Battery Drain

jo662

Free Member
I hope some one can help me with a problem on a Sargent electric box on a Swift Kontiki.
I had my batteries fail last year,they were five years old so not to bad. I replaced them,but
ever since they do not seem to last long when off grid. i have a 150 solar panel which disguises
the problem in the summer,but now winters here it’s becoming a pain. I was in France last week
and mainly off grid and the batteries only lasted 4 or 5 hours in the evening with a couple of LED
lights,a tv and heater on. In the morning the battery display read around 12.0 most mornings, which
is really bad.My question is how do I check for something that is draining the batteries?
I have 2x100ah Yuasa batteries btw.
 

Millie Master

Full Member
Van Electrics are indeed a black art!

I can't offer even an olive twig lat alone branch of knowldege about your problem which is on no doubt a Sevel built van, however I did have a very similar problem on my 08 Renaut Master which I eventually tracked down to an OE but aftremarket fitment of a RouteMaster kit that always remained live.

Perhaps Dave the electronics brain will be along soon to give you some chapter and verse advice.

Phil
 

Squiffy

Full Member
Just changing the batteries is unlikely to have caused a new problem, (Though not impossible). What is a fact is that the Amp hours of a battery claimed by manufacturers is not always correct I.e. 100ah could actually be only in reality 85ah, so if your old batteries were 110ah or even 100ah and were actually providing those amp hours it would make a significant difference to the length of time they provided usable power than if your new batteries are not giving what is claimed. The other possibility is that your solar panel has degenerated for some reason and not giving the amps that it used to. Phil.
 

wildebus

Full Member
I'd start to worry at anything below 11.5 volts.
11.5V for a battery not under any load would not be good at all!
For a typical Lead Acid Battery, 12.05V is around 50% discharged and generally the lowest you would want a battery to go to. As a rule of thumb, on any installs I do or kit I supply, I tell folk to avoid ever being below 12V
There are quite a few charts around, but this is a pretty typical one and a decent example to reference:

Note: the table above is for a battery at rest i.e. Not being used in the last 30 minutes or so and not having been charged for an hour or so.
If you take a note when say the solar panel is providing a charge, you can think the battery is at a much higher charge than it is (and conversely if under a load it can look as much more discharged than it really may be).

I hope some one can help me with a problem on a Sargent electric box on a Swift Kontiki.
I had my batteries fail last year,they were five years old so not to bad. I replaced them,but
ever since they do not seem to last long when off grid. i have a 150 solar panel which disguises
the problem in the summer,but now winters here it’s becoming a pain. I was in France last week
and mainly off grid and the batteries only lasted 4 or 5 hours in the evening with a couple of LED
lights,a tv and heater on. In the morning the battery display read around 12.0 most mornings, which
is really bad.My question is how do I check for something that is draining the batteries?
I have 2x100ah Yuasa batteries btw.
There are two possibilities
1) something is draining the batteries
2) the batteries are not as good as you think any more.

200Ah of batteries should last a lot longer than 4-5 hours with the load you describe.
How long does it take to recharge the batteries? If they seem to charge faster than you would expect (or faster than they used to for that matter), that is an indication they do not have the original capacity any longer.
a 150W solar panel is not a massive panel - if you had a substantial drain, you might have seen an issue still (although in the summer you would have less lighting use, no heater and maybe less TV if more time outside).

What Sargent unit do you have? if it is something like an EC155 and you have relied on that as the main battery charger,you will have never charged your batteries fully! I would guess your Yuasa batteries probably want a 14.4V absorption charge to fully charge them. The EC155 and its brothers put out a voltage of either 13.6V or 13.8V (can't recall which, but I think the lower). At that charge voltage, the batteries will never fully replenish and you will never get close to 100Ah into each battery. And over time that extra potential gets lost without some extra battery management. The same is true of the 'split-charge' in the Sargents - they are just a relay so when driving the batteries are relying on the lowish voltage output of the alternator.
I am making a big point of this as if it does turn out that the batteries are no longer that good and you decide to replace them, the same thing will happen if you don't improve the charging system.

