Charging tail current and BMV-712 settings?

[T_i_m]

After we upgraded my friends domestic battery shunt to one that would be capable of dealing with everything and then replacing the batteries, I thought I'd leave it all to settle down for a while.

 

Today I was sent a couple of charge trend graphs and think I remember advice about tweaking some of the settings to give the SoC meter a better chance of representing the true SoC (or as near as these things can).

 

This was this morning engine run / charge ...

 

The interesting bit at the beginning is him testing the new air fryer that I think suggests he ran it from the DB >  inverter, rather than the Travel Power (that or the TP couldn't fully cope)?

 

At the point that charge was stopped the SoC was indicated at 100%

 

But my question is more about the battery settings, if anyone could confirm / recommend what we should tweak please? (This was taken later on).

 

 

The TP is 2.5kW and the Combi, 3kW.

 

[blackrose]

My limited understanding is that even if you manage to get a shunt battery monitor to accurately reflect SoC it will soon drift and won't be accurate for very long.

 

I have a shunt battery monitor which I use to view amps in/out of the batteries and monitor tail current so I know when to stop charging. I don't even bother looking at the SoC function. I also have a Smartgauge battery monitor which doesn't have a shunt and uses voltage-based algorithms. That's what I use for battery SoC but only during discharge as that's when it's accurate.

[Tony Brooks]

I would suggest tail current of 1% to 2%, that would halve the rate of drift, but take longer/much longer to register fully charged.

[T_i_m]

4 minutes ago, blackrose said:

My limited understanding is that even if you manage to get a shunt battery monitor to accurately reflect SoC it will soon drift and won't be accurate for very long.

 

I have a shunt battery monitor which I use to view amps in/out of the batteries and monitor tail current so I know when to stop charging. I don't even bother looking at the SoC function. I also have a Smartgauge battery monitor which doesn't have a shunt and uses voltage-based algorithms. That's what I use for battery SoC but only during discharge as that's when it's accurate.

Hi,

 

I think I was hoping that I might be able to at least tweak the settings to get a 'better' SoC, reading, or at least what might be considered the most conservative, especially if currently the Nb owner might be terminating the charge earlier than they should?

 

Ie, He (in particular)  might find it easier to follow the SoC indication rather than monitor the current 'live' from the small BMV-712 display, trying to judge the end of the tail etc?

4 minutes ago, Tony Brooks said:

I would suggest tail current of 1% to 2%, that would halve the rate of drift, but take longer/much longer to register fully charged.

Hi and thanks Tony, I'll recommend that to them. 😉

[dmr]

14 minutes ago, Tony Brooks said:

I would suggest tail current of 1% to 2%, that would halve the rate of drift, but take longer/much longer to register fully charged.

 

I assume the op has lead acids rather than lithiums. My only recent experience is with Trojan lead acids, (though I have a few bad distant memories of cheap lead acids 😀)  1 to 2% is a bit low, its fine when the batteries are new but as they age the tail current creeps up. I estimate mine are about half way through their life and its currently about 3%, though a good equalise might possibly bring this down a bit????

 

The ops friend needs to monitor things for a bit before selecting a value.

[PeterF]

Looking at your charge curve you were getting to 10A and the curve was flattening out, which corresponds to a tail current of 2%, which fits with Tony's upper limit. If you know anothe hour of charging will brimg that lower then use a lower value, mine is 1% on quality AGMs. I would set the tail current to 2%. I would increase the Charged Voltage setting to around 14.2V which means that the charging has to be at high voltage, if there is solar, an overcast day can give low charge current but get above 13.2V and synchronise to 100% before the batteries are full.

 

I would set the lower setting to keep SOC on reset otherwise SOC goes to 100% if power is lost.

 

A shunt is pretty good at keeping tabs on SOC as long as you take the time to tune it to your batteries and have a good set of charged voltage and tail current to reset to 100%.

