Designing 12v charging system from scratch

[rusty69]

2 minutes ago, Mike the Boilerman said:

 

Yep, a 100% turnaround in attitude once a bit of publicity begins to hit home.

No PM yet from this chap to me asking if he can come and collect mine though. FAR too much trouble, that would be. I have to do all the running for THEIR faulty product.

Oh, they sent a couple of scantily clad female santa's to collect mine. Complete with reindeers and sleigh bells.

Ho!Ho!Ho!

[WotEver]

1 hour ago, Mike the Boilerman said:

I just feel the need to counter the relentless "SG is perfect" message/mantra pumped out continuously by WatEver.

Perhaps you would like to point us to such posts?

Or perhaps it’s just another one of your unfounded assertions...

[Tony Brooks]

46 minutes ago, Capnbob said:

Very happy to learn, Tony.  I have a feeling this was summarised somewhere, but can't track the process down now.

I am sure you actually want to learn but others reading this topic at some time may not want to or be able to. In that case the single gauge most likely to keep them on the right track is the Smartguage

Perhaps the most important thing to grasp is the reading the battery voltage can not be an accurate indication of state of charge unless when and how it was taken is tightly controlled. It must be the rested voltage and as the time it will take for a battery bank fresh off charge to lose its surface charge is likely to vary we can only give very rough guidance. So to get to the rested voltage you need:-

1. No solar input or charging from any source.

2. The batteries just sitting doing nothing for at least an hour and probably several hours  OR a load of a few maps put on the battery for a short time, say a quarter to half an hour.

3. NO LOAD on the batteries when the reading is taken.

Once you achieved that the voltage will indicate the state of charge. The meanings of the values will vary form battery make to battery make but as a fair guide:

12.7 to 12.8 = fully charged.

12.5 consider half charged for optimum battery life.

12.2 to 12.3 consider fully discharged for optimum battery life

Try not to go below 12.2 very often.

However you also need to take notice of trends so I like to take the battery voltage as I turn in at night and when I get up when it is still dark (solar on the boat).

The morning reading will tell me if I MUST recharge although its best to fully recharge every day if you can.  The evening reading gives an idea of how well charged the batteries were. Once you have a few readings under your belt you can start to spot trends than may indicate a problem with charging, a faulty battery, or excessive use.

As has been said the ammeter is used to tell you when the batteries are more of less fully charged by measuring the tail current. At around 14.4 to 14.5 volts (typical alternator charging voltage) the tail current should be about 1% to 2% of the bank's capacity to   suggest the bank is as fully charged as it is likely to get. At other voltages the fully charged point can be deduced by noting when the currant has failed to fall over one or two hours.

Do not get confused by a multi-stage charging source dropping its voltage as it goes into float. This will cause a corresponding drop in charging current.

Now we ill have people questioning the values I have put down but they are a GUIDE not set in stone. Much of batteries and charging is only rough guides so those who wish to obscure the issues have plenty of opportunity.

[WotEver]

29 minutes ago, Mike the Boilerman said:

1) You can't establish SoC whilst charging with any degree of accuracy.

Not strictly true. If you know the capacity of the bank then you can get a pretty good estimation of SoC simply by looking at the charge current, assuming your charge voltage is somewhere above 14.4V. Amp-hour ‘law’* covers it. 

* It’s not a law at all, simply a rough (but pretty accurate) guide. 

[cereal tiller]

20 minutes ago, WotEver said:

Not strictly true. If you know the capacity of the bank then you can get a pretty good estimation of SoC simply by looking at the charge current, assuming your charge voltage is somewhere above 14.4V. Amp-hour ‘law’* covers it. 

* It’s not a law at all, simply a rough (but pretty accurate) guide. 

As long as you Account for Amps. being drawn by other loads on the System while Charging.

[WotEver]

8 minutes ago, cereal tiller said:

As long as you Account for Amps. being drawn by other loads on the System while Charging.

That’s why I wrote ‘charge current’. I was referring strictly to the current flowing into the battery, not the current being supplied by the charging source. 

[WotEver]

41 minutes ago, Tony Brooks said:

Now we ill have people questioning the values I have put down but they are a GUIDE not set in stone. Much of batteries and charging is only rough guides so those who wish to obscure the issues have plenty of opportunity.

