Ideas for improving my charging/battery system?

[Sven...]

I am thinking I am often wasting diesel or undercharging my batteries, and I probably need more batteries and/or a battery management system anyway.

 

The boat is a 57' widebeam, six years old and has a 43hp Beta Marine engine. I bought the boat a month ago and soon realised the old domestic batteries were worn out, hence I replaced them. Now there are 3x Platinum sealed 110 AH leasure batteries from Alphabatteries. In theory I should have 330 AH capacity, and as you will see below, I am easily using all of it.

 

I am living aboard as a continuous cruiser with normally no shoreline power. I am also working from the boat at daytime, using a laptop and an external computer screen, lights, gadgets, etc.

 

At the moment there is an extremely simple setup for the domestic charging system: The alternator and the batteries. There is no external regulator or battery management system, no amperemeter/counter, not even a voltmeter.

 

The alternator has the name “Iskra” on it and I think this came as an original part of the Beta 43 engine installation (as it is painted like the rest of the engine). Perhaps one of you would know more about this model. I have been told that the belts were recently replaced, and to ensure they are not loose, I have tried to turn the alternator by hand, which was not possible. There is a separate starter battery with its own smaller alternator – this looks about half the size of the Iskra.

 

There is a 12V fridge, Webasto diesel central heating system, and the normal lights, etc. There is an inverter connected directly to the batteries, a 5kW Chinese model which I have learned to use with care. Finally there is a shoreline connection system and a generator, but these are totally separate from my batteries (no charger). The only way to charge the batteries now is by running the engine. Finally, I am planning to get some solar panels - but I think perhaps Spring is a better time to invest on that front.

 

I have recorded my experiences with the batteries over a typical 24-hour period and posted this separately below. Basically it seems I am using all of the 330AH capacity, easily breaking the 50% rule, but I am also a bit in the dark about what is the status of my charging, consumption and batteries at any time. I guess my questions are...

 

• If I were to install some battery regulator/management/monitoring/split charge system, what setup would you recommend? I'd like to save on diesel and not undercharge or overcharge, and being a bit of a geek, I wish to have as much status information as possible. I would also like to be able to charge the batteries from the generator or shoreline once in a while.
  
• Should I install more batteries – and how many?
  
• Any other advice as to how to improve my system?
  

 

All the best,

Sven

[Sven...]

So here is my log from the last 24 hours...

 

Thursday morning, after a night of running various appliances, I decided to start observing closely how my charging and batteries behave. I am using my trusted old digital multimeter, a Mastech M838. This is supposed to be accurate within 0.5% with DC voltages and shows all measurements with two decimal places, like 12.04V. Below I have rounded all readings to one decimal place (like 12.0V) as I guess the second decimal might be inaccurate, and it is often fluctuating anyway.

 

In the morning I started by turning off virtually all of the loads (by way of flipping all main switches in the boat). I waited for an hour and then measured the domestic batteries at the terminals. This showed 12.0 volt (actually between 12.02 and 12.04). I think this means around 20% state of charge for my type of batteries, certainly very little anyway. While I was down there, I also measured the starter battery, which showed 12.9V.

 

Next I started the engine and let it run on tickover, a bit below 1000 revs that is.

 

After a minute or two I measured voltages again. For the domestic system, I now found 13.9V at the alternator output, and 13.4V at the battery terminals. So it was charging, but I would have expected a lower voltage, as I thought the voltage is supposed to climb gradually. I also wondered if the voltage drop through the cables should be a concern.

 

For the starter battery, I now found 12.8V at the alternator and 12.7V at the battery terminals. I guess this was also then busy charging, but again found it strange that this voltage was so much lower than for the domestic system.

 

I came back after 10 minutes and found that things had changed for the starter battery. Its alternator now showed 14.6V and the terminals 14.4V. So the starter battery was more or less fully charged and I guess everything is good on that side of things. I made a mental note that, for 95% of the time I am not making use of the capacity of the starter alternator. Maybe a SmartBank or similar could improve this?

 

As for the domestic bank, nothing had changed after 10 min. Again there was 13.9V at the alternator and 13.4V at the battery terminals.

