Interesting read on battery SOC measurement.

[wincrasher]

Of course they are trying to sell you on a Smartguage. But this was referred to me as a better way of measuring SOC and you don't need a shunt.

Smart Gauge Battery Monitoring Unit Photo Gallery by Compass Marine How To at pbase.com

So what do you battery guru's make of this?

 

[avanti]

Yeah, I've seen that before. It is interesting. I have no data at all on this device, but I don't find it totally implausible that you could keep track of SOC via constant monitoring of voltage combined with a clever algorithm. In any event, Balmar is certainly not a lightweight brand in this area.

The device's manual may be downloaded here:
http://www.balmar.net/PDF/SMARTGAUGE MANUAL.pdf

 

[booster]

I have also seen that before, and have to agree with Avanti, can't cast away out of hand, but an interesting idea.

It is interesting to see how they describe things. SOC is the important thing to know, as they state, in relation to battery life in particular, but when it comes to how much time you have left to run, it is not as good because of the mentioned capacity changes with age, temp, etc. Without an ammeter, I would also be interested in how they determine totally full, and do they walk down the SOC as the battery is aging, when it could be brought back somewhat by a very full charge. All speculation, of course, and for the masses probably a good thing to have. Personally, I want an ammeter so I can see use at any time, and AH down, as well as control charging by amps, but others want simpler.

The inaccuracy he talks about is certainly there on the AH based monitors both from capacity reduction and especially on the recharge side. The recharge side is made much better by the auto reset to full based on amps and voltage that most of them do, and if you discharges are pretty consistent their auto calibrating charge efficiency is a not feature for being more accurate on the recharge side.

There is also one other thing about using the SOC exclusively, and that is at what discharge rate is it calculating. We had a very interesting thread here a while ago that addressed Peukert, and why it didn't really apply well to our uses because the discharge rate varies so much over the discharge cycle, and is rarely at the 20 hour rate. Bottom line was the the total AH used to a given SOC will always be the same at all discharge rates, as long as the very end of the discharge is at the 20ah rating amps. At that point if you are discharging faster, you will get less ah out when it gets to the rated end voltage if you are above the 20 hour current, and more amp hours if you are under the 20 hour rated current. The meter in question here would have no way of knowing your future discharge rate, so it would probably have to base it's calcs on past history somehow, and average discharge really doesn't cover it.

The talked about battery health monitoring, but I wonder how it tells you. If it gave you the actual reduction in capacity you have, now that would be very useful.

 

[Jostalli]

This is from the same guy who taught me everything about LiFePO4 batteries.

 

[GeorgeB-ClassB]

"Entertaining" reading for a novice in battery management science!

Let's hope, as booster is saying that the Peukert factor doesn't really apply well to our uses. Why couldn't we get a simple and reliable SOC meter like we have in our smartphones!

It was hard enough to program correctly my Trimetric (TM-2030) to get a very precise SOC.

Even to find the correct charged "set point voltage" which is supposed to be "slightly lower than the absorption charging voltage". I set mine to 14.0V for my two AGM 190Ah 6V batteries. (Absortion between 7.05 to 7.2 v) Don't know if it is correct, but the setting I used gave us a SOC reading quite consistent with our usage.

Even if it is not at the ultimate and precise setting our Trimetric proved very useful to understand our energy usage and helped us relax when dry camping. Not being obsessed anymore with counterintuitive simple voltage reading.

Help is welcome if the "battery gurus" have suggestions for my settings!

The other obscure setting is the "charged setpoint amps multiplier". Left it at 2% because I couldn't find that kind of info on our POWER TEC AGM battery.

Also the "sneaker wire" danger. I think our shunt was wired properly, although I have doubts because when I disconnected the main ground, the solar panels were feeding the system through a small "feedback" positive from the solar charge controller (small red cable in the next battery bank photo) The solar charge system has it's own ground (Strange- I should get the wiring diagram to understand)

The batteries all marked prior to be taken off for winter storage.
(with the small "feedback" red positive cable from the solar charge controller

18.4 Amp (at 14.3 v): the maximum charging Amp reading we got with our 400W solar panels (no Amp draw -the fridges where closed). Guess it's because our batteries are obsessively monitored and never allowed below 60% so the solar controller never allowed more amps that 18.4 amp. And maybe also because 400W of solar will never output more than 18.4 amp at 14.3 Volt even when the controller gets the green light to output it's maximum amperage.

.