[nicknorman]

53 minutes ago, T_i_m said:

After we upgraded my friends domestic battery shunt to one that would be capable of dealing with everything and then replacing the batteries, I thought I'd leave it all to settle down for a while.

 

Today I was sent a couple of charge trend graphs and think I remember advice about tweaking some of the settings to give the SoC meter a better chance of representing the true SoC (or as near as these things can).

 

This was this morning engine run / charge ...

 

The interesting bit at the beginning is him testing the new air fryer that I think suggests he ran it from the DB >  inverter, rather than the Travel Power (that or the TP couldn't fully cope)?

 

At the point that charge was stopped the SoC was indicated at 100%

 

But my question is more about the battery settings, if anyone could confirm / recommend what we should tweak please? (This was taken later on).

 

 

The TP is 2.5kW and the Combi, 3kW.

 

 

There is no definition of 100% SoC for lead acid, other than an infinity of time on charge during which the current asymptotically approaches a fixed value (the self-discharge or leakage). So it's a fudge. The Victron default of 4% is quite high. In reality if the batteries are in good condition, fully charged is less than 1% SoC. But that takes a long time to achieve. So as Tony says, 1% to 2% is reasonable. Less is better but means more charging time.

 

What seems wrong in your settings is the "charged voltage". The algorithm for setting the meter to 100% is voltage is greater than charged voltage, current is less than the tail current specified, and this condition is held for the time specified (3 minutes in this case).

 

It is easy to envisage a scenario where, with the charged voltage set to 13.2v, these conditions are met before the battery is charged. First one being you simply decide to stop charging when the battery is at around 90%. The voltage under charge is already 14.4v and when the charger is disconnected with no load, it remains above 13.2v for a while (surface charge effect). If it takes more than 3 minutes to subside below 13.2v then the meter resets to 100%.

 

Or, during charge passing 80% you put a big load on, which drags the charge voltage down to say 13.3v and thus virtually no current flows into the battery. Again, after 3 minutes the "fully charged" algorithm is satified and the meter resets to 100%.

 

I suggest a more sensible value would be 14.2v (depending a bit on the alternator's regulated voltage). This means that only when the batteries are being fed with a good charge voltage AND they are not taking much current, does the meter reset to 100%

 

Why do Victron specify the crappy 13.2v? You'd have to ask them but probably to cater for a battery charger that goes to float a bit prematurely before the current has fully subsided. When the charger goes to float (say 13.3v) the current will fall off and the meter will reset to 100%. Great! (NOT!)

[T_i_m]

3 minutes ago, dmr said:

 

I assume the op has lead acids rather than lithiums. My only recent experience is with Trojan lead acids, (though I have a few bad distant memories of cheap lead acids 😀)  1 to 2% is a bit low, its fine when the batteries are new but as they age the tail current creeps up. I estimate mine are about half way through their life and its currently about 3%, though a good equalise might possibly bring this down a bit????

 

The ops friend needs to monitor things for a bit before selecting a value.

Hi, yes, 4 x  12V, 120Ah sealed FLAs, new a few weeks ago. and a HD DB alternator on the Beta 50 engine, 70' Nb. (Starting alternator / battery separate).

 

He never uses shore power or external genny ... has 3 x 180W solar though an Epever Tracer controller, a 2.5kW TravelPower and a 12/3000/120 Victron Phoenix Combi Plus.

 

He is a(n elderly) live-aboard semi-permanent moorer. 😉 

4 minutes ago, nicknorman said:

 

There is no definition of 100% SoC for lead acid, other than an infinity of time on charge during which the current asymptotically approaches a fixed value (the self-discharge or leakage). So it's a fudge. The Victron default of 4% is quite high. In reality if the batteries are in good condition, fully charged is less than 1% SoC. But that takes a long time to achieve. So as Tony says, 1% to 2% is reasonable. Less is better but means more charging time.

