Importance of equal length battery cables

[REO]

I found a place where I could add an additional battery to my house bank. This would really be helpful, since my Albin 27 has limited space for batteries. The problem is that the space is about 5 feet "as the cable flies" from the rest of the batteries. The rest of the house batteries will have quite short and equal length connecting cables. What will happen if I add this battery? Is there a way to correctly do it?

 

[DDW]

There is a correct way to do it, and if done that way the batteries will be fine.

Parallel the batteries, but take the positive cable to the load from one end of the parallel chain, and the negative from the other end. This makes certain that any current in and out has to go through all of the parallel connections and cable drops, so that every battery is treated equally. Plenty of diagrams online for how to do this, if that explanation wasn't clear.

 

[mvweebles]

Would add that the cable should be sized properly. Do not cheap-out. My guess is this will work out okay with standard flood acid batteries. Would not work out well with LiFePO4 as they require balanced charging.

Peter

 

[C lectric]

I have a situation where there was not a practical way of modifying my EXISTING connections from batteries to the main power connection, bussbar, so the wire resistance and thus the current in each wire was balanced..

So I changed ALL the cables so they were the same lengths per polarity. Note the Neg leads must be the same lengths as each other and the Pos. must be the same lengths as each other. The Neg & Pos SETS though do not have to be the same length as each other.

I just bundled the excess up in a disorganized fashion so none of them would produce a coil. Do not coil or bundle in the same direction or that will produce a magnetic coil and field.

 

[ksanders]

In reality you are dealing with 4/0 cable probably.

so... you are talking about micro ohms of difference in resistance.

My opinion is to parallel the new batterie(s) into your existing bank and continue on with your life.

Remember the basics. Current * resistance = voltage drop.

Even if current is a large number if resistance is near zero the difference in voltage is near zero.

Some here are for sure going to disagree, but folks I have done this for a living my entire life and I am an old man. A boat is not a laboratory setup.

Parallel in the new batteries and go about your happy life.

 

[REO]

I didn't mention that my new house bank will be LifePo4. Guess I should have??

 

[mvweebles]

I know LiFePO4 batteries should be balanced very closely in voltage before hooking in parallel. I don't know what happens of they are not.

Question: is it possible to put a large bus bar or power-post halfway between them and run each battery with equal cables to there, hub-and-spoke? Would need one for hot, one for negative.

Peter

 

[mvweebles]

Some here are for sure going to disagree, but folks I have done this for a living my entire life and I am an old man. A boat is not a laboratory setup.

Parallel in the new batteries and go about your happy life.

I don't know enough to disagree, but after you posted this, OP updated batteries are LFP. I'm not positive, but suspect introduction of individual BMSs might make a difference. For AGM/FLA, I totally agree.

Thoughts?

Peter

 

[ksanders]

I don't know enough to disagree, but after you posted this, OP updated batteries are LFP. I'm not positive, but suspect introduction of individual BMSs might make a difference. For AGM/FLA, I totally agree.

Thoughts?

Peter

We are talking about microvolts of difference measured at the battery terminal when under a very high load or charge current.

In a practical application you will not see any problems.

 

[twistedtree]

I didn't mention that my new house bank will be LifePo4. Guess I should have??

Then it matters quite a bit. The cable lengths to parallel batteries need to be exactly the same.


For lead acid, provided the batteries get regularly brought back to full charge, it matters very little.

 

[twistedtree]

I know LiFePO4 batteries should be balanced very closely in voltage before hooking in parallel. I don't know what happens of they are not.
Question: is it possible to put a large bus bar or power-post halfway between them and run each battery with equal cables to there, hub-and-spoke? Would need one for hot, one for negative.

Peter

Yes, that a very good way to do it with more than a couple of batteries.

 

[DDW]

Then it matters quite a bit. The cable lengths to parallel batteries need to be exactly the same.


For lead acid, provided the batteries get regularly brought back to full charge, it matters very little.

You're gonna have to explain that to me. The cable pair between batteries (positive and negative) should be the same length. The pair length between batteries doesn't matter (if properly sized etc.) provided the load is taken from each end. At least that is my primitive understanding of it (as an EE).

Consider this cartoon:



Three batteries in parallel, one with cables twice as long. A primitive model of the battery current flow is as a resistor representing its internal resistance. It is the same if all three batteries are the same. I've represented the cable resistance as a unit resistance, with the long pair twice the short pair. Now consider the current flow from the positive load to the negative load: The same resistance is encountered through any of the three batteries, 3Ω + R. They will charge and discharge identically. This should be true for LA and LFP or any other chemistry.