 

[booster]

Your voltage setpoint should be fine. All it is used for is to make sure that the system is actively charging. It will then check the amps to determine when the batteries are full. Those amps are what the charged amps multiplier is. These are also the amps we call "return amps", "ending amps", "tail amps", "absorption transition amps" in other threads that you probably have seen. It will take the % of the battery capacity you have programmed in for your case. For AGM, the two percent is probably a bit high, especially when new. Lifeline says .5% for full, and most will go even lower than that. At 2% you will get a fully charged light before the batteries are totally full. Neither of these settings affect the AH counting, just the charged light. What will affect the AH readout is the charge efficiency, but only on recharge side. You want to get it to the point that you are around 100% full when the fully charged light comes on. It will vary with depth of discharge, etc, so not totally accurate on recharge. It does reset to 100% as soon as you start to discharge so the discharge will correct, if you got a full charge.

The wiring for the shunt looks OK from what is visible. All the grounds should be one side, and only the battery negative on the other. No other wires should go to the battery negative.

On the solar leak. Are you saying the Trimetric still reads, or does the coach power up, with the battery ground off? If the Trimetric stays on, there is a problem with wiring, most likely. If the solar comes on and powers the van, it probably is wired OK but running without battery reference, because the grounds are tied at the non battery side of the shunt. On some controllers this is bad thing to do, unless the panels are shut off. The problem comes if you are in sun, panels on, and you connect the solar to the coach without any batteries. You can get a big voltage spike that many harm electronics on the van 12v wiring.

 

[GeorgeB-ClassB]

Your voltage setpoint should be fine. All it is used for is to make sure that the system is actively charging. It will then check the amps to determine when the batteries are full. Those amps are what the charged amps multiplier is. These are also the amps we call "return amps", "ending amps", "tail amps", "absorption transition amps" in other threads that you probably have seen. It will take the % of the battery capacity you have programmed in for your case. For AGM, the two percent is probably a bit high, especially when new. Lifeline says .5% for full, and most will go even lower than that. At 2% you will get a fully charged light before the batteries are totally full. Neither of these settings affect the AH counting, just the charged light. What will affect the AH readout is the charge efficiency, but only on recharge side. You want to get it to the point that you are around 100% full when the fully charged light comes on. It will vary with depth of discharge, etc, so not totally accurate on recharge. It does reset to 100% as soon as you start to discharge so the discharge will correct, if you got a full charge.

The wiring for the shunt looks OK from what is visible. All the grounds should be one side, and only the battery negative on the other. No other wires should go to the battery negative.

On the solar leak. Are you saying the Trimetric still reads, or does the coach power up, with the battery ground off? If the Trimetric stays on, there is a problem with wiring, most likely. If the solar comes on and powers the van, it probably is wired OK but running without battery reference, because the grounds are tied at the non battery side of the shunt. On some controllers this is bad thing to do, unless the panels are shut off. The problem comes if you are in sun, panels on, and you connect the solar to the coach without any batteries. You can get a big voltage spike that many harm electronics on the van 12v wiring.

Thank you very much booster for this quick and very informative answer!
That forum is unique with all it's very helpful and knowledgeable members!

I understand better now how the Trimetric uses the settings for it's SOC reading.

I don't remember if the Trimetric was on when I disconnected the main ground. I was putting the van in storage so I decided very quickly to disconnect all cables to the batteries before leaving the van for it's 6 month rest. The solar was feeding the system through that "feedback" wire but when I disconnected it from the battery positive the whole system finally shut itself off. I understand that without batteries regulating the voltage the solar panels could harm the system. I will investigate with my upfitter that "feedback" wiring when I get the van back.

I was testing the system because I would like to install a master disconnect switch to the main ground, but obviously that is not possible with the actual wiring.

Why would the charge controller of the solar system be grounded separately and interfere with a normal disconnect switch system on the main ground?

I will ask that question to my upfitter.

 

[booster]

Why would the charge controller of the solar system be grounded separately and interfere with a normal disconnect switch system on the main ground?

When you put the solar into the system, there are several ways they can do it. Normally, and probably the way yours is, would put all the grounds together on non-battery side of the shunt and only the battery on the other side. If they had the solar ground to the battery or the the second side of the shunt, it wouldn't make any difference because solar would still be connected to a plus and minus cable, just no battery in it. That is where the problem comes. This is also assuming the solar is on the battery side of main disconnect in the positive.

We had a good discussion here a while ago as to which side of the main disconnect is the best for the solar because of the possible spike when you got in the sun if the disconnect was open.

I think Avant's was wired to the battery side, with a pullable fuse for when you wanted to kill all the power drains. I did ours that way also, and also added a switch so I can shut off the panels if I want no solar and don't want to pull the fuse.

 

[GeorgeB-ClassB]

...Normally, and probably the way yours is, would put all the grounds together on non-battery side of the shunt and only the battery on the other side.
We had a good discussion here a while ago as to which side of the main disconnect is the best for the solar because of the possible spike when you got in the sun if the disconnect was open. ..

Thanks, found it. Very clear where it's best to wire the solar input for a disconnect switch install.
http://www.classbforum.com/forums/f5/solar-controller-output-destination-3511.html

My solar controller negative is wired with all the other negative cables. No wonder everything was working with the solar when I disconnected only the main battery ground on the right side of the shunt. The positive of the controller was also grouped with all the other loads, so it was a full connection to the van house circuit.