 

What seems wrong in your settings is the "charged voltage". The algorithm for setting the meter to 100% is voltage is greater than charged voltage, current is less than the tail current specified, and this condition is held for the time specified (3 minutes in this case).

 

It is easy to envisage a scenario where, with the charged voltage set to 13.2v, these conditions are met before the battery is charged. First one being you simply decide to stop charging when the battery is at around 90%. The voltage under charge is already 14.4v and when the charger is disconnected with no load, it remains above 13.2v for a while (surface charge effect). If it takes more than 3 minutes to subside below 13.2v then the meter resets to 100%.

 

Or, during charge passing 80% you put a big load on, which drags the charge voltage down to say 13.3v and thus virtually no current flows into the battery. Again, after 3 minutes the "fully charged" algorithm is satified and the meter resets to 100%.

 

I suggest a more sensible value would be 14.2v (depending a bit on the alternator's regulated voltage). This means that only when the batteries are being fed with a good charge voltage AND they are not taking much current, does the meter reset to 100%

 

Why do Victron specify the crappy 13.2v? You'd have to ask them but probably to cater for a battery charger that goes to float a bit prematurely before the current has fully subsided. When the charger goes to float (say 13.3v) the current will fall off and the meter will reset to 100%. Great! (NOT!)

Hi Nick,

 

Thanks for that. I'll have to read it a few times to hit my 'absorbed' threshold. 😉

 

Yes, the 13.2V was the default setting so I'll get them to tweak it up to 14.2 (alternator generally 14.4+).

19 minutes ago, PeterF said:

Looking at your charge curve you were getting to 10A and the curve was flattening out, which corresponds to a tail current of 2%, which fits with Tony's upper limit. If you know anothe hour of charging will brimg that lower then use a lower value, mine is 1% on quality AGMs. I would set the tail current to 2%. I would increase the Charged Voltage setting to around 14.2V which means that the charging has to be at high voltage, if there is solar, an overcast day can give low charge current but get above 13.2V and synchronise to 100% before the batteries are full.

 

I would set the lower setting to keep SOC on reset otherwise SOC goes to 100% if power is lost.

 

A shunt is pretty good at keeping tabs on SOC as long as you take the time to tune it to your batteries and have a good set of charged voltage and tail current to reset to 100%.

Thanks for the reply and info Peter. 😉 This is a cleaner trend graph from last Wednesday:

 

 

[nicknorman]

All those hours spent burning diesel, making noise, wearing the engine, just to put 10A or so into the batteries. Makes me cringe! Lead acid really is crap!

 

... just ignore me, I am having an "I hate lead acid" moment.

[T_i_m]

Just now, nicknorman said:

All those hours spent burning diesel, making noise, wearing the engine, just to put 10A or so into the batteries. Makes me cringe! Lead acid really is crap!

 

... just ignore me, I am having an "I hate lead acid" moment.

No, you are right Nick, I feel the same, but in this particular instance the 'jump' to Li was just too big for all sorts of reasons.

 

It's partly the reason and in spite of wanting to play more with the Venus.OS / RPi /  SmartShunt ... I've just ordered a Black Friday Bluetti AC180, mostly because it has LiFePO₄ cells. 😉

 

It's partly an experiment (what could I power Off-Grid from 2 x 180W panels at home), partly some stand-by power and partly some portable power. I might still set up a parallel system using my existing LA, the Epever solar controller (or a Victron to add more action to my Venus.OS display) and a small Victron inverter.

 

I was going to build my own, possibly using Victron kit but it would add up to much more than what I paid for this AC180.

 

Plus it has a 5 year warranty so ...

[blackrose]

5 hours ago, nicknorman said:

 Lead acid really is crap!

 

... just ignore me, I am having an "I hate lead acid" moment.

 

Yes that's right, but surely we've known that for decades. Don't tell me you've only just realised because you have lithium now? I knew LA batteries were crap when I started living aboard 25 years ago even though I knew nothing about electrics.

