Victron battery monitor weirdness.

[s/v Jedi]

Originally Posted by GILow

A picture is worth a thousand words. Well, a couple of dozen in this case.

I think you can safely go to 14.6V as charged voltage in the BMV configuration. Just check that the BMV resets to 100% regularly.

The issue is this: at 14.2V it means the BMV will reset to 100% at 14.2V even in case you stop charging before finishing or even reaching absorption phase. When you increase the charged voltage setting to 14.6, it becomes more likely to get fully charged. The risk of going too high is that the BMV doesn’t reset in which case it becomes less accurate.

[rslifkin]

Originally Posted by GILow

I’ve not had much to do with AGMs yet but it is my understanding that they are less “resistive” to charge than FLA?

Also, are you charging with solar or some other source? As smac999 notes, solar fluctuations can really muddy the picture.

Both solar and shore charging depending on the situation. The fluctuations are why I set the time longer (it's either 10 or 15 minutes, can't remember). If solar output drops due to clouds, unless the batteries are pretty full then a charge rate at or below 0.5% of capacity will lead to voltage falling below 13.3 before the 10 minute mark, so it doesn't sync to 100% prematurely. FLAs should behave similarly enough.

[GILow]

Originally Posted by rslifkin

Both solar and shore charging depending on the situation. The fluctuations are why I set the time longer (it's either 10 or 15 minutes, can't remember). If solar output drops due to clouds, unless the batteries are pretty full then a charge rate at or below 0.5% of capacity will lead to voltage falling below 13.3 before the 10 minute mark, so it doesn't sync to 100% prematurely. FLAs should behave similarly enough.

Interesting observations, thank you.

I will keep an eye on the history function and see just how the FLA Trojans behave in similar situations. My old Votronic didn’t give me this sort of data so I’m still discovering what was going on in the background. I guess I’ve had such a stress free run with the Trojan setup that I’ve never felt the need to peek under the hood. But I’m curious now.

[GILow]

Originally Posted by s/v Jedi

I think you can safely go to 14.6V as charged voltage in the BMV configuration. Just check that the BMV resets to 100% regularly.

The issue is this: at 14.2V it means the BMV will reset to 100% at 14.2V even in case you stop charging before finishing or even reaching absorption phase. When you increase the charged voltage setting to 14.6, it becomes more likely to get fully charged. The risk of going too high is that the BMV doesn’t reset in which case it becomes less accurate.

I suppose you are right. The Coulomb counter seems to get it right at about the same time as the bank reached 14.2, but no harm in going higher.

The new boat has a DIY LifePO4 bank so I’m going to have to start over again with these setting when I get to setting it up.

[Stu Jackson]

Originally Posted by GILow

Thank you, you’ve made me feel good because an hour ago I slapped myself and told myself to just go with my gut feel and use 14.2.

And you’ve articulated something that was bothering me about the Victron algorithm, which was that it seemed more appropriate to mains chargers than solar.

The real issue is declaring premature condition of full. This has been a known issue for over a decade. 14.2 is a good start.


These explain WHY the defaults need to be changed because of premature declaration of full.




For everyone installing a battery monitor: The "Gotcha Algorithm" thread, a "MUST READ"

http://c34.org/bbs/index.php/topic,4922.0.html

DEFAULTS are factory settings that are made to be modified to suit your setup.

Also read this one:

Programming a Battery Monitor (by Maine Sail)

https://marinehowto.com/programming-a-battery-monitor/

[GILow]

Originally Posted by Stu Jackson

The real issue is declaring premature condition of full. This has been a known issue for over a decade. 14.2 is a good start.


These explain WHY the defaults need to be changed because of premature declaration of full.




For everyone installing a battery monitor: The "Gotcha Algorithm" thread, a "MUST READ"

http://c34.org/bbs/index.php/topic,4922.0.html

DEFAULTS are factory settings that are made to be modified to suit your setup.

Also read this one:

Programming a Battery Monitor (by Maine Sail)

https://marinehowto.com/programming-a-battery-monitor/

I dunno if I should be taking electrical advice from a guy who wires his subwoofer into the bilge pump circuit.

[GILow]

Originally Posted by Stu Jackson

The real issue is declaring premature condition of full. This has been a known issue for over a decade. 14.2 is a good start.


These explain WHY the defaults need to be changed because of premature declaration of full.




For everyone installing a battery monitor: The "Gotcha Algorithm" thread, a "MUST READ"

http://c34.org/bbs/index.php/topic,4922.0.html

DEFAULTS are factory settings that are made to be modified to suit your setup.

Also read this one:

Programming a Battery Monitor (by Maine Sail)

https://marinehowto.com/programming-a-battery-monitor/

Well… the first link is kinda stone-age now and the second link is broken.

It was the thought that counted tho.

[SailingHarmonie]

Originally Posted by rslifkin

With my AGMs, I have the Victron "full charge" settings set for 13.3v and 0.5% tail current for at least 10 minutes. It never syncs to 100% prematurely, only after completing absorb, dropping to float and then being in float for a bit. Basically, 0.5% won't happen for any sustained period at or above 13.3 volts unless the batteries are pretty much full.

But this is not right. 0.5%C is the right number for calling the batteries rigorously 100% full, but that is at absorption voltage. If it resets after dropping to float, you don’t really know what the actual SOC is. Just because you reach what you picked for voltage current at full does not mean the batteries actually are full!

