Thread: Warning on auxiliary batteries

This is a warning to D4 owners particularly, but it probably also applies to the D3 and the RRS.

It is NOT satisfactory to use the simple dual-battery systems that work on simpler vehicles. The best system for charging auxiliary batteries on a D4 is a DC-DC converter type.

Now a simple solenoid/relay operated dual battery controller may work adequately for SOME users for SOME of the time, but it will NOT work adequately for ALL users ALL of the time.

Further, there is a very real risk with the use of ANY dual battery system that the vehicle ECU will malfunction, causing anything from engine shut-down to a flat cranking battery. DC-DC converters are less likely to cause problems than solenoid types.

Apologies for the length of this post. Acting on ‘information received’ I have tried to give a full explanation with references and methods of verifying what I write.

Below is a brief explanation followed by a more detailed explanation with some references to support what I say. Where I don’t quote references, I will suggest the type of person to ask so that anyone interested can easily verify what I write.

However, it would be ideal if an un-biased moderator would show this post to an expert for comment. By expert I mean someone with no vested interest in promoting one thing over another, someone with electronic ECU skills, and someone who understands battery charging requirements. This may not be easy to find – I am sure I am not alone in being surprised how many people who should be experts aren’t. In a recent post I noted that Sneigy works for MLR. I wonder if he could be persuaded to use his position at MLR to have this post studied by engineers at Land Rover UK for comment? I don’t think it would be worth troubling anyone at LR Australia….

Anyway, here goes. First are the points made briefly, followed by a more detailed explanation of each point for anyone interested. I will try to keep it easy to understand.

IN BRIEF

1. The D4 has a multi-stage charging system (aka smart alternator) that uses a higher voltage until the cranking battery is fully charged then drops back to a float voltage of 13.2 volts. Reference, a phone conversation with Peter, service department, MLR.
2. AGM batteries will take an impossibly long time to fully charge at 13.2 volts. Even at 13.5 volts, a Fullriver AGM battery takes 60 hours charge to just 55%. Reference, Fullriver batteries, their graph included in details section.
3. Because the D4 has a multi-stage charging system, extra loads MUST NOT be connected directly to the battery posts. Hopefully when I get my D4 I will be able to find the correct connection points and advise the forum.
4. Any extra electric loads have the potential to cause problems with the car ECU. With most devices, the risk of trouble is vanishingly small but with dual battery systems there is a slim but real risk of trouble. This presumably is why Land Rover won’t honour warranties if such after-market electronic devices are fitted. To make the ECU impervious to interference by other electronic devices is possible but extremely difficult, especially in an automotive environment. DC-DC converters are less likely to cause problems than solenoid based dual battery systems.

Right, now the boring detail.

1. Multi-stage charging results in extended battery life. In a vehicle like the D4 that takes such huge advantage of electronics this is very important. Multi-stage can mean anything up to seven stages, but broadly the important stages are the bulk charging and the float charge. Broadly, single stage charging just applies a fixed voltage to the battery, regardless of the state of charge of the battery. This can result in overcharging the battery, and reduced battery life. Multi-stage charging drops the voltage down after the battery is fully charged.

The time taken to fully charge a cranking battery can be very small. This is because normally there will only be a tiny amount of power taken from the car battery, ie, that used to start the engine plus a tiny amount while the car is at rest. If accessories are used while the car is idle, then of course it will take longer for the battery to become fully charged and for the system to go into float mode.

The point here is that there will seldom be enough time to put a significant charge into auxiliary batteries before the charger goes into float mode. Most of the time there will only be 13.2 volts available for charging the auxiliary battery.

For confirmation, anyone with a D4 (or an RRS or a D3) can easily check this themselves. Buy a cheap digital multi-meter. With the battery fully charged after a reasonable run, switch off the engine and wait maybe 15 minutes. Start the engine and measure the voltage across the battery posts. You should observe that the voltage reads about 14.5~15 volts for a short time, then drops back to around 13.2 volts.


2. Below is a graph sent to me by Fullriver batteries. It shows the time taken to charge an AGM battery at various fixed voltages. As can be seen, at 13.5 volts the battery only charges to 55% after 60 hours! I asked them about 13.2 volts – they simply replied “We do not recommend charging at 13.2 volts”. Put simply, you will need MORE batteries to run your gear if you don’t charge them fully. I don’t know how much DC-DC converters cost these days – I bought mine from GSM at the Sydney Caravan show a couple of years ago, for about $330. Whatever, they are smaller and lighter than extra batteries would be, and probably cheaper.

For confirmation, ask your battery supplier what they think of the success of charging with a voltage of 13.2 volts.

3. The correct way to determine when a battery should be put into float mode – ie, fully charged - is to measure the CURRENT the battery draws when supplied with a fixed charging voltage. As the battery becomes more fully charged with this fixed voltage, the current it draws reduces. When it reduces to a specified amount, (typically a few amps) the battery is considered fully charged and the charger drops into float mode.