 

Lithium is obviously much better but it's an expensive change which many of us find difficult to justify or simply can't afford.

[nicknorman]

3 hours ago, blackrose said:

 

Yes that's right, but surely we've known that for decades. Don't tell me you've only just realised because you have lithium now? I knew LA batteries were crap when I started living aboard 25 years ago even though I knew nothing about electrics.

 

Lithium is obviously much better but it's an expensive change which many of us find difficult to justify or simply can't afford.

No I haven’t know it for decades because it was all there was. It was normal and therefore not crap, just a fact of life on a boat. It was only in 2020 when I started looking into Li that I realised.

[PeterF]

The newer graph shows the tail current getting down to about 6A or 1.5%, it would take a while longer to get to 1%. A tail current setting on the BMV of 2% is reasonable and gives room for ageing.

[Tacet]

9 hours ago, nicknorman said:

All those hours spent burning diesel, making noise, wearing the engine, just to put 10A or so into the batteries. Makes me cringe! Lead acid really is crap!

 

... just ignore me, I am having an "I hate lead acid" moment.

I understand there are engines that propel the boat at the same time as charging batteries.

[nicknorman]

Just now, Tacet said:

I understand there are engines that propel the boat at the same time as charging batteries.

I think you might be correct. Lead Acid wasn’t a big deal for us as generally we cruise  6 hours or so every day we are on the boat. This being why I felt that going Li was not justifiable. I justified it only as a lockdown passtime.

 

But now, I have seen the light (and it’s not the low voltage light on the inverter) and wouldn’t go back to the dark days of cantankerous lead acid and being glued to the Smartgauge. I would liken it to an abusive marriage, you don’t realise how bad it was until you are divorced.

[cuthound]

When BT (who were Europe's largest user of lead acid batteries when I worked there) used wet lead acid batteries they terminated charging when the tail current had remained constant for three consecutive half hourly readings.

[Tony Brooks]

35 minutes ago, cuthound said:

When BT (who were Europe's largest user of lead acid batteries when I worked there) used wet lead acid batteries they terminated charging when the tail current had remained constant for three consecutive half hourly readings.

 That is probably best, but not an option on the Victron thing. It would also probably not be acceptable to boaters who want to minimise engine running.

 

The missing bit from that is @14.2+ volts for those using solar or a battery charger that goes into float when it thinks it will.

Edited November 24 by Tony Brooks

[T_i_m]

3 hours ago, Tacet said:

I understand there are engines that propel the boat at the same time as charging batteries.

😉

 

3 hours ago, nicknorman said:

I think you might be correct. Lead Acid wasn’t a big deal for us as generally we cruise  6 hours or so every day we are on the boat. This being why I felt that going Li was not justifiable. I justified it only as a lockdown passtime.

 

But now, I have seen the light (and it’s not the low voltage light on the inverter) and wouldn’t go back to the dark days of cantankerous lead acid and being glued to the Smartgauge. I would liken it to an abusive marriage, you don’t realise how bad it was until you are divorced.

😉

 

I have had a few of those eureka moments in my life (and not just after divorcing my first wife 😉 ).

 

I think the most notable is my daily carry, a Leatherman PST II (multitool) where I feel naked, or that I have lost my wallet when I don't feel it on my hip. Even my mate on the Nb was impressed how many times I'd done something with the Leatherman before he had found the right tool (and done it efficiently etc).

 

The other was my first Garmin GPS (III+ I think). The number of times we would be out, get thrown off course by some roadworks and I would ask the Mrs to get us on back on track with the map but she couldn't. We were BOTH very happy to spend the 500+ quid to not have to have those moments ever again! 😉

 

I can imagine going to Li over LA would also be one of those moments, if as you say the moment was right.

57 minutes ago, cuthound said:

When BT (who were Europe's largest user of lead acid batteries when I worked there) used wet lead acid batteries they terminated charging when the tail current had remained constant for three consecutive half hourly readings.