 

Next I started experimenting with revving the engine. I found that I could increase the voltage by increasing the revs. At 1200-1300 revs, I read 14.4V at the domestic alternator and 13.8V at the domestic battery terminals. A further increase of revs did not make much difference. I decided to leave it at around 1200 revs, as I suppose a higher voltage can charge the battery better and quicker.

 

Now I switched on all my domestic gear – the fridge, inverter, computer equipment, lights, etc. I knew this would draw around 20A. (There is no central amperemeter, but in the past I have measured each appliance individually.) The charge at the battery terminals did not change from 13.8V. I felt happy about the situation, left the engine running at 1200 revs, and got back to my work.

 

After two hours I had a look again. The domestic voltage was now 14.6V at the alternator and 14.4V at the terminals. That looked quite good I thought, so I stopped the engine to find out more. In order to allow for any surface charge to dissipate and get a proper and consistant reading, I turned off all my domestic systems again and left the batteries to rest for an hour. Actually there is always a constant 0.1 ampere load from my inverter in standby mode, but I am not sure if this is enough to take off the surface charge, combined with waiting for one hour.

 

Anyway, 60 minutes later, my reading at the domestic battery terminals said 13.4 volt. I also noticed the starter battery now showing 13.2 volt.

 

With the engine not running, I turned on all my domestic equipment again (20A or so) and noticed the voltage at the domestic battery terminals drop from 13.4 to to 13.1V.

 

I decided to run the engine to supply my needs and charge the batteries for a few hours more (1200 revs). After two hours, around 4pm, I stopped the engine again, turned off all the load (except the constant 0.1 ampere) and went away for two hours. On coming back in the early evening, the battery again measured 13.4V at the terminals. I felt there could be some surface charge left and applied some load to be certain to get rid of this – 20A for 5 minutes. I then turned everything off again and measured 13.3V at the terminals. I felt confident that the batteries were fully charged and that they also probably had been fully charged earlier in the day.

 

Overnight I used some lights and water, and of course the 12V fridge was humming. I was also running the Webasto diesel central heating system, which I know would draw around 4 Ampere all night. (Yes I know the central heating is not necessary at this time of year, but Winter is coming soon enough, so I need to know if this all works.)

 

Next morning, Friday around 10am I disconnected everything again and measured the batteries at the terminals. The result was, 11.9V. I take this to mean the state of charge was less than 20%... Not good.

 

If the batteries' capacity is still 330AH (remember they are almost new), and if they were fully charged yesterday at 4pm, I must have used some 250AH by 10am the morning. In other words, my average consumption over 18 hours must have been close to 15A. That is more than I would expect, but not impossible. In any case, it seems I need more batteries, and some monitoring equipment to keep me much better informed.

 

Could it even be possible that my batteries are damaged? I know that just after I bought them, I went away for 5 days and forgot to turn off the fridge. When coming back, the batteries were flat. But I would hope they can survive a single incident like that. They are supposed to be good deep cycle batteries. Alternatively, because the batteries are quite new, perhaps they have not yet been cycled enough to reach their full capacity.

 

Sven

[Sir Nibble]

If you are serious about this then your first act must be to equip yourself with instruments to measure voltage and current. Readings taken under different conditions will allow experts to "read" what your system is achieving now, identify possible improvements, and check that those improvements have been achieved. Do not go out and buy bolt ons yet even if others tell you about the improvements they have had. If you can read numbers off your alternators or take photos that would help too, saying "Iskra" tells us no more than "Beta".

 

Cross posted, now I have seen your second post.

First, the volt drop between alternator and battery is unacceptably high, sorting that will make a difference.

Second, just because the charge voltage has stabilised does not mean batteries are charged. I suspect you are charging your batteries to 25% and then discharging to 20% (ish).

[KenK]

Hi Sven,

Beta do use Iskra alternators. The engine one is probably a 40 Amp unit and should be battery sensed. The domestic could be either a 70 Amp machine sensed unit or a 100 Amp battery sensed. Beta used to modify the domestic alternator with a device to improve the charging, it turned the smaller alternator into a battery sensed unit and improved the charging regime, it was certainly in use six years ago. My boat has the 40 and 70 amp Iskra alternators with the device. I fitted the Smartgauge / Smartbank system last year and I can certainly recommend it. Iskra do have a UK website if you need more information.