Is there something about LFP that I don't understand which would make this untrue?

 

[REO]

The large bus bar in the middle between the batteries may work. I will need a new bus bar for my new inverter, so relocating may not be a huge deal.The sources I use online don't show anything big enough for this use. Any suggestions for heavy duty electrical parts?

I have read and re-read the "Marine How to" information on LiFePO4 batteries. I'm fairly confident that the simple system I am building will perform well. Maybe trying to add another remote LiFePO4 battery is too risky. Has anyone done this?

 

[twistedtree]

You're gonna have to explain that to me. The cable pair between batteries (positive and negative) should be the same length. The pair length between batteries doesn't matter (if properly sized etc.) provided the load is taken from each end. At least that is my primitive understanding of it (as an EE).

Consider this cartoon:

View attachment 136859

Three batteries in parallel, one with cables twice as long. A primitive model of the battery current flow is as a resistor representing its internal resistance. It is the same if all three batteries are the same. I've represented the cable resistance as a unit resistance, with the long pair twice the short pair. Now consider the current flow from the positive load to the negative load: The same resistance is encountered through any of the three batteries, 3Ω + R. They will charge and discharge identically. This should be true for LA and LFP or any other chemistry.

Is there something about LFP that I don't understand which would make this untrue?

Oh, boy, you are challenging my 40 year stale EE skills. I may be easy pickings.


Wired as you show I think it's OK, but I don't think that's how he was going to wire them. But looking back I think we don't really know so I was assuming. I figured he's has a bank wired appropriately, and then was going to run two longer cables to a remove battery. In that case the remote battery will see more voltage drop, so share less in the current load. So it will discharge less and charge less than the other batteries.


With LA this is less of a concern since the absorb and float cycles are effectively balancing the cells. With LFP there is no such opportunity, other than balancing from electronic balancers. They might keep up, but they also might not. But if that one battery doesn't work as hard as the others, does it really matter? Probably not, other than loss of potential capacity.


Somewhere I came across an analysis of diagonal wiring like you have shown, and it was a surprise to me. If I remember correctly, diagonal wiring was good for up to there batteries, but above that it actually isn't balanced. But I can't find it. What I remember is that it struck me as being from an unlikely source.

 

[mvweebles]

https://www.pysystems.ca/resources/...-lithium-battery-bank-split-in-two-locations/

"It is highly recommended when lithium batteries are connected into a bank that common positive and negative bus bars are installed with even length cables from each battery. This is because the internal resistance of a lithium battery is far lower than a traditional flooded lead acid battery and even slight differences in cable lengths may cause individual batteries to charge or discharge at different rates."
​

 

[DDW]

Well, you wouldn't want to just run two cables to a new remote battery. You run two (equal length) cables, then a third to either the positive or negative load and disconnect the existing one. That should work regardless of how many batteries are in parallel. The main points are: pairs of cables between batteries are the same gage and length, the batteries are a true daisy chain of parallel connections (not a star or mixed topology), and the positive and negative connections to the load/charge are at opposite ends of the chain. The batteries also have to have the same characteristics. But beyond that, different lengths and even wire gages will not affect the balance - they all see the same circuit resistance. If the contact resistance at a connection is bad, then they may become unbalanced, but that is true no matter the connection topology.

 

[DDW]

https://www.pysystems.ca/resources/...-lithium-battery-bank-split-in-two-locations/

"It is highly recommended when lithium batteries are connected into a bank that common positive and negative bus bars are installed with even length cables from each battery. ...
​

If you look at my cartoon, you will see that each battery sees exactly the same cable length as every other battery.

 

[mvweebles]

Like TT, I wish I could find the article I read a couple years ago. The author put four batteries in parallel and cabled them in several different configurations. The amp draw varied considerably between the four depending on how they were cabled. The only one that gave a balanced draw was the hub-spoke config to a bus bar.

But hey, I have zero initials after my name so no creds here. I do wish I could find the article - was definitely interesting to see the actual diminished power ratings.

Peter

 

[REO]

The heavy duty bus bar hub-spoke seems like a reasonable solution. Where can I purchase one? I am becoming more convinced that an LiFePO4 battery could be added in a remote location. Has anyone done this?

 

[ksanders]

Oh, boy, you are challenging my 40 year stale EE skills. I may be easy pickings.