I do have a fuse for the solar input but no master switch yet. I could imagine installing a master disconnect on the negative battery side of the shunt and a main solar controller switch for the positive going to the battery.

Found one diagram for wiring the TriStar solar controller we have. I see the "sense" wiring. I guess it's the battery voltage sense cable. Don't see why the solar array needs to be separately grounded though?

I definitely need the wiring diagram of my van!

Thanks again for your help!

 

[booster]

The panel ground and battery ground are always separate, AFAIK, and both panel wires go directly to the controller, not though the chassis for the ground.

They probably have the "sense" wire with it own ground separate from the main ground system to get a more accurate signal. Many don't do that. Our Blue Sky senses off its battery connections. No doubt the separate would be somewhat better, but if you have enough difference between the two connections, you have other issues to address, I think.

 

[GeorgeB-ClassB]

Thanks again booster. I will investigate the wiring diagram with my up-fitter. To be continued...

 

[markopolo-ClassB]

George - from what you posted I think these are the settings you entered into your Trimetric:

P1: 14.0V ("Charged" setpoint voltage)
P2: 2A ("Charged" setpoint amps)
P3: 190Ah ("Battery Capacity" Amp-hours) - I'm assuming you have two 6V AGM's in series.

Note: P2 is a value in amperes (edit: P2 is a value in amperes (1A to 100A) on the Trimetric TM-2025. Booster pointed out that P2 is a % of the 20 hr Ah rating (0.1 to 10%) for the newer TM-2030 units.)

Your settings require both 14V or more and 2A or less flowing through the circuit to meet the "Charged" criteria.

My settings are:

P1: 14.3V
P2: 1A
P3: 390Ah

My settings require both 14.3V or more and 1A or less flowing through the circuit to meet the "Charged" criteria.

Your settings are good and will help get good life and performance from your battery bank.

My settings require that my battery bank is more fully charged before allowing the "Charged" indicator light to flash.

--------------------------------------

There's tons of great info on the Smart Gauge site: SmartGauge Electronics - SmartGauge battery monitor

The parallel wiring tutorial is a must read:
SmartGauge Electronics - Interconnecting multiple batteries to form one larger bank

I read the site with my particular RV use in mind. You have to consider the low and variable amp draws (loads) typical to small RV use.

From their site:

1. Amp hours counters are incredibly useful pieces of equipment for monitoring power usage and power return (charging).
2. However, they are useless for the purpose of monitoring the state of charge of batteries and a SmartGauge does a far better job, more reliably and is simpler to understand.
3. The simplicity of installation of SmartGauge.

I couldn't agree more on their point #1 especially when it comes to ending amps or return amps that booster mentions often. That is incredibly useful info to have. Critical info from my point of view.

I mostly disagree with the emphatic nature of point #2 because of point #1. To me, it appears that the main cause of their dislike of Ah counters is the inaccuracy buildup over time. You can avoid that error buildup by meeting your charge ending amps setpoint daily (or often) thus resetting your counter. Return daily or often to a state of full charge is easy enough to do in a small RV. It would be much harder to do in a completely off-grid house or a boat though.

If I had to choose one or the other for a small RV my first choice would be an Ah counter. Second choice would be to do as the "engineer who designed the SmartGauge" did - use both type of devices. :thumbup:

 

[GeorgeB-ClassB]

George - from what you posted I think these are the settings you entered into your Trimetric:

P1: 14.0V ("Charged" setpoint voltage)
P2: 2A ("Charged" setpoint amps)
P3: 190Ah ("Battery Capacity" Amp-hours) - I'm assuming you have two 6V AGM's in series.

If I had to choose one or the other for a small RV my first choice would be an Ah counter. Second choice would be to do as the "engineer who designed the SmartGauge" did - use both type of devices. :thumbup:

Thanks for sharing all this info markopolo. Looking forward to pursue the learning and fine tuning of battery management when I get my van back after it's winter vacation!

 

[booster]

Very interesting reading in the Smartguage links. I wish he would have gone into the series/parallel setups in the battery hookup one, though.

The Smartguage to ah counter was definitely biased, as would be expected on their site. There were a couple of things that were of particular interest to me.

He mentioned being able to check the "actual state of charge" at a few points in all the cycling. I would have to assume that would mean a rested voltage check and maybe an internal resistance check. I would think both of those would only be able to be done with no loads or charging happening. In our world, there would be exactly zero times on a 2 month trip that would happen, so that is a bit confusing. I find it hard to understand how the unit could take into account the loads on the system.

His description of Peukert and how it affects things is pretty much opposite what we .discussed in this thread.