Back to your question .... "how do I check for something that is draining the batteries?" . You need an ammeter of some kind. Either a handheld meter, or a battery monitor that has that function and watch for current use. If you don't have one, then you could at the end of the evening, make a note of the voltage, pull out plugs (such as a TV 12V plug for example) and compare the voltage in the morning. This may take a few days to see if one particular circuit is the issue.
Some people have reported they have a faulty solar panel or controller that allows backfeed into the panel when it gets dark. You could disconnect the panel from the controller to see if that makes a difference?
 

MarkJ

Full Member
An alternative to using a meter is to pull fuses one at a time from the 12V distribution box or equivalent and see if it's one particular circuit that's the problem. If there is an isolator switch (our Electroblok unit has one) you can also disconnect the 12V completely and see what happens.

We had a similar problem on our old van and it turned out to be a bit of natty wiring a previous owner had put in which by-passed the 12V box altogether - wired straight across the battery in fact - and it was feeding something we didn't use (I forget what) which had gone faulty and was on when it wasn't supposed to be.
 

SquirrellCook

Full Member
A tiny load in the range of milliamps. Sorry, but 11.5V is much much too low to be an acceptable minimum target
Frightenly when my leisure batteries are 50% discharged, this is the kind of voltage that is seen. The voltage figure is recorded by my Victron BVM. As I'm unable to see a detailed history log I don't know at what loading this takes place.
 

wildebus

Full Member
Frightenly when my leisure batteries are 50% discharged, this is the kind of voltage that is seen. The voltage figure is recorded by my Victron BVM. As I'm unable to see a detailed history log I don't know at what loading this takes place.
When you see that voltage and that SOC% voltage, the BMV will also tell you the current out (or in) at that time. Something is not correct with those numbers. Maybe there is a massive voltage drop on the voltage sense cable to the BMV so it is reading too low a figure? bad connection somewhere with extra resistance introduced? Have you cross-checked the BMV displayed voltage with measuring across the battery yourself with a Meter to see if they match?
The BMV bases its SOC calculation on the battery capacity you have entered (at the C20 rating) and the current being drawn. The voltage is not as far as I am aware used for any SOC calcuations, but as an indication for the user plus the Watts/Wh consumption (as that is Current x Voltage), so it is quite feasible to have what appears a total mismatch of SOC and Voltage even under no load or charge if the voltage to the shunt is getting 'corrupted' in some way.
What some people have found is that the resistance of the fuse (100ma) in the supplied voltage cable that goes from Battery to Shunt causes a measurable voltage drop (when I say measurable I am talking small 100ths of volts, not many 10ths) and they replace the supplied fuse with a 1A fuse to get the reading to match.

This is an important subject as incorrect battery care costs the owners hundreds of pounds and major inconveniences. So many baulk at the idea of spending money on a decent way to monitor their batteries but don't realise knowing what is going on can save them a lot more than they have spent.
 

SquirrellCook

Full Member
This is an important subject as incorrect battery care costs the owners hundreds of pounds and major inconveniences. So many baulk at the idea of spending money on a decent way to monitor their batteries but don't realise knowing what is going on can save them a lot more than they have spent.
It's all very odd. Only once have I ever been in an under voltage situation where things didn't work until the sun came up. That was before I took the Victron route. I was never convinced my "sunshine meter" was giving useful information or even charging the batteries well.
Last year I checked as well as I could the two Trojan T105's on the bench and they seemed fine.
The batteries that worry me the most are the engine start ones. I have electric doors, and slow doors are a sure sign that the batteries require charging. As mentioned before, bigger Merc's seem to suffer a constant battery drain and are supplied with a isolator switch. The down side of this is that neither the alarm or doors work once isolated! I'm shortly going to install a BVM, but its a pain to alter the negative battery cable on it. At the same time I'm going to take a new fused supply for the doors and alarm. This way most of it can be isolated and still secured. Oh what fun ;(
 

wildebus

Full Member
...For a typical Lead Acid Battery, 12.05V is around 50% discharged and generally the lowest you would want a battery to go to. As a rule of thumb, on any installs I do or kit I supply, I tell folk to avoid ever being below 12V
There are quite a few charts around, but this is a pretty typical one and a decent example to reference:

Note: the table above is for a battery at rest i.e. Not being used in the last 30 minutes or so and not having been charged for an hour or so.
If you take a note when say the solar panel is providing a charge, you can think the battery is at a much higher charge than it is (and conversely if under a load it can look as much more discharged than it really may be)....
I put the above message up a couple of days ago. I thought I would add a couple of charts to see how well those charts relate to the 'real world' and if they are useful....

This chart is a graph of my own battery bank over a period of 11 days. I have taken the start from when I just removed the mains battery charger and the batteries were reported as 100% SOC (so Full) and stops when I decided to plug in the mains charger again and the batteries had got down to 55%
1572950492611.png


Not tried this before, but I can make my own version of that table and see how it matches up. So here goes:

Battery SOCMy Battery VoltageTable Voltage
100%13.10V13.00V
90%12.72V12.75V
80%12.59V12.50V
70%12.37V12.30V
60%12.16V12.15V
55%/50%12.11V (55%)12.05V (50%)
Now the original table reading is based on a battery that has been allowed to rest before taking the readings. My system is live, so the voltage readings will fluctate somewhat as loading comes on (the 'ragged' voltage line in the chart is the result of the fridge coming on around 20 times or so a day and when on, causing the voltage to dip) and charging happening from the solar panels (the little bumps around the middle of each day).

But you see while the voltage bounces up maybe 0.25V during that charging time, the SOC only increases by 1-2%, and not the 15% or so the voltage table would suggest (which is why it is so important when reading a voltage on a meter to know if charging is occuring).

Effect of loading on the voltage?
Well my batteries in the graph above are always under load - I am pulling a pretty constant 1A day and night due to the various kit I have always running (Victron GX, Router, MiFi, BMV and illiumination on the bunch of USB sockets dotted around the van) so the voltage will always be very slightly down compared to a resting voltage.
I have added back the current in the drill-downs below so you can see the loading and how that when the current draw increases (fridge goes on), the voltage dips (by around 0.1V for the extra 4.5A in fact).
This first chart shows the voltage at the regular 1A draw when the battery is at 62.4%
1572951940123.png


This next chart actually captures the in-rush of the fridge compressor (it happens every time, but the sampling doesn't always grab it)
1572952218231.png

When the demand shoots up from the normal 1.0A to 41.15A (a fair old current draw), the voltage dips from 12.20V to 12.05V - a 0.15A drop.
Again, if by happenchance you looked at a voltmeter at that precise time and then used the table to check the SOC, you would say your batteries were down to 50% whereas in reality they are at 62.4% (as the SOC chart shows).

So yes, a Voltmeter and a Voltage vs SOC chart is a useful tool when used correctly and will give you a pretty good idea of your batteries. It can be upto 10 points out if the batteries are under a fairly higher load or charge (most people would not draw more than 15A or charge greater than 15A at any time) but still gives a decent indication.
An SOC Meter (Victron BMV, NASA Marine BM, plus numerous others) are a lot better however as they will give you a much closer indication of the batteries State of Charge and can be read at anytime without considering loads and charging.
 

wildebus

Full Member
I put my electric heater on in the underbed garage today to prevent freezing of pipes (so just on the frost setting) and also turned up the electric water heater a smidge. On checking the power use I thought it might be quite interesting to show the effect of current draw on voltage levels at a variety of current levels in a fairly short space of time, going from the Electric Compressor fridge coming on (a modest 3A) to the Oil Rad and Water Heater on together (a rather high 243A)
1573336256711.png
 
Top