I also did 5 years with BT (Factory Technician but only in electronics) fixing 300 BAUD modems the size of a desktop PC and PCM / TDM kit etc.

 

We did go on courses at places like Stone in Staffs and Stud Street in London and we did have visits to exchanges (Strouger ... with big '50V' clear cased battery banks) and Martlesham to see the new 'Fibre Optic' being spliced. 😉

 

 

So when should he have the Travel Power unit switched on?

 

I ask because todays trend chart:

 

 

... showed the toaster load coming off the battery via the Combi and I need to check exactly when / how it would be recommended to have the TP turned on?

 

Like, I'm pretty sure it wouldn't matter if it was on with the engine off (no negative impact on the DB etc) and if on and no real load being drawn, it wouldn't add much of a load to the engine?

 

So if the engine was running then the TP would cover up to ~2.5kW of 240V load (assuming the Combi was set to allow that on the input current knob?) and cover the load it can before drawing more via the Combi inverter?

 

It could also assist with charging the DB but I'm not sure it would, given the DB alternator might have lifted the voltage above what it considered warranted it's involvement?

 

He shouldn't have to adjust the big input selector switch that goes between shore and TP either should he?

 

He seems to turn the TP off and I'm not sure if that's habit from when it wasn't hooked up for a long time (a year I think) or because things work better in most cases with it off or he thinks it should only be used when moving?

 

Ironically, he can't  easily use it when moving as the 1200+ RPM would mean he would empty the cut behind him. ;-(

 

Edited November 24 by T_i_m

[cuthound]

59 minutes ago, T_i_m said:

😉

 

😉😉

😉

 

I also did 5 years with BT (Factory Technician but only in electronics) fixing 300 BAUD modems the size of a desktop PC and PCM / TDM kit etc.

 

We did go on courses at places like Stone in Staffs and Stud Street in London and we did have visits to exchanges (Strouger ... with big '50V' clear cased battery banks) and Martlesham to see the new 'Fibre Optic' being spliced. 😉

 

Yes I have fond memories of Stone and Martlesham.

 

The clear lead acid batteries were the "small" ones.

 

The attached photo shows the larger ones. The largest I came across were 50 volt 15,050Ah batteries, and there were 5 of them in parallel!

[T_i_m]

1 minute ago, cuthound said:

 

Yes I have fond memories of Stone and Martlesham.

 

The clear lead acid batteries were the "small" ones.

 

The attached photo shows the larger ones. The largest I came across were 50 volt 15,050Ah batteries, and there were 5 of them in parallel!

 

Feck, now that IS a battery! 😉

 

So is that each plate connected to it's series partner individually and open to the elements?

 

In true BT style I was sent on a PCM signalling course at Stone after bench repairing them for 6 months. It was interesting when we got to the actual cards I had been repairing and the Instructor getting some bits not quite right. I gently offered a question / correction (that he said he would check on the break). He did and it was nice to see him glancing at me whilst still coving that area and me giving him reassuring nods like I was secret bidding. 😉 

 

It was fully, after leaving BT but still being in the Datacomms game, if we had a visiting engineer for something I didn't cover (like our PABX), I could always tell the Ex BT guys because of how friendly they were (and they would generally accept a cuppa). 😉

 

 

[nicknorman]

6 hours ago, T_i_m said:

 

 

So when should he have the Travel Power unit switched on?

 

I ask because todays trend chart:

... showed the toaster load coming off the battery via the Combi and I need to check exactly when / how it would be recommended to have the TP turned on?

 

Like, I'm pretty sure it wouldn't matter if it was on with the engine off (no negative impact on the DB etc) and if on and no real load being drawn, it wouldn't add much of a load to the engine?

 

So if the engine was running then the TP would cover up to ~2.5kW of 240V load (assuming the Combi was set to allow that on the input current knob?) and cover the load it can before drawing more via the Combi inverter?

