diy solar

diy solar

Need help debugging a charging issue using a Multiplus and shore power

rmaddy

Full-time Solar-powered Trailer Life
Joined
Nov 16, 2019
Messages
3,737
Location
USA
Today, for the first time, I had my complete electrical system all connected in my cargo trailer so I made my first attempt to charge my LiFePO4 batteries using the Multiplus inverter/charger while connected to shore power. After roughly 90 minutes of charging the battery monitor has detected excessive difference in voltage between the batteries and charging has been stopped.

Details:

Two SOK 12V 206Ah batteries in series for a 24V system.
Victron Multiplus 24/2000/50 inverter/charger.
Victron BMV-712 battery monitor and shunt. I have this setup to do midpoint monitoring to ensure the two batteries stay close to the same voltage.

Lots of other parts but everything was turned off during most of the charging.

Before I started charging, both batteries were reading the same voltage at roughly 13.2V and the battery monitor showed roughly 26.4V. The shore power was connected, through an extension cord and adapter, to a normal 15A, 110V outlet at my house. Once the inverter/charger was connected it started showing that it was bulk charging the batteries. Through the VictronConnect app on my phone I could see the batteries were getting charged at roughly 40A. The charging battery voltage was initially showing 27.53V and slowly going up. The last value I noticed, which was about 10 or 15 minutes before the alarm, the charging voltage was up to about 27.77V. I had left the trailer during that 10-15 minutes and it is when I came back that I heard the alarm. The inverter/charger has stopped charging and the battery monitor was reporting an excessive voltage difference between the two batteries of just under 2%. I turned the inverter switch off and stopped the alarm. Using a multimeter I measured the voltage of the two batteries at their terminals. One battery was showing about 13.4V and the other was roughly 13.7V. Again, both had the same voltage just before I started charging. The battery with the lower voltage is the one connected to the shunt.

What am I doing wrong here and what needs to be changed to help ensure the two batteries charge equally?

My electrical schematic is attached (hopefully it's clear enough as attached).

During the time I was charging both of the two big red switches were on. The battery protect was turned off via VictronConnect so no 12V or 24V loads were being used. All of the AC breakers were off so no AC loads were being used. The breaker to the charge controller was open so effectively that was out of the picture and my solar panels are not yet installed. So the only components in play should have been the batteries, shunt, inverter/charger, battery monitor, and the smart dongle.

Schematic2.png
 
You need to charge the two batteries to full in parallel to 14.4-14.6V.

(+) charger lead should be on (+) of battery 1
(-) charger lead should be on (-) of battery 2

Once fully charged, the batteries will be top balanced and should have a much closer voltage throughout the majority of their operating range.

Additionally, these 12V balancers will work to keep them at the same voltage by transferring charge from high to low:


I use 8 of these on my Trojans to keep them balanced.
 
You need to charge the two batteries to full in parallel to 14.4-14.6V.

(+) charger lead should be on (+) of battery 1
(-) charger lead should be on (-) of battery 2

What kind of charger is best for this process? I'd like to avoid spending $$$ on some fancy charger that will get used once or twice.

And to be clear, I should disconnect my batteries from my system, connect the two negative leads together and connect the two positive leads together (with the same length cables I presume) and then connect the charger leads to the two batteries as you described. Then once the charge is complete, connecting my batteries back into my system should resolve my issue?

Thanks.
 
What kind of charger is best for this process? I'd like to avoid spending $200 on some fancy charger that will get used once or twice.

And to be clear, I should disconnect my batteries from my system, connect the two negative leads together and connect the two positive leads together (with the same length cables I presume) and then connect the charger leads to the two batteries as you described. Then once the charge is complete, connecting my batteries back into my system should resolve my issue?

Thanks.

Any 12V battery charger that will charge them to 14.4-14.6V. Automotive 12V chargers might work.

Yes. Completely remove batteries from system. Break the series connection and wire them in parallel. Charge them. Once fully charged, restore original configuration.
 
I have a charger on order and should have it in a few days. I'll post my results after I get the batteries fully charged.
 
Battleborn recommends that you do this once per year for their batteries when wired in series. IMHO, the addition of $50 of those balancers (one per battery) should make this unnecessary.
 