Is Peukert misunderstood? - Class B Forums

The big difference is that he claims a rapid discharge actually removes more energy from the battery than a slow one, which is not what the information in the thread indicated, which was no change, and 200ah out was 200ah out, whether fast or slow discharge. Remaining capacity would be the same in both. This is a huge deal in the very real world of how we use our vans. We can be discharging at our normal of less than 1 amp, then to 4.5 amps when the frig comes on, then have 100 amps for a microwave run, and even 150+ amps if the hairdryer is used. That would put most of our use at 100+ hour use with other use at about 3 hour rate. If he is correct, unless we are incredibly lucky to average to 20 hour rate, we should have no clue what we have left, as our monitors don't use Peukert. The Trimetric also does not use Peukert (which we don't have any more) and their explanation why made lots of sense (it is linked in the other thread). If the Smartguage is designed to use Peukert, and if he is not correct on how Peukert works, I would think the guage could not be very accurate except under very consistent drawdown. I could be all off on this, but it really just doesn't make sense to me at this point, as the theories are on both ends of the scale and very either/or in nature.

I seemed to remember being kind of miffed that the Trimetric used a % of capacity to set the fully charged transmission amps, but the readout was in amps, so I went and looked at the TM2030 manual, and it does say it is % of programmed ah capacity.

 

[markopolo-ClassB]

Good catch on the Amp value or % of Ah rating setting for P2. I'll edit my previous post.

It is an amp value for the TM-2025 that I have and is a % of Ah rating for the TM-2030 that GeorgeB has.

Edit: That would make Georges settings:

P1: 14V
P2: 2% (3.8 amps)
P3: 190Ah

I think that would be a bit too easy to reach and the batteries would still have a ways to go to get fully charged. My guess is that his setup easily and routinely surpasses those settings though.

 

[GeorgeB-ClassB]

Good catch on the Amp value or % of Ah rating setting for P2. I'll edit my previous post.

It is an amp value for the TM-2025 that I have and is a % of Ah rating for the TM-2030 that GeorgeB has.

Edit: That would make Georges settings:

P1: 14V
P2: 2% (3.8 amps)
P3: 190Ah

I think that would be a bit too easy to reach and the batteries would still have a ways to go to get fully charged. My guess is that his setup easily and routinely surpasses those settings though.

Thanks, true it's a percentage value.

 

[booster]

Good catch on the Amp value or % of Ah rating setting for P2. I'll edit my previous post.

It is an amp value for the TM-2025 that I have and is a % of Ah rating for the TM-2030 that GeorgeB has.

Edit: That would make Georges settings:

P1: 14V
P2: 2% (3.8 amps)
P3: 190Ah

I think that would be a bit too easy to reach and the batteries would still have a ways to go to get fully charged. My guess is that his setup easily and routinely surpasses those settings though.

I would agree that the 2% is probably too high. What we don't know is if he will ever get there if it is lower, depending on how the charger works. We charge to return amps, so it is easy for us as it takes care of itself. Marko can watch the amps and if he isn't there he can get the PD to run another cycle. We don't know how George's charger is setup at this point, so hard to tell. Usually, if they time to absorption before the batteries are totally full, the monitor won't ever show fully charged , even if they do get full on float, because they are looking for full charge voltage on the return amps at the same time, and once in float it doesn't have the voltage. Even at 2%, if the charger only does a short timed absorption like 2 hours, he may not get to the fully charged setting on a fairly deep discharge. Some of the boat folks know that, and also know they won't ever be able to get full, so they set the return amps to show when they are at 90% or so as an indicator of where they are or to shut off the engine or generator.

Perhaps if George knows the charger make and model we can check it out and see how it will work.

 

[avanti]

I have seen my Trimetric report SOC values > 100%.

 

[markopolo-ClassB]

That's probably because you had not yet reached the "Charged" criteria to automatically reset the SOC counter. It just adds and subtracts amps in and out but amps in are reduced by the setting in P10 - Assumed efficiency factor. P10 is 94% by default on the TM-2025 but can be set between 60% and 100%.

Mine has shown exactly 100% for over a month (plugged in continuously) because it reaches the charged criteria daily (PD charger 15 min boost to 14.4V every 21 hours) and resets the counter.

 

[booster]

I have seen my Trimetric report SOC values > 100%.

Yep, that is actually pretty common and usually do to the charge efficiency setting, which as we have talked about other places, varies with a lot factors including depth of discharge. About all you can do is select a charge efficiency that is good at the midpoint of the range you actually get, and sometimes you will be over 100 when the charge light goes on, and sometimes under. That is why they do the reset to 100% as soon as you start discharging after the charge light is on. It makes the discharge cycle accurate. If you get a full charge every time, you will only be off a little bit on each of the recharges, but both charge and recharge accuracy will normally get worse with every cycle that doesn't get full. Unless of course you get lucky and have the right amount of overs and unders on the recharge side.