In a vehicle it is therefore necessary to measure the current being drawn by the battery to determine its state of charge. If another load is placed directly across the battery posts, the current drawn by that load will fool the charging system into thinking that the battery is not fully charged, and it will thus keep the voltage at the higher level instead of dropping into float mode. This could result in over-charging and possibly reduced life of the cranking battery. It may also result in a fault code being logged.
The car system may measure the current going into either the positive or negative battery post via a current shunt. If it is in the negative lead, then it should be simply a matter of connecting any additional load to the chassis of the vehicle, NOT the negative battery post. If the current is measured in the positive lead, it will be more difficult to locate the correct point for connection of leads. I do not know the answer to this, and Land Rover won’t tell me, they just say that adding aftermarket accessories will void the warranty. When I get my D4 next month, I will investigate and report back.

Remember, this applies equally to solenoid based and to DC-DC converter based systems.

I don’t know how you can confirm this. If Land Rover know, they ain’t saying!

4. Damage to the ECU is not easy to discuss in simple terms. Broadly, problems are caused by voltage spikes getting in to the wrong part of the ECU. I will discuss this in two parts – CAUSES of voltage spikes, and RESULTS of voltage spikes.

4a CAUSES. Whenever there is a sudden change in the amount of current flowing through a wire, there is a possibility that a voltage spike will be generated. The greater the change of current, the greater the spike. The more sudden the change of current, the greater the spike. Also inductive coils, like those used in relays and solenoids cause voltage spikes in exactly the same way that the ignition coil generates a spark in a petrol engine. Suddenly connecting auxiliary batteries to the cranking battery by a solenoid gives a sudden large change in current and can cause a voltage spike. Likewise, the application or removal of voltage from a solenoid or relay coil can cause a voltage spike. It is possible to filter these spikes to an extent with the use of diodes etc, but that is rarely 100% effective.
A DC – DC converter is much better in this respect. The current drawn normally builds up slowly (slow in ECU terms – maybe the current builds up in 1/200,00ths of a second, that is slow in ECU terms and unlikely to cause a spike). Also, the current drawn is limited, usually to a maximum of about 1.5 times the rated output current.

4b RESULTS. Ideally, there would be no results. All voltage spikes would be filtered out before they reached the ECU and there would be no problems. However, this is incredibly difficult in an automotive environment. There are numerous switches and sensors all over the vehicle connected back to the ECU, and the wires leading to them act like antennas picking up spikes generated by wires leading to other equipment, such as dual battery systems. Some of these sensors produce only tiny voltages and must be amplified before the ECU can deal with them. This makes them more sensitive to spikes.

It is much easier for the designers to protect against KNOWN spikes such as those caused by starter motors etc than it is to protect against UNKNOWN spikes such as are generated by after-market equipment.
It is also very difficult to test if a particular installation has the potential to cause problems. This is because the processor in the ECU has a very busy and hugely complicated life. It is continuously polling all the switches and sensors for data, then processing that data and instructing various systems in the vehicle to make appropriate changes. A micro-processor can only do ONE thing at a time. Thus when a spike appears, the ECU processor could be involved in any of many thousands of different things. The results, if any, of the spike will depend upon exactly what the processor was doing at the time of the spike. Most of the time there will be no result. But there will be times when there COULD be a result, and that result will be quite unpredictable.

Thus to fit an after-market dual battery system and try it out and have everything work fine is meaningless. The test would need to be run a huge number of times with the vehicle in a huge number of states to be meaningful. No, the best thing the designer of the car can do is protect against LIKELY spikes in the design stage with correct hardware layout and ‘defensive’ programming.

It is clear to me that the Land Rover engineers in this respect have done a fantastic job on the D4 (and presumably also on the D3 and RRS). The electronics are hugely complicated and are a good part of the reason why these cars are just so damn good! To grumble because this makes it slightly more complicated to run auxiliary batteries seems very ungrateful. Besides, troubles with the ECU seem comparatively rare even with dual battery systems fitted.

However, I have seen on this forum posts relating to unexplained drops into ‘limp home’ mode, unexplained flat batteries and others. It would be very interesting to do a poll on this site to find out of all those who HAVE had ECU problems how many have dual battery systems fitted, and what type?

For confirmation of this part of the post, show this to someone who designs micro-processor based systems – a professional hardware and software designer, not the sparky at work.

In brief....

If youve direct coupled the batteries through say, a dampned solenoid like they've been using in pinball machines since they started putting electronics in pinball machines driving a set of contacts that've got a a choke on it you'll not get your initial spike on connection or disconnect.

and the cranking battery wont take a full charge till the aux battery has approached full charge... (assuming youve actually flattened it as much as your claiming)

not sure why youd want to bring a computer hardware engineer or software engineer to solve a problem thats been solved by electrical engineers and automotive electrical engineers (as well as aeronautical and maritime engineers) for years (all the way back to steam driven electrics) with no great hassles.

Just out of curiosity have you yet tried to see what a D2+ does when you just manually hook the 2 batteries together with a pair of decent jumper cables?