had a software bug for a couple of years.
 
Gilbert Smith said:
I frequently end up with a flat battery in my Forester05, despite the
garage test rig reporting that the alternator and battery are in
perfect condition.

What am I doing wrong ? I am driving without the heater fan or
sidelights on. If you run the battery low by parking with the lights
on (say), there is no way the battery will ever recover unless you
drive with the fan on. (I am not joking). Short journeys or long.

Charging lead-acid batteries is not rocket science, all you do is put
14.6v across the terminals until the charge current shows it to be
fully charged, then maintain a 14.2v trickle charge. Try plugging a
voltmeter into your cigar lighter socket and watch what happens on
your lovely Subaru, then compare it to any other make.

You haven't said what the charging voltage reads with the engine
running and everything else off. Is it higher than battery voltage
with the engine off? Is it higher at 2000 rpms than at idle?

Have you bothered to check the current draw when the engine is off
and all accessories, interior lights, etc. are off?

Disconnect the negative terminal and put an ammeter between the post
and the terminal. After waiting several seconds for any capacitors to
recharge, the current should be less than 0.05 amps. Anything higher,
start pulling fuses one by one until you discover the circuit with the
problem.

Bob
 
Bob Bailin said:
You haven't said what the charging voltage reads with the engine
running and everything else off. Is it higher than battery voltage
with the engine off? Is it higher at 2000 rpms than at idle?

All modern cars including Subarus can charge at max rate when the
engine is idling. Increasing the revs makes no difference to any of
them including mine.

The charging voltage is 14.7v during the periods when it is delivering
any charge to the battery.
Have you bothered to check the current draw when the engine is off
and all accessories, interior lights, etc. are off?

Disconnect the negative terminal and put an ammeter between the post
and the terminal. After waiting several seconds for any capacitors to
recharge, the current should be less than 0.05 amps. Anything higher,
start pulling fuses one by one until you discover the circuit with the
problem.

There is absolutely no need to measure current flow, the charge rate
of the battery is entirely down to the voltage across the terminals.
14.2v for minimum trickle charge recommended for normal use when fully
charged, 14.7v for a healthy charge rate.

The system is perfectly capable of doing this continuously, indeed it
does exactly this with the lights on. It is only in conditions of
minimum load that it fails to maintain even a trickle charge, despite
the battery state being below 50%.

Gilbert
 
Gilbert Smith said:
All modern cars including Subarus can charge at max rate when the
engine is idling. Increasing the revs makes no difference to any of
them including mine.

The charging voltage is 14.7v during the periods when it is delivering
any charge to the battery.


There is absolutely no need to measure current flow, the charge rate
of the battery is entirely down to the voltage across the terminals.
14.2v for minimum trickle charge recommended for normal use when fully
charged, 14.7v for a healthy charge rate.

The system is perfectly capable of doing this continuously, indeed it
does exactly this with the lights on. It is only in conditions of
minimum load that it fails to maintain even a trickle charge, despite
the battery state being below 50%.

Gilbert

If this were any other make of car, I'd say you had a bad voltage
regulator on your alternator. Has your garage tried installing a
new or rebuilt alternator as a test? It's one of the benefits of
going to a dealer with these problems. Yes, I'm aware that
alternators on this car cost upwards of $450 list.

You mentioned in another post that you think the ECU controls
the alternator output. I think you're mistaken. The voltage
regulator on most modern alternators consists of a small logic
board, not just static components. Hence the term "smart".

If the swapped alternator behaves properly, and you're not
covered under warranty, you might be able to save $$$ by
replacing just the voltage regulator, only (!) $170 list.

Bob
 
Bob Bailin said:
If this were any other make of car, I'd say you had a bad voltage
regulator on your alternator. Has your garage tried installing a
new or rebuilt alternator as a test? It's one of the benefits of
going to a dealer with these problems. Yes, I'm aware that
alternators on this car cost upwards of $450 list.

You mentioned in another post that you think the ECU controls
the alternator output. I think you're mistaken. The voltage
regulator on most modern alternators consists of a small logic
board, not just static components. Hence the term "smart".

If the swapped alternator behaves properly, and you're not
covered under warranty, you might be able to save $$$ by
replacing just the voltage regulator, only (!) $170 list.

Bob

The dealer told me the ECU was the voltage regulator. The car is still
in warranty (just) and I am talking to the directors at the dealership
to get them to accept the fault. It would be nice if it was the
alternator, but my suspicions are aroused by the fact that it behaves
perfectly with the fan or lights on.

