Solutions to 12 Volt Batteries and Chargers Posted Here

[BrockWI]

Yes, my initial thought was some odd phantom load and I believe that is still in play. And yes I have checked it with a clamp on meter and found it is usually in the 50 ma range, but I have found it at 800 ma twice and power cycling the car would fix it. I am not going to check each time I power the car down to see if it really shut off and I have no faith any dealer or service tech is going to catch that or even if they did if they could fix it.

I do believe though that if the battery was properly charged it would carry it through the next power cycle or until the car was charged next. So it is a double whammy to the aux battery. My initial thought was to replace the aux battery with something larger, about twice the size, then an occasional 800 ma load wouldn't matter. But since I have been charging weekly I think that is carrying me through.

 

[RegGuheert]

If I can manage to let the LEAF sit for another 24 hours, I'll see what the voltage is at this time tomorrow. My guess: 12.70V.

O.K. Today's reading, another 24 hours after yesterday's reading is 12.71V. Slightly better than my projection. Here are the voltage results for this test, so far (each -> represents 24 hours):

12.87V->12.78V->12.71V

It seems the battery is holding charge a bit better than it did a couple of years ago when I made the graph in the OP.

Today the voltage is 12.64V. That is another drop of 70mV, very slightly better than the result achieved a couple of years ago. Here are the voltage results for this test, so far (each -> represents 24 hours):

12.87V->12.78V->12.71V->12.64V

Unfortunately, the LEAF will not sit for another 24 hours since it will need to be charged and driven in the morning. The good news is that the battery appears to be in very good condition at this point!

 

[69800]

That's unfortunate for you and costly. Did you consider that there might have been an excessive current demand on the lead-acid battery when the vehicle is not in use? This can be problematic for late model vehicles where a CAN bus fails to 'sleep' causing ECUs to not power-down. You might consider using
a clamp-on amp meter to measure the 'sleep' current draw at the battery, or carefully place a series
amp meter in the circuit without breaking the battery's connection. The 'sleep' current should be less
than about .050 amps (50 ma) and ideally less than 30 ma. In some cases, it might take as long as
a few minutes before the vehicle enters the 'sleep' mode.

It is agreed that there is a small phantom draw on the leaf. Most modern cars have them these days. With a proper charging system on a daily driver this is not a problem. The problem with the leaf is the algorithm the are using is insufficient to fully charge the battery. There are at least 100 posts scattered about this forum that confirm this. And by reading other sites such as battery university it is re confirmed. A battery cannot be fully charged a 13.1 volt where the leaf spends 95 percent of its time. I have a volt meter in my car and it rarely spend more than a few minutes at 14.6 volts where true absorption charging takes place.

 

[AndyH]

Has anyone measured the unloaded output of the charger on the Leaf?

This is important because we cannot tell the set voltage by logging the battery voltage. If one has a partially discharged battery of any type connected to a variable power supply, one can turn the voltage up to 50 and the amperage to 10 and the voltage at the battery will still be 12.0. We wouldn't be able to measure a voltage above the safe limit of the battery until it's more than 100% charged. Since the car's DC-DC and/or charger isn't fully charging the car, any voltage logged will reflect the battery, not the charger output.

The experiment I'd do is to connect a logger to the battery, then plug the car in for a 100% charge - and then leave the car plugged-in for at least another 12 hours after the traction battery's full.

 

[lorenfb]

If one has a partially discharged battery of any type connected to a variable power supply, one can turn the voltage up to 50 and the amperage to 10 and the voltage at the battery will still be 12.0.

Not necessarily, as it's a function of the state/condition of the battery and it's output impedance.
A battery with a high output impedance will not be able to fully sink the 10 amps and the voltage
will rise above 12 volts but not necessarily to 50 volts. The proper test for any battery is a load test
to determine its output voltage (basically its output impedance) under a heavy load, i.e. at greater
than its ampere-hour current rating, but less than its cold cranking ampere rating (CCA).

Using the LeafDD, one can monitor the Leaf's battery's output impedance (delta V/ delta I) over time
and as a function of the battery temperature, which provides a measure of the Leaf's battery's condition.

 

[RegGuheert]

Since the car's DC-DC and/or charger isn't fully charging the car, any voltage logged will reflect the battery, not the charger output.