 

cheers

 

Ken

[Sven...]

OK so here are pictures...

 

First an overview picture of the engine and alternators:

 

 

Here is the domestic "Iskra" alternator...it seems there is a label on it, but this has been overpainted and some cleaning of the label did not help:

 

 

Again the Iskra:

 

 

Connection point between the alternator and the batteries:

 

 

Batteries - the white one is the starter battery. The heavy red cable on top goes to the inverter.

Yes I know it should ideally be connected at the other side of the main switch..:

 

 

The smaller starter battery alternator - again no inscriptions except a logo on the back that looks like "ND":

 

[Proper Job]

I'm sure you'll get a lot more advice but here's my two penny worth.

 

I consider myself as a low consumer of power. 12v fridge during the Summer (cold box in the cockpit during Winter), a couple of hours a night TV/computer, lights (rarely more than one on at any one time) , pumps etc.

 

I have 5 x 110Ah domestic batteries. I charge via a generator & 40 amp Stirling charger.

 

I would suggest that you need to at least double your battery bank size, set the fridge thermostat to '1' to minimise compressor run time and generally become very aware of electricity usage.

 

A big recommendation from me would be: fit solar panels. I have 2 x 85W and an MPPT controller. I haven't run my generator for months (but I do cruise a lot at weekends).

 

As a wide beam you'll have plenty of room on the roof.

 

Fit a smartgauge, decent ammeter (volt meter is part of the smart gauge). I don't let my batteries go below 50% SOC and charge up to 80/85% SOC. Obviously if I'm out cruising they are charged up to 100% via the engine alternator.

 

Edited to add:

Make sure your batteries are wired up in a 'balanced' way. See the smart gauge web site for diagrams.

Edited September 10, 2010 by Proper Job

[RLWP]

Putting aside the issues around fires caused by your inverter cables...

 

The negative terminal on the starter battery needs cleaning up, it's growing corrosion underneath

 

Richard

[blackrose]

Here is something I saved, originally posted by Chris W.

 

Determining Battery Bank % Charge

The simplest and a really good way in terms of accuracy to determine battery bank charge in terms of % is simply to measure the OFF-LOAD battery bank voltage (off-load meaning that there must be no equipment or appliances running from the batteries at the time of testing and the batteries must not be on charge).

 

For wet lead-acid batteries the figures are as follows:

 

12.7v = 100% charge

12.6v = 90% charge

12.5v = 80% charge

12.4v = 70% charge

12.3v = 60% charge

12.2v = 50% charge (never let your batteries fall below this for a long life)

12.1v = 40% charge

 

(For other types of battery similar tables are available.)

 

 

Subtracting the above measurement from 100% will tell you the amount of % charge remaining.

 

Ideally, voltage readings should be taken 24 hours after any charging or discharging of the batteries, but of course this is often impractical or impossible, especially for liveaboards.

 

If taking readings directly after charging, switch on the tunnel light for 5-10 minutes to disperse what is known as the "surface charge" on the battery plates, otherwise you will get an artificially high reading.

 

Again, all readings must be OFF-LOAD or they will mean nothing.

 

Voltage readings are temperature dependant. The figures given above relate to readings taken at an ambient temperature of 22 deg C. You will need to add 0.09v for every 5.6 deg C decreased temperature, or subtract 0.09v for every 5.6 deg C increased temperature.

 

One other proviso: because you are trying to measure 0.1v you need an accurate voltmeter with at least +/- 0.5% accuracy. A lot of the cheap meters are +/- 2% so they are no good for this fine measurement.

 

 

Battery Charging Regimen

As well as making sure your batteries never fall below the 50% capacity shown above, another thing to avoid is discharging the batteries by less than 5% (12.65v) before recharging – it’s better to discharge a bit more.

 

Finally, when recharging your batteries always try to recharge fully.

 

 

Edit: Battery Monitors (This is me, not Chris W)

I have a Smartgauge and a BEP DC monitor. The Smartgauge gives battery voltage and state of charge as a % while the BEP monitor gives battery voltages, state of charge (not very accurately) and amps in & out.