Update. I got a new charger (Victron 12V 17A). I put it in Li-Ion mode. I removed my batteries and connected them in parallel and connected the charger leads, negative on one battery and positive on the other battery. It did a bulk charge with 14.2V for about an hour and then it switched to float for about 2 hours before it switched to storage mode which meant the batteries were charged. I took it all apart and reinstalled the batteries into my system. I reset the BMV-712 to tell it the batteries were at 100% SOC. For the next several hours I ran two roof fans and nothing else. Initially the battery monitor with midpoint monitoring showed the two batteries differed by 0.4% (0.06V). As the day went on that difference dropped to zero. When I shut it all down for the evening it showed 99.1% SOC. I never turned on the inverter today.

After I have a chance to drain the batteries a bit over the next couple days i'll see what happens when the inverter tries to charge them while connected to shore power.
 
After balancing my batteries last week and putting them back in my trailer I have been slowing draining the batteries with light use. Over the last 8 days I have not connected to shore power. While working on the trailer I've been using the roof fans, charging my phone, and some other minor electrical uses. As of this morning the battery monitor showed the batteries were at 53% so I decided it was time to connect to shore power and let the inverter/charger recharge the batteries.

Just before connecting to shore power, the battery monitor was showing 26.19V and 0V deviation between the two batteries.

I plugged the shore power to my house (15A outlet) and the inverter showed it was bulk charging at roughly 43A. In just a few minutes the battery monitor showed about 0.02V deviation between the batteries but it stayed at that value during the entire bulk charging period which lasted just under 3 hours. So far so good. Moments after the inverter/charger switched to absorption mode the battery monitor showed 100% SOC, 28.8V, a charge current of about 20A, and the deviation had risen to about 0.06V.

Within 20 minutes of being in absorption mode the deviation was roughly 0.46V. Like in my first post above, the battery with the lower voltage is the one connected to the shunt. At this point I turned off the inverter to stop the charging and I took a break. I checked it about two hours later and the battery monitor showed the voltage has dropped back to 26.63V (there was about a 20W draw during those two hours) and the deviation had dropped back to 0.

So while the bulk charging seemed to go fine, the voltage deviation between the batteries kicked in during absorption charging. And a while after terminating all charging, the deviation returned to 0.

I'm wondering if the problem is how I have some things hooked up to the batteries. As shown the diagram in the first post, the left battery negative is connected to the shunt and the left battery positive is connected to the right battery negative (obviously since the two batteries are in series). The right battery positive goes to the main positive busbar with an ANL fuse in between. The right battery negative has two other connections - 1) The power lead wire connected to the B2 terminal on the shunt for doing the mid-point voltage monitoring, and 2) The supplied temperature sensor wired into the inverter.

I believe those are wired correctly but the right battery is the one that gets the higher voltage when the inverter/charger switches to absorption charging mode. Could it be that it gets extra juice due to those extra wires?

Or what else could make two batteries in series charge equally during bulk charging, then charge unevenly during absorption charging, and then return to equal charge a bit after charging is terminated?

Maybe I should disconnect those two extra wires from the right battery and let the charger get into absorption charging mode and then manually check the voltage of the two batteries for a while with a voltmeter.
 
After balancing my batteries last week and putting them back in my trailer I have been slowing draining the batteries with light use. Over the last 8 days I have not connected to shore power. While working on the trailer I've been using the roof fans, charging my phone, and some other minor electrical uses. As of this morning the battery monitor showed the batteries were at 53% so I decided it was time to connect to shore power and let the inverter/charger recharge the batteries.

Just before connecting to shore power, the battery monitor was showing 26.19V and 0V deviation between the two batteries.

I plugged the shore power to my house (15A outlet) and the inverter showed it was bulk charging at roughly 43A. In just a few minutes the battery monitor showed about 0.02V deviation between the batteries but it stayed at that value during the entire bulk charging period which lasted just under 3 hours. So far so good. Moments after the inverter/charger switched to absorption mode the battery monitor showed 100% SOC, 28.8V, a charge current of about 20A, and the deviation had risen to about 0.06V.

Within 20 minutes of being in absorption mode the deviation was roughly 0.46V. Like in my first post above, the battery with the lower voltage is the one connected to the shunt. At this point I turned off the inverter to stop the charging and I took a break. I checked it about two hours later and the battery monitor showed the voltage has dropped back to 26.63V (there was about a 20W draw during those two hours) and the deviation had dropped back to 0.

