What a battery monitor may not tell you

[booster]

There have been a few battery monitor discussions lately, which I think is good as monitors are a very good thing for many of us.


We have had monitors for a long time, so we know exactly what to expect, but for new, or non technical, users that is not the case.


I just ran a test of what could be an easy to have use/charge cycle to watch how the monitor and charging reacted.


I ran the batteries down about 20%, did a short drive to put some charge back in ( could have been longer drive and lower batteries), then parked the van inside so no solar and plugged it in to shore power.


As expected, the shore charger didn't recognize that a charge cycle was needed because of the residual surface charge from the drive holding the voltage at above 12.8v. The monitor still showed about 10% down. The charger went to float and I let it run for 12 hours to test the question of if float will fill the batteries in the top off area of the charging curve. At that point the monitor had just shown 100% charge and net of zero in and out amps, so it would appear charged fully. We know that is not the case , though, as you always need to return 10-20% more amps than used. I then forced the charger into absorption voltage and after a couple of minutes the charging amps settled at about 10 amps, which would indicate we needed about another 3 hours of absorption charging in our system to get to the 2 amp fully charged point.


This answers a couple of the questions that have come up, I think, at least partially.


The first is if the typical charger that stops absorption early will get the batteries to full on float? The answer would be that it will not in an overnight charge situation which is very common for most of us.


The other question is if what you see on the monitor is a accurate and/or correct? I think that the answer to that is all of the above, unfortunately. The zero in/out amps is accurate, but it doesn't mean the batteries are full. The 100% SOC is a guess that the monitor made based on what it thought the charge efficiency would be. On this particular monitor, which does not light a separate "fully charged light", the only way to confirm full would be to force an absorption charge and look at the amps to the batteries.


The no fully charged light situation is on both our monitors, which are both built into the remotes of the shore charger and solar controller. My guess is that they don't have the lights because the shunt based setup for them is an option. For this reason, a standalone monitor with a fully charged indicator is a better choice for indicating really full charging, especially for new users or the non technical ones. On that kind of monitor, if the light isn't lit, the batteries aren't full yet, and that is very easy for everyone to use.

[Davydd]

Are you discussing lead acid batteries only?

[GeorgeRa]

If I recall correctly you have Magnum inverter with ME-ARC50 and ME-BMK shunt. Did you test this combo?

I think any independent shunt battery monitor with interrupted charging by different sources will be prone to errors like for example residual surface charge. An appropriate load to remove surface charge before connecting new charge source could help.

[RossWilliams]

Here is a link to a very thorough discussion of using battery monitors in the context of marine batteries. It does not seem to address lithium, but it details the limitations for lead acid batteries and some that probably apply to both. It both describes amp-counting battery monitors limitations and gives specific instruction on how to better deal with those limitations.

I would read it before you invest in a battery monitor that will not give you the accurate information you are looking for. If you have a battery monitor, I would read it to get a better understanding of how to get the most out of it. I was particularly interested in his description of how improperly using one can lead to shorter battery life.

[markopolo]

Interesting info from Booster. Somewhat similar to what can happen with a basic PD converter when plugging in and there's sufficient surface charge to put it straight into normal 13.6V mode.

Originally Posted by booster

....................... For this reason, a standalone monitor with a fully charged indicator is a better choice for indicating really full charging, especially for new users or the non technical ones. On that kind of monitor, if the light isn't lit, the batteries aren't full yet, and that is very easy for everyone to use...............

An exception would be like how the Samlex EVO inverter charger works. My EVO goes into bulk mode whenever it is reconnected to shore power regardless of voltage. The batteries can be at 13.2V on solar float and the EVO still always goes through the full charge cycle when reconnected to shore power. That charge cycle; bulk, absorption then float can be very quick (like minutes) if the batteries are actually full (to my programmed criteria) because bulk to absorption is triggered by voltage and absorption to float is trigger by current flow. Note: It doesn't have an SOC monitor.

Basically, a "full" battery indication needs to meet both the desired voltage and the desired current flow and you get that with a standalone monitor such as a Trimetric for example as Booster noted above. You can set the charge indicator light to only come on if both 14.4V AND 2A or less current in criteria are met. That's close enough to fully charged for most folks.

In theory it would apply to lithium batteries as well but no one wants to or seems to recommend ever getting them to 100%. The charging goal with lithium is to get them to some point under 100% SOC like maybe 90% or 95%.