Since the impedance between the output of the charger and the battery is extremely low, the voltage of the output of the charger is virtually the *same* as the voltage of the battery (they will not differ by more than a few millivolts).

 

[Bob]

Are you able to configure your voltage logger as a current logger and use it to log battery current when idle over a one-week period? Even better would be to simultaneously measure battery voltage and current but that would require a two-channel logger. That would give us a very good picture of dark current and also the top-off algorithm.

Depending on the voltage logger, you could get by with an 0.1 ohm series resistor. Or you could measure the current sense element already in series with the battery ground and rely on that voltage drop for sensing.

It is rare that I leave my Leaf sit idle for many days, but if it were helpful, I could setup an experiment like this when I next expect a long idle period. I'd have to lash up logging with a computer and DMMs, but that should be possible.

Bob

 

[AndyH]

Since the car's DC-DC and/or charger isn't fully charging the car, any voltage logged will reflect the battery, not the charger output.

Since the impedance between the output of the charger and the battery is extremely low, the voltage of the output of the charger is virtually the *same* as the voltage of the battery (they will not differ by more than a few millivolts).

What you said is accurate but is not what I'm talking about. I'm not suggesting that the connection between the data logger probes and the battery terminals is bad. I'm also not commenting on the state of the connection between the charger and the battery.

When a charger or power supply is connected to a battery or cell charged to less than 100%, the voltage measured at the battery will be somewhat above the resting voltage of the battery or the cell, but isn't likely to reflect the actual voltage setting of the charger. There's a voltage rise when charging that's somewhat like the voltage sag under load.

For example: Take a 12V 10Ah SLA that is 80% charged and connect a power supply set to 50VDC and 2A. The voltage measured at the battery terminals will rise from about 12 to about 15VDC, but won't read 50V. Adding the battery into the circuit will appear to 'pull the voltage down'. One would be measuring the voltage of the charger/battery system but would not be measuring the output voltage of the charger or the open circuit voltage of the battery.

It seems that some have assumed that the charger isn't providing proper 3-stage (though only 2 stages are needed to fully charge the battery - CC/CV - float is only for storage to overcome self discharge) voltage levels, but so far no logging performed is capable of determining that. It appears all that's happened so far is that the statement in the owner's manual that the 12V battery might not be fully charged if the car's used for short trips is verified.

That's why I suggested plugging the car in, charging the traction pack to 100%, and then leaving the car connected for at least another 12 hours to see if the car actually will attempt to bring the 12V battery to a full charge.

 

[RegGuheert]

Are you able to configure your voltage logger as a current logger and use it to log battery current when idle over a one-week period? Even better would be to simultaneously measure battery voltage and current but that would require a two-channel logger. That would give us a very good picture of dark current and also the top-off algorithm.

Depending on the voltage logger, you could get by with an 0.1 ohm series resistor. Or you could measure the current sense element already in series with the battery ground and rely on that voltage drop for sensing.

It is rare that I leave my Leaf sit idle for many days, but if it were helpful, I could setup an experiment like this when I next expect a long idle period. I'd have to lash up logging with a computer and DMMs, but that should be possible.

This is actually extremely difficult to do, and it is one of the reasons why three-stage chargers have not shown up on vehicles until recently. It is also one of the reasons why the three-stage chargers that do exist today do such a poor job of charging the 12V battery.

The problem is that the currents flowing into and out of the battery range from about 100A in to 100s of amps out. A 0.1 ohm resistor dissipates over 1000W if you have over 100A flowing through it. As a result, the resistance of the shunt used by vehicle manufacturers is much lower than 0.1 ohm. Then the problem becomes that it is extremely difficult, if not impossible, to accurately measure the mA-level currents needed to properly terminate the absorption portion of a three-stage charger in an automobile.

That is why I am the only one who, to my knowledge, has measured the tare current of the LEAF to the nearest mA. I had to use a trick to be able to make that measurement using the 300mA scale on my Fluke multimeter.