Confirm: @28.8V, 20A current, .06V deviation between batteries A & B?

After another 20 minutes @ 28.8V, deviation increased to 0.46V? How can this be? Did one battery drop and the other raise?

So while the bulk charging seemed to go fine, the voltage deviation between the batteries kicked in during absorption charging. And a while after terminating all charging, the deviation returned to 0.

Once you hit absorption, you're in the extreme upper leg of the charge curve where voltages change very rapidly. @ 20A (0.1C), you're already at an extremely high state of charge - likely in the 95%+ range (100% according to your BMV), so voltage deviations may manifest very quickly even though the batteries are at very similar states of charge.

Voltages settle after charge or discharge current is removed, so it's not surprising the voltage deviation goes away once charge is removed and

I'm wondering if the problem is how I have some things hooked up to the batteries. As shown the diagram in the first post, the left battery negative is connected to the shunt and the left battery positive is connected to the right battery negative (obviously since the two batteries are in series). The right battery positive goes to the main positive busbar with an ANL fuse in between. The right battery negative has two other connections - 1) The power lead wire connected to the B2 terminal on the shunt for doing the mid-point voltage monitoring, and 2) The supplied temperature sensor wired into the inverter.

I believe those are wired correctly but the right battery is the one that gets the higher voltage when the inverter/charger switches to absorption charging mode.

1618204969938.png

Could it be that it gets extra juice due to those extra wires?

No. All components within a series circuit must experience the same current.

Or what else could make two batteries in series charge equally during bulk charging, then charge unevenly during absorption charging, and then return to equal charge a bit after charging is terminated?

They are always charging equally as they are both receiving exactly the same amount of current. One is just at a higher state of charge than the other and its voltage increases faster as a result.

Maybe I should disconnect those two extra wires from the right battery and let the charger get into absorption charging mode and then manually check the voltage of the two batteries for a while with a voltmeter.

Good to confirm values with a separate meter, but there's nothing to be gained by disconnecting the BMV as charging doesn't work as you imagine.

Hypothesis: One of your batteries has a slightly higher self-discharge rate than the other. After 8 days of use, it lost a very small amount of its stored charge putting it 1-2% lower than the other battery. Upon recharge and attaining near 100% SoC, the slight difference in SoC between the two batteries manifested itself as a voltage deviation.
 
Confirm: @28.8V, 20A current, .06V deviation between batteries A & B?

After another 20 minutes @ 28.8V, deviation increased to 0.46V? How can this be? Did one battery drop and the other raise?

At this point I'm not sure. I was tracking so many numbers I'm forgetting the exact sequence now. I'll need to take better notes next time to be sure. I know the deviation was increasing quickly once absorption charging kicked in.

Once you hit absorption, you're in the extreme upper leg of the charge curve where voltages change very rapidly. @ 20A (0.1C), you're already at an extremely high state of charge - likely in the 95%+ range (100% according to your BMV), so voltage deviations may manifest very quickly even though the batteries are at very similar states of charge.

That charge current was 20A at the very initial transition from bulk to absorption. But it tapered off fairly quickly down to a much smaller current.

Voltages settle after charge or discharge current is removed, so it's not surprising the voltage deviation goes away once charge is removed and



View attachment 44767



No. All components within a series circuit must experience the same current.

This diagram is from the BMV-700/702 Quick install manual (and how my setup is wired). The Quick install specific to the 712 has a similar diagram. One difference (but it shouldn't matter, right?) is that the 712 Quick install guide shows the B2 wire going to the positive terminal of the battery connected to the shunt.

They are always charging equally as they are both receiving exactly the same amount of current. One is just at a higher state of charge than the other and its voltage increases faster as a result.

I'm still confused why there's virtually no deviation for hours during high current bulk charging and then suddenly a big deviation kicks in as the charge current drops a lot during absorption.

Good to confirm values with a separate meter, but there's nothing to be gained by disconnecting the BMV as charging doesn't work as you imagine.

I wasn't planning to disconnect the BMV. I was planning to remove the inverter's temperature sensor (if that actually needed for LiFePO4?) and the mid-point monitoring lead.