 

It could also assist with charging the DB but I'm not sure it would, given the DB alternator might have lifted the voltage above what it considered warranted it's involvement?

 

He shouldn't have to adjust the big input selector switch that goes between shore and TP either should he?

 

He seems to turn the TP off and I'm not sure if that's habit from when it wasn't hooked up for a long time (a year I think) or because things work better in most cases with it off or he thinks it should only be used when moving?

 

Ironically, he can't  easily use it when moving as the 1200+ RPM would mean he would empty the cut behind him. ;-(

 

The TP alternator is spinning anyway so if the TP is on with zero load isn’t going to add any significant engine load. Our TP only comes on when the ignition is on so presumably the 12v supply for the “magic box” comes from the engine electrics. In fact we have the TP remote panel anyway, so the switch on the actual TP box is always on. When the engine is started the TP comes on in standby mode, I have to press a button on the remote panel to get it to actually start operating.

 

Anyway, the point is that there is no real downside to having the TP on. As you say, it would cover loads like the toaster rather than sapping alternator/battery power.

 

If it were mine I would set up the Combi to supplement the alternator (if it isn’t already). Your graphs illustrate how poor a standard alternator regulator is. Ideally the current would be at max (bulk mode) until the voltage reached 14.4v and then the voltage would stay at 14.4v as the current gradually subsided (absorption mode). But you can see a massive overlap where the alternator is neither in pure bulk nor pure absorption - there is a lengthy period where the voltage gradually rises as the current subsides. By the time the voltage has reached the regulated value of 14.4v, the current is about half maximum. This is the crappy soft regulation of a single-transistor analogue regulator with only error feedback (the P in PID - the ID is absent). Using the Combi fed from the TP would likely increase the voltage and hence the current during the first half of the charging until the voltage reaches 14.4v

Edited November 24 by nicknorman

[PeterF]

Based on the chart, once the charge current falls to 50A the voltage is 14.4V, so running the TP for battery charging will be beneficial until the current falls to around 50A.

[nicknorman]

3 minutes ago, PeterF said:

Based on the chart, once the charge current falls to 50A the voltage is 14.4V, so running the TP for battery charging will be beneficial until the current falls to around 50A.

Yes. But leaving it on beyond that, whilst not having benefit, also doesn't really have a downside so there is no need to worry about leaving it on.

[T_i_m]

25 minutes ago, nicknorman said:

If it were mine I would set up the Combi to supplement the alternator (if it isn’t already).

Thanks for the feedback Nick (all noted).

 

It has been suggested that the TP was on for this mornings charge so would that mean that the Combi isn't configured to supplement the TP (assuming the Combi input current knob was set to at least 2.5A) or something is broken etc?

5 minutes ago, PeterF said:

Based on the chart, once the charge current falls to 50A the voltage is 14.4V, so running the TP for battery charging will be beneficial until the current falls to around 50A.

Given that looks to be about an hour Peter, would be worth doing.

 

So to test the TP we could turn off the Combi (or put it on charge only) and then see if we can run the toaster (I believe we did that, and ran a full washing machine load, assuming it hasn't failed since) on the TP alone?

 

Then it might be a matter of checking the Combi settings as Nick suggests and see if it's set up correctly to support the TP if required (and / or charge if it's not needed)?

[nicknorman]

12 minutes ago, T_i_m said:

Thanks for the feedback Nick (all noted).

 

It has been suggested that the TP was on for this mornings charge so would that mean that the Combi isn't configured to supplement the TP (assuming the Combi input current knob was set to at least 2.5A) or something is broken etc?

If the TP is on and working and selected by the rotary switch to feed the Combi, the Combi should be in Charger mode and not in Inverter mode, provided the mains power demand isn't greater than the rotary knob setting. Has this been checked - ie that the TP is actually sending power to the Combi?