Wired as you show I think it's OK, but I don't think that's how he was going to wire them. But looking back I think we don't really know so I was assuming. I figured he's has a bank wired appropriately, and then was going to run two longer cables to a remove battery. In that case the remote battery will see more voltage drop, so share less in the current load. So it will discharge less and charge less than the other batteries.


With LA this is less of a concern since the absorb and float cycles are effectively balancing the cells. With LFP there is no such opportunity, other than balancing from electronic balancers. They might keep up, but they also might not. But if that one battery doesn't work as hard as the others, does it really matter? Probably not, other than loss of potential capacity.


Somewhere I came across an analysis of diagonal wiring like you have shown, and it was a surprise to me. If I remember correctly, diagonal wiring was good for up to there batteries, but above that it actually isn't balanced. But I can't find it. What I remember is that it struck me as being from an unlikely source.

OK, Now you have to explain it to me, why lithium batteriers care about the tiny voltage difference.

Lets imagine for a moment that the OP has for round numbers 100 amps of charging capability maximum over his normal boat loads. That is honestly about all he is going to get unless he has parallel chargers.

Now lets imagine that he has 4/0 wire and has a 10' run to the new battery.

Ancor brand 4/0 marine wire has .05 ohms per 1,000 feet.

Two 10' runs (positive and negative) = .001 ohm

100 amps will yield a voltage drop of .01 volts.

But remember... he is not getting 100 amps of current into the new battery. He is getting his charge current divided by the number of batteries in parallel. Most lithium batteries I've seen are about 100AH at 12 volts. I'm guessing here but he is getting his 100 amps divided by 4 or 5 or maybe even 10 if he has a 1000 AH bank.

So in reality we are talking about a real world voltage difference on his new bank of around 0.002 volts

I just do not see it affecting anything from a practical standpoint. I would even bet that the charge controllers on most chargers are not accurate to .002 volts.

 

[Ka_sea_ta]

I have to agree with Kevin on this, just remote the battery using at least 2 /0 cable. The voltage drop of the cable will be negligible especially when you include crimp connectors, screw down terminal lugs, the batteries internal wiring and electronics of the BMS. You guys are way over thinking this, This will be installed on a boat not in a lab. Look at the internals on this highly rated Epoch battery https://panbo.com/epoch-batteries-elegant-drop-in-lifepo4/ do you think each one of the those connections and the FETs on the BMS are matched for internal resistance. The output wiring to the terminals is a doubled 8 gauge wire.

 

[mvweebles]

The heavy duty bus bar hub-spoke seems like a reasonable solution. Where can I purchase one? I am becoming more convinced that an LiFePO4 battery could be added in a remote location. Has anyone done this?

Not cheap - but these 600A bus bars are decent. There are cheap knock-offs, but for electrical, I stick with Blue Sea Systems.

https://www.bluesea.com/products/2104/PowerBar_600A_BusBar_-_Four_3_8in-16_Studs

Peter

 

[SteveK]

Bookmarked

https://www.solar-electric.com/lib/wind-sun/Iota_balanced_charging.pdf
https://thecampingnerd.com/how-to-wire-multiple-12v-or-6v-batteries-rv/

 

[SteveK]

OK, Now you have to explain it to me, why lithium batteriers care about the tiny voltage difference.
............................

Kevin, In my research there were several sites that insisted that LFP batteries are charged to the same or very close level before they are connected in parallels otherwise they will forever not be equal in charge/discharge.
Don't understand it, just what I read.

 

[twistedtree]

OK, Now you have to explain it to me, why lithium batteriers care about the tiny voltage difference.

Lets imagine for a moment that the OP has for round numbers 100 amps of charging capability maximum over his normal boat loads. That is honestly about all he is going to get unless he has parallel chargers.

Now lets imagine that he has 4/0 wire and has a 10' run to the new battery.

Ancor brand 4/0 marine wire has .05 ohms per 1,000 feet.

Two 10' runs (positive and negative) = .001 ohm

100 amps will yield a voltage drop of .01 volts.

But remember... he is not getting 100 amps of current into the new battery. He is getting his charge current divided by the number of batteries in parallel. Most lithium batteries I've seen are about 100AH at 12 volts. I'm guessing here but he is getting his 100 amps divided by 4 or 5 or maybe even 10 if he has a 1000 AH bank.

So in reality we are talking about a real world voltage difference on his new bank of around 0.002 volts

I just do not see it affecting anything from a practical standpoint. I would even bet that the charge controllers on most chargers are not accurate to .002 volts.

Here is my understanding of the issue.