 

Personally if I could only have one of these battery monitors I would definitely choose the Smartgauge. It's nice to be able to see amps going in when you're charging and amps out as you switch things on, but it's not essential (unless of couse you need to know how many amps your alternator is putting out).

Edited September 10, 2010 by blackrose

[Gibbo]

It makes a change to get such comprehensive information. We usually get a few snippets, half of which are wrong anyway

 

1. The cabling from the domestic alternator is too thin, the isolator switch (if fitted) may also be introducing losses.

2. The domestic alternator is not pulley'd up properly. You need a bigger pulley on the engine. This is a very common problem with Betas.

3. Your batteries could well have lost substantial capacity. Just 1 discharge down to 0% will knock 25% off the actual capacity. Possibly for ever.

4. I don't think you have enough batteries for your useage. Working from home using a laptop along with everything else I'd be looking at doubling them.

 

That's a start. There are a million other things you could do to help but from your readings I can be certain that adding an alternator controller will make not one jot of difference. As is the norm with modern alternators.

 

Sorting out those losses between the domestic alternator and the batteries will make a huge difference.

 

After discharging, waiting 30 minutes or so will get rid of the surface discharge and allow a reaosnable estimation of state of charge to be had from the voltage. Following a charge you need to wait a lot longer. Like about 24 hours as an absolute minimum. Even up to 72 hours with certain batteries. Taking a reading after a few hours is pointless.

 

PS. Get used to the idea that recharging from 50% to 100% is absolutely not going to happen in less than about 6 to 8 hours. And that's an optimistic time. This is absolutely regardless of what equipment you install and completely regardless of what anyone else tells you.

 

 

Voltage readings are temperature dependant. The figures given above relate to readings taken at an ambient temperature of 22 deg C. You will need to add 0.09v for every 5.6 deg C decreased temperature, or subtract 0.09v for every 5.6 deg C increased temperature.

 

 

All good info apart from this bit which is wrong.

 

The temperature coefficient of a lead acid battery is extremely tiny. It's actually zero when fully charged and still only 430uV/°C/cell at almost flat.

 

In effect this means that even a huge temperature change of 20°C on a battery at 75% SoC will affect the terminal voltage by less than 0.04 volts.

 

For the purposes of checking battery SoC it can be completely ignored.

 

I'm not sure where Chris got those figures but they wrong.

Edited September 10, 2010 by Gibbo

[RLWP]

 

It looks to me that there are corrosion products between the cable tags too on top of the battery clamp

 

Richard

[Dovetail]

The only thing I can add and probably already mentioned is that your three batterys are a max of 330 AH and you will only get approx half that from the batterys so perhaps say a max 170 AH before needing to be charged.

[Keeping Up]

 

...

 

All good info apart from this bit which is wrong.

 

The temperature coefficient of a lead acid battery is extremely tiny. It's actually zero when fully charged and still only 430uV/°C/cell at almost flat.

 

In effect this means that even a huge temperature change of 20°C on a battery at 75% SoC will affect the terminal voltage by less than 0.04 volts.

 

For the purposes of checking battery SoC it can be completely ignored.

 

I'm not sure where Chris got those figures but they wrong.

I suspect he may have got them from this page on the Adverc site - that appears to be the value they have used in their graph. I have long suspected that my Adverc reduces the charging voltage too much when the engine compartment warms up

[nb Innisfree]

I agree with Gibbo (who would disagree!) when your charge voltage reaches 14.6/14.8 you need to carry on charging for at least 8 hours to get your domestics anywhere near 100% SoC

 

To minimise fuel usage you could charge to 14.8v daily then stop but do an 8-10 hour charge weekly, this won't give you maximum battery life but it will be a reasonable compromise between battery life and fuel usage.

[Arnot]

I am thinking I am often wasting diesel or undercharging my batteries, and I probably need more batteries and/or a battery management system anyway.

 

• If I were to install some battery regulator/management/monitoring/split charge system, what setup would you recommend? I'd like to save on diesel and not undercharge or overcharge, and being a bit of a geek, I wish to have as much status information as possible. I would also like to be able to charge the batteries from the generator or shoreline once in a while.
  
• Should I install more batteries – and how many?
  
• Any other advice as to how to improve my system?
  