[RossWilliams]

Here is a quote from the link I posted above:

"We have not yet seen an Ah counter’s SOC be within 10% of actual tested capacity and most are far worse than a 10% error. Is this the fault of the Ah counter? No, it is the fault of improper use, lack of understanding and poor programming."

He explains in detail the reasons for this and how to avoid adverse consequences from it.

[Davydd]

Originally Posted by markopolo

In theory it would apply to lithium batteries as well but no one wants to or seems to recommend ever getting them to 100%. The charging goal with lithium is to get them to some point under 100% SOC like maybe 90% or 95%.

What does "in theory" mean? Who actually limits lithium batteries to under 90% to 95% SOC? I keep my computer, tablet and cell phone lithium batteries at 100% routinely and will replace them for functional obsoleteness before the batteries deplete. But they aren't RV batteries. I keep my Advanced RV batteries at 100% though ARV does stop charging and letting batteries drop down to 90-91% before charging begins again back to 100%. That was a software change about two years ago for me I suppose to answer "in theory" jitters or for people who think they are going to will their great grandson the RV. Before it was 100% and stop charging and then maintain 100% while plugged in and using your batteries. It is impossible to not go to 100% with their setup. Going on 4.75 years there is imperceptible drop off in performance and that is with an older BMS since improved.

I suppose you can get anal in theory on this and limit your range but why? The in theory advocates also says you should maintain at 65% SOC optimally and that means less than half battery capacity usage. That's idiocy if you maintain 20-65% SOC for minimal gain. I am more concerned about the other end. Who else limits lithium battery usage at 20% SOC by automatically disconnecting, limits charging when batteries go below freezing and maintains battery temperatures optimally always about 41 deg. F?

OK, getting off my soapbox. I'm not an in theory guy, I'm speaking of actual use in having, I believe, the first ARV 800ah build of lithium ion batteries in 2014.

[Bud]

"OK, getting off my soapbox."

Please don't do that too much.

I tend to agree and wonder why I'm limiting by 3hp ebike to 90% charge, switch selectable 80, 90, 100% SOC. It will out live me! Probably true with your B.

Tesla does limit charge and discharge. Then if there is some catastrophic event, they reprogram the usual charge and discharge limits.

[markopolo]

I recall davydd reporting more than once that Advanced RV stopped charging to 100% in favor of charging to 99% ............

Originally Posted by Davydd

..............So, I drive into a campground at a full 99% SOC. I was concerned that I could be at a 90% SOC but learned ARV has programmed only the shore power to charge to 99% and then let electrical usage drop it to 90% before starting charging again.

Originally Posted by Davydd

.....................
Advanced RV is always tweaking their systems. Last year I had an upgrade to my 800ah lithium ion system. Instead of staying fully charged all the time I now float from 90% to 99% SOC. In other words, when I reach 99% charging shuts down until SOC drops to 90%. That is tweaking for longer life. Batteries shut down at 20% SOC but I have set Autogen (auto starting the engine) at 40% SOC but have never dropped to that on our road trips. So I am theoretically keeping my batteries in a middle SOC range. The Autogen SOC is programmable. I don't know what the upper end is off hand but I've programmed it to 90% SOC in my impatience to see if Autogen actually worked. I have once turned Autogen off to see if the batteries actually shut down at 20% SOC. It did.

[markopolo]

My understanding of lithium batteries for RV's is that it is not necessary to charge them to 100%. They are not diminished or affected by not charging to 100%. They might get diminished by charging to 100% but someone else who knows more can answer that.

Lead acid batteries do need to be charged to 100%. They benefit from it.

I would think that with either type of battery, the best type of SOC meter would reset when both voltage and current (amps) set-points are reached. If the SOC meter doesn't do that and instead functions as just a coulomb counter then it seems inevitable that it would become less and less accurate over time.

[Boxster1971]

Originally Posted by RossWilliams

Here is a link to a very thorough discussion of using battery monitors in the context of marine batteries. It does not seem to address lithium, but it details the limitations for lead acid batteries and some that probably apply to both. It both describes amp-counting battery monitors limitations and gives specific instruction on how to better deal with those limitations. . .

Thanks Ross - that link is a great resource. I've read many of Rod “RC” Collins' electrical articles. He is a legend in the Marine Electrical world.

[booster]

Originally Posted by GeorgeRa

If I recall correctly you have Magnum inverter with ME-ARC50 and ME-BMK shunt. Did you test this combo?