So the question becomes: How can I criticize Nissan for not handling the charge termination well when the vehicle's cannot possibly measure its current accurately? The answer is that the problem that exists does not have to do with measurement accuracy. Instead, it has to do with the current that does NOT get measured when the vehicle is turned off. The charging system needs to keep track of how long the vehicle was off and use knowledge of the car's off-state currents (programmed by the designers) to than replace the proper amount of charge when it is next turned on (or, for an EV, when it is next charged). Unfortunately, it seems clear that the existing system assumes that the answer to this question is either "zero" or "damned little" was lost while the vehicle was turned off, which is invalid in some cases. The result is a charging system which virtually never tops off the battery's charge level.

 

[RegGuheert]

For example: Take a 12V 10Ah SLA that is 80% charged and connect a power supply set to 50VDC and 2A. The voltage measured at the battery terminals will rise from about 12 to about 15VDC, but won't read 50V. Adding the battery into the circuit will appear to 'pull the voltage down'.

It does not "appear to pull the voltage down". It ACTUALLY pulls the voltage down. At any given point in time, a power supply can control EITHER its output current OR its output voltage but not both simultaneously (or sometimes it regulates output impedance or output power).

In the case you mention, the settings for the power supply are 50V with a 2A current limit. As soon as the current reaches 2A, the power supply switches from regulating its output voltage to regulating it's output current.

One would be measuring the voltage of the charger/battery system but would not be measuring the output voltage of the charger or the open circuit voltage of the battery.

That's right. But the voltage of the charger/battery system is all there is in the vehicle charging application. (You can also measure current, but as I said in the previous post, it is not an easy task. You also have a constantly-changing load current attached to the battery, which further muddies the waters.)

As others have said, if you want to charge and/or test your battery, you can take it out of the system to do that. But the same doesn't apply to the charger: if you want to see how (and how well) it charges a battery, then you need that battery in the circuit.

 

[RegGuheert]

Unfortunately, the LEAF will not sit for another 24 hours since it will need to be charged and driven in the morning. The good news is that the battery appears to be in very good condition at this point!

I need to plug in the LEAF now, but it has sat for another 19 hours since the last reading, so here is the result:

12.87V->12.78V->12.71V->12.64V->12.61V (All readings 24 hours apart except the last pair, which are 19 hours apart.)

So there is little to no sign of sulfation now. (A sulfated battery will drop relatively much more quickly through the voltages representing the high SOCs to some voltage representing a lower SOC where it will then resume a more normal discharge voltage curve. The portion of capacity lost to sulfation is that range above that lower SOC where it acts like a normal battery.)

I will charge for about three hours (and I will try to measure how long the charger stays at 14.5V) and then will drive about one hour today and then will let it rest before charging again. I will try to get a reading preceding the second charge to see how it compares with this 12.61V reading immediately before plugging in.

 

[RegGuheert]

...(and I will try to measure how long the charger stays at 14.5V)...

I plugged in and ran in the house for a couple of minutes. By the time I came out, it had already dropped down to 13.1V. So I only know that the charging at 14.5V lasted less than five minutes. That's unfortunate, since it has up to three hours (today) to do the job properly.

 

[69800]

On a side note but related. Look at the performance of you halogen high beams when the car is at 13.1 volts most of the time. Here is an interesting thread on LED high beams to overcome the low voltage we run at. But lets keep the light bulb thread over there and not here.

http://www.mynissanleaf.com/viewtopic.php?f=37&t=19598&p=419806#p419806" onclick="window.open(this.href);return false;

Halogen bulb light output is severely compromised with decreased voltage. The drop in light output is not linear, it is exponential with the power 3.4. For example, let's consider a 9006 low beam bulb rated 1000 lumens at 12.8 Volts and plug in different voltages:

10.5V : 510 lumens
11.0V : 597 lumens
11.5V : 695 lumens
12.0V : 803 lumens
12.5V : 923 lumens
12.8V : 1000 lumens Rated output voltage
13.0V : 1054 lumens
13.5V : 1198 lumens
14.0V : 1356 lumens Rated life voltage
14.5V : 1528 lumens

 

[69800]

Has anyone measured the unloaded output of the charger on the Leaf?