 

All the best,

Sven

Congratulations Sven, I think that this first time a query of this type has had almost all the information necessary for a considered reply. It does my heart good to see it (and Gibbo’s pacemaker )

 

So, let’s do the obvious stuff first.

 

The starter battery charging is fine, just don’t worry about it. You need so little power to start an engine (usually less than 0.2Ah) that a bicycle dynamo would put it back in in minutes.

 

As an observation, it would be a good idea to check your fanbelt tensions. From the images there seems to be a bit of rubber dust around (although this may not be recent) and that is usually a sign of slipping belts. As a rule of thumb, if you can make the belt slip by turning the pulley with a spanner then the belt is too loose. This test should turn the engine.

 

All the measurements you did are good and useful but it might help to check the voltage between the domestic alternator casing and the domestic battery negative. This will tell you how much you are losing through the negative path. Any more than about 0.2 volts at full output could be improved on.

 

The domestic altenator is an Iskra and generally they are pretty good. The one you have should be able to just about reach 120A with perfect wiring but most will do 90A. Another plus is that they are geared to give quite a high output at reasonbly low revs. The ones I have tested don’t seem to give amy more after 1300 rpm. It would be well worth while adding a new cable from the domestic alternator output to the domestic battery directly avoiding the existing engine wiring. I would suggest 25mm csa cable the slight benefits from using large cable tend to be negated by not being able to terminate it with the correct stud diameter. This should reduce the voltage drop from the alternator to the batteries you have measured but do check this when the job is done.

 

I don’t actually agree with Gibbo about the gearing, my experience is that if they are geared up further then it tends to make the idling unstable as the alternator regulator hysterysis beats with the mechanical diesel pump governor and I think that the benefit would be disproportionate to the cost.

 

Your batteries are probably still OK but it is true to say that totally flattening them will reduce their service life a bit. It’s a bit of a generalisation but batteries tend to have a limit to the power they can store and release in their life so the deeper you discharge them the fewer cycles you get. With a fairly standard leisure battery, discharging to 80% down regularly on a daily basis would probably translate to a battery life of about a year, perhaps slightly less.

 

You already know that it is a rule of thumb to have a battery that will support the nomal daily usage without discharging more than 50%. Smaller than this tends to lead to premature failure, larger than this tends to batteries dying from old age before they are worn out. My experience is that liveaboards with your sort of usage tend to need four to six leisure batteries.

 

I would suggest you invest in monitoring PDQ and given your obvious grasp of the nature of the problem and ability to interpret hard data results, despite the ease and simplicity of the Smartguage, I would suggest the Victron BMV-600S. It shows amps and volts, consumed energy in Ah and calculates from a theoretical capacity ceiling the time to go. It also stores some historical data and if you want to get really hi-tech, you can plug a laptop in an datalog the readings into Excel. I do a lot of government work and always specify this monitor because I can get some idea of what has gone wrong. Some of the current drain figures you quote have been based on assumptions about the battery capacity and may not be correct, a BMV-600S would give you hard facts. One of the major benefits of these is that you can easily see what minimum speed you need to run your engine at to maintain maximum charging and also when you can turn it off again.

 

If you want to charge your battery from a generator or shoreline occasionally then you really need a decent sized charger (50A plus) and these are expensive for a goodun. Given that you have a chinese inverter at the moment (and I don’t believe the 5kw for one moment) it might be a wise move to invest in a Mastervolt or Victron pure sine combi. It will give you a good charger, a good inverter and a low quiescent current draw. As well as this the automatic switching will keep the mains wiring simple.

 

Now the batteries. As an alternative to merely adding a couple of leisure battereis, given that you have just bought the boat and use it to live and work on it would be worth considering a big investment in this area if you can afford it. Leisure batteries are cheap and cheeful and provide a solution to the power storage problem of sorts. However, if you not only rely on your power system for living and earning but use it heavily (as you do) then 2v storage cells would be a better and cheaper solution in the longer term. A 600Ah bank installed would probably set you back about £2K but you would be able to use the top 80% safely so this equates to a 1000Ah leisure bank. In addition, it would last ten to fifteen years and the low impedance would allow you to top up the power far more quickly thus saving fuel. One of my customers reports a saving of 40% plus on fuel costs!