Hypothesis: One of your batteries has a slightly higher self-discharge rate than the other. After 8 days of use, it lost a very small amount of its stored charge putting it 1-2% lower than the other battery. Upon recharge and attaining near 100% SoC, the slight difference in SoC between the two batteries manifested itself as a voltage deviation.

If this is true then any thoughts on a solution? I've only made two attempts to charge my batteries from the inverter so far and both resulted in things going bad during absorption charging. I assume this isn't normal. Are there settings I should change? Should I retry the exercise of fully charging the batteries in parallel? Or ...?

Thanks.
 
Does lithium need the absorption phase or is it counter productive to battery life? Bulk charge to 3.45-3.5 and end charging cycle? Still trying to learn the differences.
 
At this point I'm not sure. I was tracking so many numbers I'm forgetting the exact sequence now. I'll need to take better notes next time to be sure. I know the deviation was increasing quickly once absorption charging kicked in.

I question this because 0.46V deviation between two 12V batteries already at 14.37/14.43V is almost catastrophic and should likely see the battery's BMS triggering high voltage shut off. I don't see how it's possible that only one battery could shoot to 14.9V while the other remains at 14.4V, nor is it plausible that the low one would drop.

That charge current was 20A at the very initial transition from bulk to absorption. But it tapered off fairly quickly down to a much smaller current.

The above confirms that your batteries are at a very high state of charge once absorption was hit. The BMV registered 100% meaning you had input the correct amount, or you triggered its 100% criteria.

This diagram is from the BMV-700/702 Quick install manual (and how my setup is wired). The Quick install specific to the 712 has a similar diagram. One difference (but it shouldn't matter, right?) is that the 712 Quick install guide shows the B2 wire going to the positive terminal of the battery connected to the shunt.

Shouldn't matter.

I'm still confused why there's virtually no deviation for hours during high current bulk charging and then suddenly a big deviation kicks in as the charge current drops a lot during absorption.

Because that's how LFP works. The voltage curve is extremely flat with comparatively violent changes at full and empty:

1618236239704.png
The above chart shows an empty to full charge based on LFP's upper limit of 4.2V (this is rarely used because there is very little to be gained in capacity by taking it to 4.2, and it shortens their life). I have sketched in two green lines. The upper is the 3.65V line corresponding to 14.6V. As you can see, the red line changes very little until the very end.

The lower green line represents your current, 20A, or 0.1C. This is a low current for that level of charge, and many cells indicate that batteries/cells are full once 0.05C is reached, i.e., 10A in your case.

In other words, your charger should be configured to stop charging at 28.8V and 10A. If you were still charging at a lower rate, it's possible you were over-charging the battery.

I wasn't planning to disconnect the BMV. I was planning to remove the inverter's temperature sensor (if that actually needed for LiFePO4?) and the mid-point monitoring lead.

No value in that. Those are sense and BMV power wires. They do not pass meaningful current, and the temperature sensor doesn't factor into it.

If this is true then any thoughts on a solution? I've only made two attempts to charge my batteries from the inverter so far and both resulted in things going bad during absorption charging. I assume this isn't normal. Are there settings I should change? Should I retry the exercise of fully charging the batteries in parallel? Or ...?

First, you need to change your perspective. These two attempts are completely different. 13.4 and 13.7 are HUGE disparities encountered well before full charge. The second attempt encountered a 0.06V deviation @ absorption and 20A of charge - when the batteries were 1) indicated at 100% SoC by the BMV AND 2) just by voltage/current criteria alone at an EXTREMELY high state of charge. This was AFTER allowing the batteries to go uncharged and used for 8 days.

Your problem has gone from an axe wound to a mild hangnail.

IMHO, if my hypothesis is right, this is something you will experience on a regular basis because the batteries themselves perform slightly differently, and it will be worse the longer you go between charges. This is likely the result of batteries that perform within spec, but do not perform well with each other in series.

Minimize the issue:
  1. Set proper charge termination criteria on your Multiplus. Limit absorption time to 10 minutes. If it allows tail current, specify it at 10A
  2. Set BMV charged current to 28.6V, 7% tail current, charged detection time to 1 min.
  3. Purchase and install the balancers I linked in post #2.
 
IMHO, if my hypothesis is right, this is something you will experience on a regular basis because the batteries themselves perform slightly differently, and it will be worse the longer you go between charges. This is likely the result of batteries that perform within spec, but do not perform well with each other in series.