I think any independent shunt battery monitor with interrupted charging by different sources will be prone to errors like for example residual surface charge. An appropriate load to remove surface charge before connecting new charge source could help.

Yes, this was run on that system, and I was quite certain it would behave as shown as I caught on to residual charge thing a long time ago. I even spoke to a programmer at Magnum about it, and it is on their list to correct because the unit should not look at voltage to determine if a full charge is needed if running in return amp mode for float transition. As long as it going to float by measuring charge amps, you just run a charge cycle every time the chsrger is plugged in. If the batteries are full, it will see low amps in a couple of minute and go to float so no harm. The Magnum actually will work that way in CC/CV mode with amps charing control set. The Blue Sea charger we had also worked that way, as does our Blue Sky solar controller with amp controlled charging.


So, this an error that shouldn't even happen with the Magnum if the software was fixed.


The reason I did run the test was to show what can happen if you have poor charging control, which includes nearly all shore chargers. It is not the monitor's fault that the charger is incapable of fully charging a battery, and without a full battery calibration point a monitor is very handicapped. The monitor is capable of giving the information that the poor chrarging is happening if you know to look for it when you put in the monitor, and correct the problems.


Many timer chargers look at voltage to determine whether to charge or not and they will wrong probably 98% of the time they don't run a charge. Of course, if they do run a charge cycle on a 90+ full battery they likely will overcharge it, so can't win.


My opinion for all is to not let those that disparage the monitors but ignore the real problems of charging control. They just don't want to know how bad their batteries are being charged.

[booster]

Originally Posted by Boxster1971

Thanks Ross - that link is a great resource. I've read many of Rod “RC” Collins' electrical articles. He is a legend in the Marine Electrical world.

He lost me when he said Peukert should be programmed into the monitor, as IMO all it will do is introduce more errors, not eliminate them. We have discussed that several times and Peukert really doesn't mean that faster discharge uses more energy from the batteries, only that the arbitrary cutoff voltage will happen earlier due to voltage drop.

[booster]

Originally Posted by markopolo

My understanding of lithium batteries for RV's is that it is not necessary to charge them to 100%. They are not diminished or affected by not charging to 100%. They might get diminished by charging to 100% but someone else who knows more can answer that.

Lead acid batteries do need to be charged to 100%. They benefit from it.

I would think that with either type of battery, the best type of SOC meter would reset when both voltage and current (amps) set-points are reached. If the SOC meter doesn't do that and instead functions as just a coulomb counter then it seems inevitable that it would become less and less accurate over time.

Very well put, I totally agree.

[RossWilliams]

the best type of SOC meter would reset when both voltage and current (amps) set-points are reached

The article linked above suggests even with that reset process it is unreliable. He is talking abut a marine environment and some of what he has to say may be unique to that. But for those of us trying to operate off-grid, the marine industry has a very long history of doing exactly that. We need to see if we can learn from their experience.

In theory I think it is true that, unlike lead-acid batteries, lithium batteries don't need to be fully charged. But, in practice, lithium batteries are expensive and most of us want as much juice available as possible. There is no point in paying for unnecessary capacity you can't use.

[RossWilliams]

" He is a legend in the Marine Electrical world."

I can see why. He not only seems to know what he is talking about from long experience, but he explains it clearly.

[booster]

Originally Posted by RossWilliams

The article linked above suggests even with that reset process it is unreliable. He is talking abut a marine environment and some of what he has to say may be unique to that. But for those of us trying to operate off-grid, the marine industry has a very long history of doing exactly that. We need to see if we can learn from their experience.

In theory I think it is true that, unlike lead-acid batteries, lithium batteries don't need to be fully charged. But, in practice, lithium batteries are expensive and most of us want as much juice available as possible. There is no point in paying for unnecessary capacity you can't use.

I am not going to get into a proxy discussion through a third party when they other side is not available. That is not fair to anyone.


This is my comment and applies to battery monitor settings, other opinions may vary.


Programming in a charge efficiency into a battery monitor is an exercise in futility, and no you can not get the CE from a manufacturer of the batteries unless you give them a specific set of conditions as CE changes with DOD, recharge rate, plus all the other things like voltages, temps, etc, so I would totally disagree with the article on that. If you have varying DOD, etc, like we all do, you will have a different charge efficiency every time you charge the batteries. The chrge efficiency is not even the same during a charge cycle, being high t the beginning and very low at the end. If you just tell the monitor it is 100% it will accurately count the AH back in without inaccurately modifying them so you have a better feel for what is going on and will see the actual over 100% replacement AH that are necessary. SOC will behave the same way. The good is that CE does not in any way have an effect on the full charge determination when a monitor shows full battery charge, or the recalibration to 100% at that point, which means the discharge AH out will be correct as will SOC if the capacity is set right.