Well said

[cereal tiller]

3 minutes ago, WotEver said:

That’s why I wrote ‘charge current’. I was referring strictly to the current flowing into the battery, not the current being supplied by the charging source. 

Yes, of course.I might Swap my SG for MtB's without him Knowing? could Produce interesting results

[MtB]

On 10/12/2017 at 11:27, WotEver said:

Amp-hour ‘law’* covers it

So what is the Amp-hour law then please?

A brief google for the term seems to focus only on Peukert.

On 10/12/2017 at 12:04, cereal tiller said:

Yes, of course.I might Swap my SG for MtB's without him Knowing? could Produce interesting results

 

I'd be up for that provided you have actually checked the calibration of yours using a DVM previously checked against a reference source, and found it correct.

Have you?

It would be a most interesting experiment, so see how fast you trash your own battery bank!

[cereal tiller]

6 minutes ago, Mike the Boilerman said:

So what is the Amp-hour law then please?

A brief google for the term seems to focus only on Peukert.

 

I'd be up for that provided you have actually checked the calibration of yours using a DVM previously checked against a reference source, and found it correct.

Have you?

It would be a most interesting experiment, so see how fast you trash your own battery bank!

I sense a Smoke Screen ,Manifesting

[MtB]

On 10/12/2017 at 12:16, cereal tiller said:

I sense a Smoke Screen ,Manifesting

 

The smoke screen is yours. You avoided answering my question.

You have no idea if yours is correctly calibrated, have you?!

[WotEver]

19 minutes ago, Mike the Boilerman said:

So what is the Amp-hour law then please?

A battery will demand (pretty closely) the same current as the missing charge in Ah. 

So, using simple quantities, a 100Ah battery at 80% SoC will demand approximately 20A charge current. That’s why tail current works. 

[MtB]

On 10/12/2017 at 12:30, WotEver said:

A battery will demand (pretty closely) the same current as the missing charge in Ah. 

So, using simple quantities, a 100Ah battery at 80% SoC will demand approximately 20A charge current. That’s why tail current works. 

 

That's most interesting and useful. Thank you!!

So approaching it the other way around, if I discharge my battery to 80% and bung it on charge, I can use this to calculate the actual capacity (provided it goes straight to absorption at 14.4v), yes?

Now, how can I find out when mine are at 80%....?

I know, I'll use my calibrated DVM!

[WotEver]

5 minutes ago, Mike the Boilerman said:

So approaching it the other way around, if I discharge my battery to 80% and bung it on charge, I can use this to calculate the actual capacity (provided it goes straight to absorption at 14.4v), yes?

I can’t see why not. 

How are you determining 80%?

[MtB]

On 10/12/2017 at 12:43, WotEver said:

I can’t see why not. 

How are you determining 80%?

 

Won't the Smartgauge tell me? 

[WotEver]

20 minutes ago, Mike the Boilerman said:

 

Won't the Smartgauge tell me? 

Yes, absolutely. But not yours, they’re broken and you’ve not bothered to return them. 

[MtB]

On 10/12/2017 at 13:11, WotEver said:

Yes, absolutely. But not yours, they’re broken and you’ve not bothered to return them. 

 

Ah but Cereal Tiller wants to swap his suspected good Smartgauge for one of my broken ones. 

He's gone all quiet now though...

[nicknorman]

52 minutes ago, Mike the Boilerman said:

 

That's most interesting and useful. Thank you!!

So approaching it the other way around, if I discharge my battery to 80% and bung it on charge, I can use this to calculate the actual capacity (provided it goes straight to absorption at 14.4v), yes?

Now, how can I find out when mine are at 80%....?

I know, I'll use my calibrated DVM!

It’s a pretty rough rule of thumb. If you discharge to 80% SoC and then put on charge, you will initially get much more than 20% of the capacity as current. If you discharge to 50% SoC and then recharge, passing 80% you will have less charge current than if you discharged to 70% and then recharged. This of course is due to the accessibility or otherwise of the chemicals deeper within the plate. The first part of the charge does the surface bit which is easy, if the batteries have been deeply discharged getting the charge into that bit is a lot slower.

Similarly battery temperature has quite a big effect. I’ve noticed our charging current mid-charge is a lot less now, than in summer.

All that said, it is a useful rule of thumb but only if you know “how you got to this point”. Fundamentally it reflects the gradually diminishing capability of a battery to accept charge as its SoC increases.