The other day I stopped at the end of a trip leaving the radio on for
15 minutes before re-starting (ie: didn't go through the 'off' posn).
The battery volts never went above 12v the whole way home. I hope it
is repeatable.

It would be really nice if someone else could do the same check as me,
I guarantee you will be surprised by the fluctuations - who knows,
maybe they are all like this ?

Gilbert
 
Gilbert Smith said:
The dealer told me the ECU was the voltage regulator. The car is still
in warranty (just) and I am talking to the directors at the dealership
to get them to accept the fault. It would be nice if it was the
alternator, but my suspicions are aroused by the fact that it behaves
perfectly with the fan or lights on.

The other day I stopped at the end of a trip leaving the radio on for
15 minutes before re-starting (ie: didn't go through the 'off' posn).
The battery volts never went above 12v the whole way home. I hope it
is repeatable.

It would be really nice if someone else could do the same check as me,
I guarantee you will be surprised by the fluctuations - who knows,
maybe they are all like this ?

Gilbert

If you reported this problem before your warranty ran out, then you're
covered. Also, I believe your dealer is mistaken about the ECU
controlling the alternator. You wouldn't have a design with two
methods of voltage regulation, too complicated.

Since the function of the voltage regulator is to control the alternator
output based on system load and yours doesn't respond the way it
should, then it should be a no-brainer to put in a working alternator
and compare it.

Perhaps it's failing when the engine gets good and hot and your dealer
is testing it after it's cooled off.

They could, if they wanted to, hook you up with a portable ODBC II
monitor and monitor the system voltage level while you have the
car. When they play it back, they'll see what you're talking about.

But trying a replacement alternator costs them nothing but the time
to install it. If it's not the solution, they can put it back on the shelf.

Bob
 
Bob said:
If you reported this problem before your warranty ran out, then you're
covered. Also, I believe your dealer is mistaken about the ECU
controlling the alternator. You wouldn't have a design with two
methods of voltage regulation, too complicated.

Since the function of the voltage regulator is to control the alternator
output based on system load and yours doesn't respond the way it
should, then it should be a no-brainer to put in a working alternator
and compare it.

Perhaps it's failing when the engine gets good and hot and your dealer
is testing it after it's cooled off.

They could, if they wanted to, hook you up with a portable ODBC II
monitor and monitor the system voltage level while you have the
car. When they play it back, they'll see what you're talking about.

But trying a replacement alternator costs them nothing but the time
to install it. If it's not the solution, they can put it back on the shelf.

Bob
Hi,
I think the dealer techs ae incompetent. Time to try another dealer.
 
I just talked to the Subaru importers, and they know all about this
problem, including having looked for ways of fixing it without
modifying the ECU software. They can offer no solution other than
driving with the fan or lights on all the time. Apparently it affects
all models since 2004, and is due to trying to get the best results
from carbon emission tests by keeping the generator off-load.
Gilbert.
 
Gilbert said:
I just talked to the Subaru importers, and they know all about this
problem, including having looked for ways of fixing it without
modifying the ECU software. They can offer no solution other than
driving with the fan or lights on all the time. Apparently it affects
all models since 2004, and is due to trying to get the best results
from carbon emission tests by keeping the generator off-load.
Gilbert.

I have yet to experience a dead battery (due to my own negilgence OR the
'smart voltage regulatoer' issue you mention) in my '06 WRX. If the
system was incapable of keeping the battery charged, shouldn;t I have
had a dead battery several times by now?

I'm in the US - are you? Could this be specific to another market?

Carl
 
Carl 1 Lucky Texan said:
I have yet to experience a dead battery (due to my own negilgence OR the
'smart voltage regulatoer' issue you mention) in my '06 WRX. If the
system was incapable of keeping the battery charged, shouldn;t I have
had a dead battery several times by now?

I'm in the US - are you? Could this be specific to another market?

Carl

I am in UK, so quite possibly the US market is different. Do you have
charges based on CO2 emissions ? Are you using daytime driving lights,
or aircon frequently ?

When I talked to Subaru UK they were quite open about it, in fact I
only had to say I was monitoring the battery voltage and they went on
to describe exactly what I was seeing, and the things they had tried
while trying to fix it.

Gilbert
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
13,966
Messages
67,560
Members
7,448
Latest member
zeushead01

Latest Threads

Back
Top