Minimize the issue:
  1. Set proper charge termination criteria on your Multiplus. Limit absorption time to 10 minutes. If it allows tail current, specify it at 10A
  2. Set BMV charged current to 28.6V, 7% tail current, charged detection time to 1 min.
  3. Purchase and install the balancers I linked in post #2.
Thanks for all of the great info. I got the batteries down to 80% from 100% today and I should be able to get it down below 50% tomorrow so then I'll try charging again with these changes in place and see how it goes. I'll take better notes on the results.

BTW - I just found this in the manual for the battery charger I used last week to fully charge my two batteries while connected in parallel:

The charger is designed for lead-acid and Li-ion batteries 14-225Ah. Do not use for any other purpose.

With my two 206Ah batteries in parallel that is of course 412Ah. Could this mean that my attempt to fully charge the two batteries to the same 100% SOC using that charger didn't work as it could have?
 
Lead acid should be charged with a range of currents. That charger provides too low a current to safely charge > 225Ah lead acid batteries. Shouldn't matter with yours.

Only charging to 14.2V was likely a little short of 100%, but the long float periods should have topped them off.
 
I have detailed results that are rather interesting I think. Between yesterday and today I ran the batteries from 100% SOC to 53%.

The inverter is set to 28.6V for absorption charging and 27.2V for float. The absorption time is set to 1 hour. It can't be set to 10 minutes. It only allows 1 hour increments. There is no tail current setting in VE.Config for the Multiplus. I did update the BMV with a charged voltage of 28.6V, 7% tail current, and 1 minute charged detection time.

During the entire 4 hour charging cycle that I monitored today I was drawing a constant 63W from lights, phone charger, and other little things connected to the system.

Using a voltmeter I measured both batteries at 13.07V. The BMV showed 26.15V so that essentially matched.

I then connected to shore power. The inverter started bulk charging at roughly 41A and seemed to go down to about 35A over time. The bulk charging lasted about 2 hours 50 minutes and both batteries had the same voltage. My last measurement at 99% showed 13.79V on both batteries. So far so good.

As soon as the inverter switched to absorption charging I started recording the BMV status screen from VictronConnect. I recorded for the next hour and a quarter and I measured both batteries regularly with a voltmeter. The results are where it gets interesting.

For the next 39 minutes the voltage of the left battery went from 14.27V down to 14.06V at a fairly linear pace. The right battery went up from 14.32V (there was a 0.05V difference moments after absorption charging started) to 14.54 at the same rate the other battery went down. So during those first 39 minutes of absorption charging the two batteries kept a total between 28.58V and 28.60V. Meanwhile the charging current went from about 33A down to about 1A in the first 10 minutes. That drop was quick at first and then slowed down. The charge current then spent the next 29 minutes bouncing around between 0.2A and 0.5A.

At about the 39 minute mark into the absorption phase the charge current dropped to 0.0A and it got weird. The left battery was showing a steady 14.06V at that point. But the right battery's voltage was bouncing around like crazy. It was hard to see but the voltmeter was showing very rapid changes that seemed to be within the 14.2V to 14.8V range. Then for the next two minutes there was a roughly -0.2A current then back to 0.0A and then at the 41:30 mark of absorption the charge current went back to the very low positive current it had. Over the next 8 minutes of absorption charging the charge current was mostly between 0.2A and 0.5A with the occasional swing as high as 1.2A and bouts of 0.0A and a couple of -0.2ishA periods. The last 9 minutes of the 60 minutes of absorption charging the current was at 0.0A and the BMV showed the battery voltage fluctuate wildly between 28.3V and 28.8V but mostly between 28.5V and 28.7V (which makes sense since the inverter is set for an absorption charge of 28.6). During that last 9 minutes of voltage bouncing the left battery showed stable readings from 14.01V down to 13.99V. It was only the right battery that was jumpy whenever the charge current went to 0.0A (and only when it was at 0.0A).

After an hour of absorption the charger switched to float. Both batteries were stable and slowly dropping in voltage. I watched them for 15 minutes before shutting it all down. Just after switching to float, the left battery was at 13.88V and the right was at 14.32V. After 15 minutes they were at 13.53 and 13.70. So the gap was getting smaller as I've seen in the past. I suspect the voltages were the same after about another 30 minutes.