In any event, for the purposes of the battery monitor, it is not required to get to a rigorous 100% SOC for the synch function to get usefully accurate SOC data.

We set our charged voltage to 0.3 volts under the absorbsion voltage to account for those hot days when the battery temp drives the voltage down a bit, and the 4%C for 3minutes works well on the current side while the solar system is topping up the batteries on a sunny day.

[smac999]

Originally Posted by rslifkin

With my AGMs, I have the Victron "full charge" settings set for 13.3v and 0.5% tail current for at least 10 minutes. It never syncs to 100% prematurely, only after completing absorb, dropping to float and then being in float for a bit. Basically, 0.5% won't happen for any sustained period at or above 13.3 volts unless the batteries are pretty much full.

You will probably never hit 0.5% tail current even if you leave the batteries charging for weeks plugged into ac. They will constantly take amps even if full. So yours probably never does an auto resync. Most batteries are considered full at ~2%. I guess maybe at float. But my 600ah bank will keep taking ~8a all day long.

[smac999]

Originally Posted by s/v Jedi

I think you can safely go to 14.6V as charged voltage in the BMV configuration. Just check that the BMV resets to 100% regularly.

The issue is this: at 14.2V it means the BMV will reset to 100% at 14.2V even in case you stop charging before finishing or even reaching absorption phase. When you increase the charged voltage setting to 14.6, it becomes more likely to get fully charged. The risk of going too high is that the BMV doesn’t reset in which case it becomes less accurate.

The problem is if it’s 40c mid summer the absorb charging value drops to ~14.2v. And never gets near 14,6v

My 0.5v below absorb recommendation leaves buffer for volt temp comp. with lithuim batteries I would go 0.2v under absorb since absorb value will never change. Which more or less means 14-14.2 works for all batteries types

[s/v Jedi]

Originally Posted by smac999

The problem is if it’s 40c mid summer the absorb charging value drops to ~14.2v. And never gets near 14,6v

My 0.5v below absorb recommendation leaves buffer for volt temp comp. with lithuim batteries I would go 0.2v under absorb since absorb value will never change. Which more or less means 14-14.2 works for all batteries types

Yes, I messed up because I wasn’t considering the chemistry and am all about LFP now. I agree, the temperature compensation must be taken into account for lead acid chemistry

[rslifkin]

Originally Posted by SailingHarmonie

But this is not right. 0.5%C is the right number for calling the batteries rigorously 100% full, but that is at absorption voltage. If it resets after dropping to float, you don’t really know what the actual SOC is. Just because you reach what you picked for voltage current at full does not mean the batteries actually are full!

In any event, for the purposes of the battery monitor, it is not required to get to a rigorous 100% SOC for the synch function to get usefully accurate SOC data.

We set our charged voltage to 0.3 volts under the absorbsion voltage to account for those hot days when the battery temp drives the voltage down a bit, and the 4%C for 3minutes works well on the current side while the solar system is topping up the batteries on a sunny day.

Originally Posted by smac999

You will probably never hit 0.5% tail current even if you leave the batteries charging for weeks plugged into ac. They will constantly take amps even if full. So yours probably never does an auto resync. Most batteries are considered full at ~2%. I guess maybe at float. But my 600ah bank will keep taking ~8a all day long.

My setup definitely auto-syncs, I've monitored it to make sure. Plenty of times I've seen the log on VRM show that it went from something like 99.6% to 100% instead of slowly creeping the last bit to full. My AGM bank (currently 415ah) will get down to under 0.5% (2 amps) in float fairly easily when full, and if I leave the boat on shore power for a day or 2, it'll get down to about 0.1 amps in float.

For the solar controllers, I drop from absorb to float at 4 amps of charge acceptance (1%). The battery manufacturer (Fullriver) indicates that full charge is absorb until 1.2 - 2% of capacity, then 8 hours in float before considering them full. Because they'll never get 8 hours of float from solar, I push them a little further in absorb.

Based on the behavior I've observed with these batteries and when my charging sources drop from absorb to float, I've never encountered a situation where I get a premature BMV sync. Basically the batteries won't hold 13.3+ volts with 2 amps or less going in for 10+ minutes unless they're actually very close to fully charged. Doing it this way leads to it syncing later than if I were targeting to have it sync during absorb, so I've found it to have a very low risk of a premature sync to full.

Another concern for me is that my alternator charging is currently done via a pair of ACRs with internally regulated alternators, so they only put out about 14 volts. If I had the sync voltage set too high, I'd never get a sync from an extended motoring session (powerboat) unless solar pushed the voltage up all the way to my normal absorb voltage (14.7v).

[GILow]

So, it sounds like I should stick with 14.2?

Talk of temperatures prompted me to see how the batteries had fared in a recent heat wave. They got warm. (We slept out on the foredeck for that week.)

But oddly the voltages seemed to have increased over those few days. I thought they were supposed to decrease.

[rslifkin]

Make sure your temperature compensation setting is entered as negative. Otherwise it might be compensating the wrong direction and increasing (rather than decreasing) voltage at higher temperatures.

[Stu Jackson]

Originally Posted by GILow

Well… the first link is kinda stone-age now and the second link is broken.

It was the thought that counted tho.

you're welcome


stone age? No, really. It IS all about the logic behind the algorithm, and premature call of full. Which remains valid. All of it.



Sorry Maine Sail's link died, but you could try his website to see if anything like it is still there. He's posted so many great things, but the organization of his website is not great or easy to click through at all.