 

Especially if you intend to invest in storage cells now or in the medium future, an external regulator for your domestic alternator would be a good investement but it doesn’t need to be a whizzy multi stage digital doohdah with lots of LED’s. A simple external regulator that can be set to the voltage appropriate the the battery bank you have will do just fine. In addition with a larger battery, a Smartbank to parallel the alternators would bring down charging times.

 

Hope this helps…

 

Regards

 

Arnot

[Gibbo]

I suspect he may have got them from this page on the Adverc site - that appears to be the value they have used in their graph. I have long suspected that my Adverc reduces the charging voltage too much when the engine compartment warms up

 

Yes I would agree that's where he's got the figures from. But he's interpreted them wrong.

 

Those are the ideal adjustments to the charge voltage that are required under varying temperature conditions.

 

They have got nothing whatsoever to do with the off load terminal voltage of the battery.

 

Two completely unrelated subjects.

[Gibbo]

I dont actually agree with Gibbo about the gearing, my experience is that if they are geared up further then it tends to make the idling unstable as the alternator regulator hysterysis beats with the mechanical diesel pump governor and I think that the benefit would be disproportionate to the cost.

 

I have to pull you up on this. When the alternator is producing full output (ie when the high gearing is actually needed) the regulator will be switched hard on. There will be no regulator action.

 

If you've seen this unstable idling at highish load it's purely because the engine won't produce enough power at tickover or the governor is sluggish. Every time I've seen this (which is a lot) it's been cured with a service. Cleaning the crud off bob weight pivots sorted it on my Gardner

 

 

Your batteries are probably still OK but it is true to say that totally flattening them will reduce their service life a bit. Its a bit of a generalisation but batteries tend to have a limit to the power they can store and release in their life so the deeper you discharge them the fewer cycles you get. With a fairly standard leisure battery, discharging to 80% down regularly on a daily basis would probably translate to a battery life of about a year, perhaps slightly less.

 

 

And this bit

 

I've just (as in over the last 3 weeks) dragged two leisure batteries (Yuasa L36-90) from an initial 80Ahr capacity down to 35Ahr capacity by dischagring them to 10% SoC 5 times. Yes they were properly charged etc. This is whilst developing the new wizzbang 4 channel embedded combined amp.hour counter/smartgauge algorithm/temperature monitoring/peukert meddling/state of health calculating military battery monitor and data logger

 

 

I would suggest you invest in monitoring PDQ and given your obvious grasp of the nature of the problem and ability to interpret hard data results, despite the ease and simplicity of the Smartguage, I would suggest the Victron BMV-600S. It shows amps and volts, consumed energy in Ah and calculates from a theoretical capacity ceiling the time to go. It also stores some historical data and if you want to get really hi-tech, you can plug a laptop in an datalog the readings into Excel. I do a lot of government work and always specify this monitor because I can get some idea of what has gone wrong. Some of the current drain figures you quote have been based on assumptions about the battery capacity and may not be correct, a BMV-600S would give you hard facts. One of the major benefits of these is that you can easily see what minimum speed you need to run your engine at to maintain maximum charging and also when you can turn it off again.

 

 

I agree that in this case an amp.hour counter is probably more useful. As long as he doesn't want to know the SoC without having to work it out for himself.

 

PS.....

 

OP. I've just noticed your battery isolators. You need to change them. They're not even close to being man enough. I suspect they account for a good lump of the voltage difference you're seeing between the alternator and the domestic battery.

Edited September 11, 2010 by Gibbo

[smileypete]

Yes I would agree that's where he's got the figures from. But he's interpreted them wrong.

 

Those are the ideal adjustments to the charge voltage that are required under varying temperature conditions.

 

They have got nothing whatsoever to do with the off load terminal voltage of the battery.

 

Two completely unrelated subjects.

What do you think of this table of open circuit voltage in the Battery FAQ? http://jgdarden.com/batteryfaq/carfaq4.htm#ocv_soc

 

Gives about 0.044V difference from 4°C to 27°C regardless of SoC.

 

cheers,

Pete.