[Iain_S]

1 hour ago, Mike the Boilerman said:

 

Answering your questions in order...

1) You can't establish SoC whilst charging with any degree of accuracy. The best you can do is measure the tail current to see when you are approaching fully charged. For this you use the ammeter. A voltmeter can be used to estimate the state of charge of a battery rested for half an hour, by accurately reading the voltage and looking it up on a chart. Or memorising the chart once you've done it a lot!

2) Tail current is the charge current as the battery approaches the mythical state 'fully charged', when being charged at a specific voltage specified by the battery manufacturer, usually (but not always) 14.4.

3) A Smartgauge might confidently display 100% when the battery is nothing like fully charged. This is the trap. So use 'tail current' instead. The main use for a SG is to know when to start charging. When it gets down to 50%, start charging. Also after two days of use, start charging regardless. These 'rules' are compromises really, as in a perfect world you must "fully recharge you battery after every discharge" to roughly quote the Trojan manual. 

I have fitted Smartgauges to three boats. The installations are slightly different in that the first one has Smartbank, but not connected to Smartgauge, the second has Smartgauge connected to Smartbank, which was modified to Smartbank Advanced, and the third has Smartgauge only(Two alternators, so no split charge system needed). I have never checked against a calibrated Votmeter, but the voltage readings agree with my meter and, in the third case, agree with the Merlin Powergauge (an Amp Hour counting meter) which was fitted originally. 

Boat one was/is a shared boat. In our normal use, the batteries were fairly fully charged when the boat was in use, but tended to be ill treated at changeover time, as typical use was a short day on the last day of the trip, along with extensive use of the hoover, followed by a short day with the next owner. The Smartgauge was not really used as to establish that the batteries were fully charged, but was very useful in ensuring that they were not left excessively discharged at changeover time.

Boat two tends to do relatively short days, and the engine is run at times to charge the batteries. If they are taken to 100% on the Smartgauge, they will typically drop to 99% or 98% after an hour or so. If the engine is run for an hour or two after the 100% is first indicated, the batteries will still be at 100% after a couple of days with master switches off. I conclude that when charging, that Smartgauge is within 2%.

Boat three is similar to boat one, but it typically takes a couple of days for the batteries to start behaving themselves, and not show excessive discharge over night. Comparison between the Smartgauge and the Powergauge indicates that the battery capacity increases a fair bit over the first three or for days use, so it is likely that the Smartgauge is a bit inaccurate, at least at first. However, the batteries are borderline. They were fitted by the boatyard, but the boat was left for several weeks with the cabin master switch on, resulting in a continuous discharge of just under an amp, and totally flat cabin batteries. The remaining capacity, by comparison of Powergauge AmpHours out and Smartgauge percentage is about a third of the bank's nominal capacity, and so Smartgauge would not be expected to be accurate. (I'm sure it's in the manual somewhere that Smartgauge doesn't work properly when batteries have lost 50% capacity)

So in answer to Mikes 1) : It's fairly accurate, and after charging, will regain its accuracy.

2) If Smartgauge is displaying 100%, the batteries are near full charge.

The above assumes that the Smartgauge was correctly calibrated at the factory 

[cereal tiller]

36 minutes ago, Mike the Boilerman said:

 

Ah but Cereal Tiller wants to swap his suspected good Smartgauge for one of my broken ones. 

He's gone all quiet now though...

Been out Shopping,Offer still on.......

[rusty69]

6 minutes ago, cereal tiller said:

Been out Shopping,Offer still on.......

As is my offer Mtb of your two SG's in exchange for crisp ten pound note.

[cereal tiller]

1 minute ago, rusty69 said:

As is my offer Mtb of your two SG's in exchange for crisp ten pound note.

You are still thinking about those Scantily clad Ladies? go easy on those Tablets ,the ones labelled Deludamol?

[rusty69]

1 minute ago, cereal tiller said:

You are still thinking about those Scantily clad Ladies? go easy on those Tablets ,the ones labelled Deludamol?

Well, as long as Mtb doesn't personally deliver them dressed as a scantily clad santa, I don't care.

I don't know what mtb looks like, but suspect you do, and now have an indelible image in your head

[MtB]

47 minutes ago, cereal tiller said:

Been out Shopping,Offer still on.......

 

And the answer to my question is...?

[cereal tiller]

10 minutes ago, Mike the Boilerman said:

 

And the answer to my question is...?

42