I have also already described why Peukert does not apply to RV, and likely marine, monitor and battery use. The Trimetric monitor does not use Peukert in their monitors and has an excellent explanation of why that has been linked in the discussions on this forum. It is posted on their website also.

[booster]

This discussion has brought up most of what I thought it would and shows what major points may be being missed or ignored.


A battery monitor is not a remedy for anything, as all it does is watch and tell you what is happening. If you have a system that is capable of doing what it should, store energy in batteries and make it available for use in the best ways, it will watch and show how much energy is available, used, etc quite accurately. Handy when you have large variations in power use in particular.


What monitors are the most often recommended for is tracking down what is happening with systems that aren't doing their job well, or to evaluate and adjust a new system or component in a system. For these purposes in particular, there is not really any better way to determine if your lead acid batteries are fully charged without sitting there for hours with an ammeter and voltmeter, watching them. Anyone can watch a monitor and know if their system is working properly, no tech knowledge needed.



There is one system used by some that does make the monitor an active part of the charging, and that is using the output contacts on a Victron monitor to shut of charging, especially engine charging. A monitor can also be used with any charger that you can force the charging stages to what you want to manually control the float transition based on the monitor lighting the fully charged lamp. Both of these are basically compensations for the poor quality of the available charging equipment, IMO.


A good point that has been brought up is that it is best to have good knowledge of what a monitor is capable of, and of what it can't do. It is not really necessary to know the minutia of all the tech behind it though. Just understand what information is accurate enough and what it means in the real world. Of course this only applies when the system is good enough to do it's job, and that is probably a much larger and more common problem than any monitor problems.

[markopolo]

That's a good summary from Booster.

The best chargers will:
- let you program the bulk current
- let you program the absorption voltage
- let you program the ending amps
- have temperature compensation

Booster has been pointing out for years now that the only practical way to know that your battery is charged is to charge to ending amps. We'll often mention 14.4V & down to 2 amps (or so) or less acceptance as being a good indication that your battery is charged on this forum. Or you'll see 0.5% to 2% of bank capacity used for the amps in set-point. (adjust the criteria as needed for bank capacity, battery brand specifics such as voltage; age, loads such as propane & C0 detectors and bi-directional relays etc. if using internal charger controls as opposed to shunt based monitoring) This means both the voltage and the amps-in set-points must be met. It's not either/or, it's AND.

Think of the above as the foundation of the system. Then adding a Trimetric Battery meter, for example, to that system will give you good information.

The Marine How To article says the same stuff but uses different words. Tail Current from the article is commonly referred to as Ending Amps on this forum and other sites. You have to read for a while to get to what that author describes as "Known Full Reset". The described known full reset is predicated on the charging system being able to fully charge the batteries to BOTH the required voltage AND ending amps.

From the article:

By using a known-full manual re-set you are not guessing when your battery is full, you know when they are full.

If you use a Trimetric for example there's no need to have to do a manual reset as you are able to program the two criteria (voltage and amps-in) to your requirements. If your charging system is capable of triggering the Trimetric "Charged" indicator light daily or at least often then the inaccuracies in the coulomb counting will be minor and not worth worrying about IMO.

If the budget has limits then I'd spend the money on a charger with programmable parameters and temperature compensation before adding a battery monitor. It goes without saying that it would be nice to have both.

[RossWilliams]

Here is a quote from the article after he describes several real life scenarios:

"There are many scenarios we can paint that can cause some of these monitors to reset falsely & prematurely thus creating more counting errors.

For the average boater, simply turn off or program out auto-sync…

For the die hard electrically minded individuals out there it is possible to program auto-sync to still work, but this is a recipe you’ll need to figure out for your own system and your system only. There are far too many variables to give a cookie-cutter recipe for auto-sync especially when we start mixing in solar or wind etc.."

I take that to apply equally to the average rv owner. For someone with lithium batteries having the monitor reset to 100% SOC when it is really at 65% SOC may not be a huge problem. But someone with AGM batteries may end up running their batteries to 0% SOC and permanently damaging them.