Any ideas on what's causing this and possible remedies? I can't imagine that it's normal during absorption charging for one battery to slowly lose voltage while the other slowly gains such that the total is always the correct 28.6V absorption charge.

What about one battery's voltage jumping around only while the charge current shows 0.0A and the same time the other battery's voltage holds steady?

One other tangent. During this whole exercise today I noticed that the BMV's display of the mid-point voltage difference was always exactly half of the actual voltage difference that I measured between the two batteries. For example, when I measure one battery at 14.10V and the other at 14.50V the BMV showed the difference was 0.20V, not the actual 0.40V.

Thanks to anyone that manages to read this whole post without passing out.
 
I have detailed results that are rather interesting I think. Between yesterday and today I ran the batteries from 100% SOC to 53%.

The inverter is set to 28.6V for absorption charging and 27.2V for float. The absorption time is set to 1 hour. It can't be set to 10 minutes. It only allows 1 hour increments. There is no tail current setting in VE.Config for the Multiplus. I did update the BMV with a charged voltage of 28.6V, 7% tail current, and 1 minute charged detection time.

Reduce absorption by 0.2V and assess behavior. Lower BMV to 0.2V below absorption.

Using a voltmeter I measured both batteries at 13.07V. The BMV showed 26.15V so that essentially matched.

Good.

I then connected to shore power. The inverter started bulk charging at roughly 41A and seemed to go down to about 35A over time. The bulk charging lasted about 2 hours 50 minutes and both batteries had the same voltage. My last measurement at 99% showed 13.79V on both batteries. So far so good.

This confirms that you're getting the batteries to extremely high states of charge without significant voltage deviation. 35A is closer to 0.17C - a fairly low rate for LFP.

As soon as the inverter switched to absorption charging I started recording the BMV status screen from VictronConnect. I recorded for the next hour and a quarter and I measured both batteries regularly with a voltmeter. The results are where it gets interesting.

For the next 39 minutes the voltage of the left battery went from 14.27V down to 14.06V at a fairly linear pace.

This may be a simple reduction in voltage in response to lowering charge current. I'm a little surprised, that it's that significant, but it may be indicative of this battery being at a slightly lower SoC than the other.

The right battery went up from 14.32V (there was a 0.05V difference moments after absorption charging started) to 14.54 at the same rate the other battery went down. So during those first 39 minutes of absorption charging the two batteries kept a total between 28.58V and 28.60V. Meanwhile the charging current went from about 33A down to about 1A in the first 10 minutes.

This was my goal in cutting the absorption time to 10 minutes. but I forgot that even Victron is limited in this regard. The only time I use my genny on the Quattro is when I'm desperate for juice.

That drop was quick at first and then slowed down. The charge current then spent the next 29 minutes bouncing around between 0.2A and 0.5A.

At about the 39 minute mark into the absorption phase the charge current dropped to 0.0A and it got weird. The left battery was showing a steady 14.06V at that point. But the right battery's voltage was bouncing around like crazy. It was hard to see but the voltmeter was showing very rapid changes that seemed to be within the 14.2V to 14.8V range.

This almost sounds like the battery's BMS was cycling on and off in response to a single cell going over voltage, which would be expected at 14.8V.

Then for the next two minutes there was a roughly -0.2A current then back to 0.0A and then at the 41:30 mark of absorption the charge current went back to the very low positive current it had. Over the next 8 minutes of absorption charging the charge current was mostly between 0.2A and 0.5A with the occasional swing as high as 1.2A and bouts of 0.0A and a couple of -0.2ishA periods. The last 9 minutes of the 60 minutes of absorption charging the current was at 0.0A and the BMV showed the battery voltage fluctuate wildly between 28.3V and 28.8V but mostly between 28.5V and 28.7V (which makes sense since the inverter is set for an absorption charge of 28.6). During that last 9 minutes of voltage bouncing the left battery showed stable readings from 14.01V down to 13.99V. It was only the right battery that was jumpy whenever the charge current went to 0.0A (and only when it was at 0.0A).

After an hour of absorption the charger switched to float. Both batteries were stable and slowly dropping in voltage. I watched them for 15 minutes before shutting it all down. Just after switching to float, the left battery was at 13.88V and the right was at 14.32V. After 15 minutes they were at 13.53 and 13.70. So the gap was getting smaller as I've seen in the past. I suspect the voltages were the same after about another 30 minutes.