Edited September 11, 2010 by smileypete

[nb Innisfree]

As regards full alternator output at tickover speed, surely if the prop takes, say, 3BHP at tickover and alternator takes, say, 3BHP and engine produces less than that total at tickover then alt gearing needs to be reduced? Our alt at 2:1 reduction gives full output and absorbs approx 6 BHP at startup with 1100RPM, prop seems to absorb a similar amount, this overloads the engine slightly so I need to increase revs to 1200 until charge rate reduces, if I wanted to idle at 1100RPM or less I would have to reduce alt pulley ratio. So therefore alt pulley ratio is dependant on power output of engine at tickover?

 

What does puzzle me is the fact that when our (24v) alt is on full 100 amp output at 1200RPM and beginning to charge domestics and w/m heater cuts in then we have to increase revs considerably to maintain full output (from 1200RPM to approx 1800RPM or more)

[Arnot]

I have to pull you up on this. When the alternator is producing full output (ie when the high gearing is actually needed) the regulator will be switched hard on. There will be no regulator action.

 

If you've seen this unstable idling at highish load it's purely because the engine won't produce enough power at tickover or the governor is sluggish. Every time I've seen this (which is a lot) it's been cured with a service. Cleaning the crud off bob weight pivots sorted it on my Gardner

 

True enough as stated but I didn’t mention the alternator loading. The problem I have noticed tends to occur when the batteries are almost charged and the regulator is having to work for it’s living. Under these conditions, at lower engine speeds and fast idle, the engine often wangs about like a demented zebedee. The result is that all the wiring, control cables and pipework to the engine take a battering and often the control wiring plug comes undone.

 

No doubt with a Gardner that has a governor response that can be plotted on a calendar it would be a different story…

 

And this bit

 

I've just (as in over the last 3 weeks) dragged two leisure batteries (Yuasa L36-90) from an initial 80Ahr capacity down to 35Ahr capacity by dischagring them to 10% SoC 5 times. Yes they were properly charged etc. This is whilst developing the new wizzbang 4 channel embedded combined amp.hour counter/smartgauge algorithm/temperature monitoring/peukert meddling/state of health calculating military battery monitor and data logger

 

Interesting (and scary)! I may well stand corrected here. I have seen this problem before but not in every case by any means. Do you think this reduction in capacity is permanent or may they recover with some nursing?

 

I have four Mastervolt 270Ah AGM’s here under test and they have collapsed to about 20Ah on a C20 load. I also have four Elecsol 270Ah “carbon fibre” leisure’s with a similar problem at under a year old. Both have ample current output when charged but just won’t hold it.

 

I also have twelve Squadron Energy 270Ah gel batteries that were installed in a 24V network for three years and discharge to 50% or lower six days a week for all this time. These mostly still have a reasonable capacity but won’t deliver even moderate currents (C20) without the voltage collapsing to the mid tens after about 20Ah but then they carry on for ages, some to as high as 230Ah.

 

I am still testing these (12 x20 hours takes a while on a single channel logger!) because they were installed without balancing links and I wanted to see if there was any correlation between the battery condition and their place in the network.

 

Regards

 

Arnot

[Gibbo]

What do you think of this table of open circuit voltage in the Battery FAQ? http://jgdarden.com/batteryfaq/carfaq4.htm#ocv_soc

 

Gives about 0.044V difference from 4°C to 27°C regardless of SoC.

 

I'll ask Bill (owner of batteryfaq) where he got that info.

 

It's pretty close at 0% SoC showing a coefficient of 367uV per °C per cell. I'm guessing that someone has come across that data at 0% SoC then simply translated it to apply across the range to make up that table not realising that it doesn't apply.

 

My own tests actually resulted in a temp coefficient of 420uV at 0% SoC and zero at 100% SoC. From this reference....

 

http://wiki.xtronics.com/index.php/Sealed_Lead_Acid_Battery_Applications

 

"I always worried about the Temperature coefficient of lead-acid batteries. dE/dT turns out zero in a fully charged battery. There is a temp-co for a discharged battery but dE/DT = -0.000 43 V/C (per cell) so it can be safely ignored. I think others have confused batteries dE/dT with dI/dT which does have a noticeable temp-co. "

 

I'm as certain as it's possible to be that this information is correct as it agrees almost exactly with my own test results and what are the chances of two completely independant people coming up with almost exactly the same results unless they are both correct?