This is important because we cannot tell the set voltage by logging the battery voltage. If one has a partially discharged battery of any type connected to a variable power supply, one can turn the voltage up to 50 and the amperage to 10 and the voltage at the battery will still be 12.0. We wouldn't be able to measure a voltage above the safe limit of the battery until it's more than 100% charged. Since the car's DC-DC and/or charger isn't fully charging the car, any voltage logged will reflect the battery, not the charger output.

The experiment I'd do is to connect a logger to the battery, then plug the car in for a 100% charge - and then leave the car plugged-in for at least another 12 hours after the traction battery's full.

I can't remember where I read it but Ingineer measured the output and I was surprise at how many amps it put out. It was way more than adequate but I cant quote the number.

 

[RegGuheert]

I can't remember where I read it but Ingineer measured the output and I was surprise at how many amps it put out. It was way more than adequate but I cant quote the number.

He measured it at over 130A.

 

[TimLee]

He measured it at over 130A.

When I left the LEAF in Ready in the garage one day and forgot about it, discovered that it does not shutdown the 12V once the main pack has shutdown.
So the 12V was dead, under 4V.
When I jump started the LEAF from ICE battery observed in LEAF Spy Pro that the DC to DC was charging the 12V at around 120 amps.
Stayed there a couple minutes and then trailed off down below 20 amps over about 8 minutes.
It is one heck of a battery charger.

The shortcoming is in how Nissan designed it's use.
And I think the speculation that they don't have a means to measure current drain while the vehicle is Off is correct. That is why their protocol often fails to charge the 12V enough.

They did other stupid things too.
An Auxiliarry mode with no warning when voltage is too low.
Only an auto shutoff after one hour.
An On mode that waits till voltage drops to 10.5V before starting to recharge the 12V.
Both of those leave the vehicle user with lots of ways to grossly shorten the life of the 12V battery.

 

[BrockWI]

Yes the DC-DC converter is beefy. Not sure why it is so beefy, but it is and I like it I was surprised when testing my 12v 2000w sine wave inverter on it and loaded it with 1650w heater. The 12v battery in the car stayed at 13.0v under that load. I also started a 1 hp air compressor and it started up and ran just fine, likely about 4000 w inrush on start-up or about 333a at 12v.

 

[lorenfb]

For example: Take a 12V 10Ah SLA that is 80% charged and connect a power supply set to 50VDC and 2A. The voltage measured at the battery terminals will rise from about 12 to about 15VDC, but won't read 50V. Adding the battery into the circuit will appear to 'pull the voltage down'. One would be measuring the voltage of the charger/battery system but would not be measuring the output voltage of the charger or the open circuit voltage of the battery.

Which is what I said;

"Not necessarily, as it's a function of the state/condition of the battery and it's output impedance.
A battery with a high output impedance will not be able to fully sink the 10 amps and the voltage
will rise above 12 volts but not necessarily to 50 volts."

It's basic high school electric shop!

 

[lorenfb]

That is why I am the only one who, to my knowledge, has measured the tare current of the LEAF to the nearest mA. I had to use a trick to be able to make that measurement using the 300mA scale on my Fluke multimeter.

Hardly! Anyone with a low scale clamp-on amp meter can very easily measure the Leaf's 'sleep' state
current. And anyone that has an in-series amp meter can also easily and ACCURATELY measure that
current by carefully NOT interrupting the 'sleep' current when attaching the meter, thus presenting
a Leaf's ECU from entering its active mode increasing the current.

"I had to use a trick"
Please!

This thread is not like other threads, e.g. Global Warning, Politics, Guns, & FCBEVs, where hyperbole
can be typically presented without actual test data to support one's case.

 

[lorenfb]

When I jump started the LEAF from ICE battery observed in LEAF Spy Pro that the DC to DC was charging the 12V at around 120 amps.
Stayed there a couple minutes and then trailed off down below 20 amps over about 8 minutes.
It is one heck of a battery charger.

Good data!

The shortcoming is in how Nissan designed it's use.
And I think the speculation that they don't have a means to measure current drain while the vehicle is Off is correct. That is why their protocol often fails to charge the 12V enough.

Most all late model ICE vehicles with extensive CAN systems don't attempt to monitor the 'sleep' current
of the vehicle given how the various ECUs communicate and operate on the various CAN buses and
bus gateways even when the ICE vehicle is in its 'sleep' mode.