To be expected as their voltages settle.

Any ideas on what's causing this and possible remedies?

Snippet of post #13:

Your problem has gone from an axe wound to a mild hangnail.

IMHO, if my hypothesis is right, this is something you will experience on a regular basis because the batteries themselves perform slightly differently, and it will be worse the longer you go between charges. This is likely the result of batteries that perform within spec, but do not perform well with each other in series.

Minimize the issue:

  1. Set proper charge termination criteria on your Multiplus. Limit absorption time to 10 minutes. If it allows tail current, specify it at 10A
  2. Set BMV charged current to 28.6V, 7% tail current, charged detection time to 1 min.
  3. Purchase and install the balancers I linked in post #2.

Since you're still experiencing it, and you can't set tail current or absorption less than 1hr, reduce voltage by 0.2V until this problem goes away.

I can't imagine that it's normal during absorption charging for one battery to slowly lose voltage while the other slowly gains such that the total is always the correct 28.6V absorption charge.

When you reduce current, voltage drops.

If battery A is at 98% and battery B is at 100%, as current decreases, battery A voltage may decrease due to reduced current, but battery B, since it's full, will still increase in voltage even at lower currents.

Again, you're likely dealing with two batteries that are at slightly different states of charge and may have different self-discharge characteristics.

What about one battery's voltage jumping around only while the charge current shows 0.0A and the same time the other battery's voltage holds steady?

BMS.

One other tangent. During this whole exercise today I noticed that the BMV's display of the mid-point voltage difference was always exactly half of the actual voltage difference that I measured between the two batteries. For example, when I measure one battery at 14.10V and the other at 14.50V the BMV showed the difference was 0.20V, not the actual 0.40V.

Maybe the BMV is reporting a ± value?

Your batteries continue to be at a very slight difference in state of charge. Recommend you cut absorption voltage by 0.2V until this behavior goes away. If you can't get it to go away by 28V, the balancers are probably warranted, and/or another parallel charge that pushes to a higher voltage.

Something to consider:

If you buy battleborn batteries, they want you to notify them that you intend to place them in series, so they can match your batteries. Additionally, they recommend a parallel charge once per year to keep the batteries in sync. This is all we're dealing with here - a very small difference in the SoC of your batteries... they're not quite top balanced.
 
Thanks again for taking the time to go through these details with me. It's been very helpful.

Reduce absorption by 0.2V and assess behavior. Lower BMV to 0.2V below absorption.
Regarding the BMV, do you mean the "Charged Voltage" setting? At the moment the inverter's absorption charge is 28.6V and so is the BMV's "Charged Voltage". If I understand this I should try 28.4V and 28.2V respectively, correct?

This confirms that you're getting the batteries to extremely high states of charge without significant voltage deviation. 35A is closer to 0.17C - a fairly low rate for LFP.

My shore power is plugged into a regular 15A home outlet which is probably keeping the charge current low. The inverter is actually setup to charge at up to 70A but that probably requires more than 15A coming in. The SOK data sheet recommends 40A charge current but I've been told that is per battery so in theory I can charge at 80A.

When you reduce current, voltage drops.

If battery A is at 98% and battery B is at 100%, as current decreases, battery A voltage may decrease due to reduced current, but battery B, since it's full, will still increase in voltage even at lower currents.

This confuses me. Why would a full battery get fuller at a lower current?

Your batteries continue to be at a very slight difference in state of charge. Recommend you cut absorption voltage by 0.2V until this behavior goes away. If you can't get it to go away by 28V, the balancers are probably warranted, and/or another parallel charge that pushes to a higher voltage.

Seems like a reasonable approach. Winter is returning where I live for the rest of the week so I probably won't be able to do more on this until next week.

But can you clarify what behavior I really should be seeing if everything was working perfectly? I assume during absorption both batteries should be keeping the same voltage as each other, not one going down while the other goes up.

The bulk charge phase is high current at the absorption charge voltage. Then the absorption phase is still the same voltage but low current. What should the battery voltages actually be doing during this low current absorption phase?

If you buy battleborn batteries, they want you to notify them that you intend to place them in series, so they can match your batteries.

I thought I brought that up when ordering the SOK batteries but I can't be sure at this point.