 

Further, if it is wrong, several devices that I know of wouldn't work as they actually rely on this being correct!

[Gibbo]

True enough as stated but I didn’t mention the alternator loading. The problem I have noticed tends to occur when the batteries are almost charged and the regulator is having to work for it’s living. Under these conditions, at lower engine speeds and fast idle, the engine often wangs about like a demented zebedee. The result is that all the wiring, control cables and pipework to the engine take a battering and often the control wiring plug comes undone.

 

No doubt with a Gardner that has a governor response that can be plotted on a calendar it would be a different story…

 

 

But under those conditions there will be very little load on the engine. Admittedly the load will be in pulses, but the pulses will be in the tens of milliseconds range and I can't really see how even the flywheel would struggle to cope with that let alone the engine. I don't know what's going on there.

 

 

Interesting (and scary)! I may well stand corrected here. I have seen this problem before but not in every case by any means. Do you think this reduction in capacity is permanent or may they recover with some nursing?

 

I have four Mastervolt 270Ah AGM’s here under test and they have collapsed to about 20Ah on a C20 load. I also have four Elecsol 270Ah “carbon fibre” leisure’s with a similar problem at under a year old. Both have ample current output when charged but just won’t hold it.

 

I also have twelve Squadron Energy 270Ah gel batteries that were installed in a 24V network for three years and discharge to 50% or lower six days a week for all this time. These mostly still have a reasonable capacity but won’t deliver even moderate currents (C20) without the voltage collapsing to the mid tens after about 20Ah but then they carry on for ages, some to as high as 230Ah.

 

I am still testing these (12 x20 hours takes a while on a single channel logger!) because they were installed without balancing links and I wanted to see if there was any correlation between the battery condition and their place in the network.

 

 

I suspect they may come back with some TLC. I know these "leisures" won't as I pulled one apart to find the bottom full of shed material.

 

The ones that show a sudden drop in voltage but then continue to produce lots of amp.hours sound like a semi b&gg*red cell. I don;t think they'll ever be right again.

 

You'd love our new 4 channel one with logger

 

 

[Arnot]

I suspect they may come back with some TLC. I know these "leisures" won't as I pulled one apart to find the bottom full of shed material.

 

The ones that show a sudden drop in voltage but then continue to produce lots of amp.hours sound like a semi b&gg*red cell. I don;t think they'll ever be right again.

 

I think you are probably right here it certainly looks like cell reversal given the rapid decline from 12.5ish to 10.5ish. Anyway, when I have finished cooking the last six batteries I am going to see if I can work out which cell(s) in each of them has departed this world. I suspect that there may be a pattern that emerges, what would your guess be?

 

Would you be interested in the test data?

 

You'd love our new 4 channel one with logger

Cor! Battery porn! It looks magnificent, you would cry if you saw my setup. You don't want a beta tester do you?

 

Regards

 

Arnot

[Gibbo]

I think you are probably right here it certainly looks like cell reversal given the rapid decline from 12.5ish to 10.5ish. Anyway, when I have finished cooking the last six batteries I am going to see if I can work out which cell(s) in each of them has departed this world. I suspect that there may be a pattern that emerges, what would your guess be?

 

Would you be interested in the test data?

 

 

If it's logged properly I'd love it. The more I have the more odd things I spot.

 

Cor! Battery porn! It looks magnificent, you would cry if you saw my setup. You don't want a beta tester do you?

 

We've got one. They just don't know it

[Denis R]

OP. I've just noticed your battery isolators. You need to change them. They're not even close to being man enough. I suspect they account for a good lump of the voltage difference you're seeing between the alternator and the domestic battery.

 

I think we've probably been down this street before (several times), but could you give a steer on which isolator you reckon would be the 'isolator of choice'?

[Gibbo]

I think we've probably been down this street before (several times), but could you give a steer on which isolator you reckon would be the 'isolator of choice'?

 

Lucas 219SA is my favourite for these size installations. Durite do a clone which is almost as good but not quite.

 

The ones shown in the photos I have destroyed at 30 amps after a couple of hours. 100 amps in less than 5 minutes. They are utter shyte.