Again, thanks for your patience and help in this matter.
 
Thanks again for taking the time to go through these details with me. It's been very helpful.


Regarding the BMV, do you mean the "Charged Voltage" setting? At the moment the inverter's absorption charge is 28.6V and so is the BMV's "Charged Voltage". If I understand this I should try 28.4V and 28.2V respectively, correct?

Correct.

My shore power is plugged into a regular 15A home outlet which is probably keeping the charge current low. The inverter is actually setup to charge at up to 70A but that probably requires more than 15A coming in. The SOK data sheet recommends 40A charge current but I've been told that is per battery so in theory I can charge at 80A.

Understand. I'm limited on my quattro as well because of the 15A limit on one leg of my genny.

This confuses me. Why would a full battery get fuller at a lower current?

Because of the influence of the other battery. The charger is maintaining a total voltage.

Seems like a reasonable approach. Winter is returning where I live for the rest of the week so I probably won't be able to do more on this until next week.

But can you clarify what behavior I really should be seeing if everything was working perfectly? I assume during absorption both batteries should be keeping the same voltage as each other, not one going down while the other goes up.

Correct - if the batteries were perfectly top balanced with one another.

The bulk charge phase is high current at the absorption charge voltage. Then the absorption phase is still the same voltage but low current. What should the battery voltages actually be doing during this low current absorption phase?

If they were in perfect balance, they should be doing nothing but staying at the absorption voltage as the current tapers.

I thought I brought that up when ordering the SOK batteries but I can't be sure at this point.

They were likely not shipped fully charged, and they needed to be top balanced in parallel with the 12V charger.

Again, thanks for your patience and help in this matter.

If you can do so conveniently, something else to try would be to put the 12V charger on the lower voltage battery and top it off (leave system intact, but don't use it). You might find that evens it out OR you might find the problem moves to that battery, which just verifies we're dealing with batteries that are slightly out of balance with each other.
 
I finally have an update and it seems to be good too. Due to weather and schedule it took a couple of weeks.

tl;dr - I did a shore power charge and the max deviation I saw during the 1 hour absorption was 0.02V. Yeah!

Last week I ended up removing my two batteries and brought them into the house. I connected them in parallel and let them sit together for a few hours. The batteries were at about 51% SOC. Then I hooked up the Victron battery charger. I had adjusted the charge settings a tiny bit higher. I let the batteries bulk charge, then absorption for an hour, then float for a while. It think it took about 13 hours before the charger switched to storage. I disconnected the charger but I left the batteries connected in parallel for two days. The first time I did this parallel charge I disconnected the batteries as soon as the charger was done.

After being left sitting for a couple of days connected in parallel I got the batteries put back in place and in series. Yesterday (a week later) the batteries were down to 40% SOC and the deviation was still 0.0V. I plugged in the shore power for the first time since the last test. I got the SOC up to about 95% during bulk charge but then I couldn't sit and monitor any more so I disconnected shore power and let it sit over night. This morning I plugged shore power back in to let it finish. The bulk charge switched over to absorption when the BMV showed about 98.7% SOC and the deviation was still at 0.0V. Last time things went crazy at this point. Today it went very well. During the first 45 minutes of absorption the SOC slowly trickled up to 100% while the voltage on the BMV showed 28.56 - 28.6V and the charge current slowly decreased from about 40A to 0A. The best part is the deviation was no more than 0.01V. In the last 15 minutes of the 1 hour absorption charge the deviation maxed out at 0.02V.

Then it switched to float charge and the battery voltage slowly settled down to the float voltage of 27.6V and the deviation slowly dropped back to 0.0V.

This is so much better than any prior attempts. I think doing the 2nd full and proper parallel charge to get the batteries really at an equal SOC helped a lot. I don't think my first parallel charge attempt was allowed to fully complete.

Thanks again snoobler for your help. I feel so much better now that the batteries aren't going crazy during absorption charging. Being at the same SOC was, of course, the key to this.

So this leaves me with two questions:

1. Is the temporary 0.02V deviation toward the end of the 1 hour absorption charge acceptable long term? If that happens every time, is that OK?

2. At what point should the BMV reach 100% during a charge session like this? VictronConnect lets you reset the SOC to 100% to calibrate. Should I do that at some point and if so, when?
 
Back
Top