Let's take your example of 100A charge current, and I'll say there are 4 batteries in parallel. Each battery is connected to a bus bar with 10' 4/0 cables, so 20' round trip, and .001 ohms each. I think this is the same as your example, but I have added that there are 4 such batteries in parallel.


Let's assume the cable for one battery are 6" longer, so in total 21' rather than 20'. Those longer cables will be .00105 ohms vs the .001 ohms for the other three batteries.


Now apply the charge current. The current will divide up according to the resistance of each set of cables. If my math is correct, the 10' cable batteries will get 25.1A each and the 10.5' cable battery will get 24.1A. That three shorter cable batteries will reach full charge and their voltage will spike before the long cable battery does. As the fully charged battery voltages jump up, more current will go to the lower charged battery, but because the end of charge happens so quickly, the low battery is unlikely to reach full charge. Then on discharge the longer cable battery isn't being drawn down as much, so overall it works less than the others.



Anyway, this is my understanding of why LFP manufacturers call for equal length cables. With parallel inverters, Victron also calls for it for both DC cables and AC output cables in order to get even loading between the parallel inverters.

 

[ksanders]

Here is my understanding of the issue.


Let's take your example of 100A charge current, and I'll say there are 4 batteries in parallel. Each battery is connected to a bus bar with 10' 4/0 cables, so 20' round trip, and .001 ohms each. I think this is the same as your example, but I have added that there are 4 such batteries in parallel.


Let's assume the cable for one battery are 6" longer, so in total 21' rather than 20'. Those longer cables will be .00105 ohms vs the .001 ohms for the other three batteries.


Now apply the charge current. The current will divide up according to the resistance of each set of cables. If my math is correct, the 10' cable batteries will get 25.1A each and the 10.5' cable battery will get 24.1A. That three shorter cable batteries will reach full charge and their voltage will spike before the long cable battery does. As the fully charged battery voltages jump up, more current will go to the lower charged battery, but because the end of charge happens so quickly, the low battery is unlikely to reach full charge. Then on discharge the longer cable battery isn't being drawn down as much, so overall it works less than the others.



Anyway, this is my understanding of why LFP manufacturers call for equal length cables. With parallel inverters, Victron also calls for it for both DC cables and AC output cables in order to get even loading between the parallel inverters.

OK, good explanation of the theory, Thanks!

So... Lets use Victrons approach here and use your "test setup" of four 100AH batteries and assuming we are starting at 25% charge.

You turn on the charger and it ramps up to 14.2 volts.

After 179 minutes and 16 seconds the first three batteries reach full charge and their current drops off.

The fourth battery has received 72.01 amp hours of current resulting in a charge state of 97%

The battery charger switches to float mode of 13.6v sits there with one battery a bit undercharged.

so in theory you are 100% correct.

But... How does the charger know the batteries are at a full charge and switch to float mode at 13.6 volts??? The current has to drop below the residual current number, often set at 4%, or the absorption voltage timer must expire.

This is where it gets tricky... Why??? Because the residual current number of 4% does not include the loads that are on the battery bank. Remember, your boats loads do not stop just because you are charging, and those loads are not all that easy to predict with any accuracy.

so... what really happens is the charger never reaches it's residual current number and continues to charge at 14.2 volts until the absorption charge timer expires. Or... You pick a number based on what you think your loads might be and your batteries are sometimes undercharged.

I would argue that although your logic and theory are 100% correct, in real applications there is enough "slop" in the system that the cable lengths are not as important as one might theorize.

 

[Insequent]

Here is an article on battery interconnections that I saved a while back. Not sure if it is the same one that a couple of posts above mentioned was from an unlikely source?

SmartGauge Electronics - Interconnecting multiple batteries to form one larger bank

 

[twistedtree]

OK, good explanation of the theory, Thanks!

So... Lets use Victrons approach here and use your "test setup" of four 100AH batteries and assuming we are starting at 25% charge.

You turn on the charger and it ramps up to 14.2 volts.

After 179 minutes and 16 seconds the first three batteries reach full charge and their current drops off.

The fourth battery has received 72.01 amp hours of current resulting in a charge state of 97%

The battery charger switches to float mode of 13.6v sits there with one battery a bit undercharged.

so in theory you are 100% correct.

But... How does the charger know the batteries are at a full charge and switch to float mode at 13.6 volts??? The current has to drop below the residual current number, often set at 4%, or the absorption voltage timer must expire.

This is where it gets tricky... Why??? Because the residual current number of 4% does not include the loads that are on the battery bank. Remember, your boats loads do not stop just because you are charging, and those loads are not all that easy to predict with any accuracy.

so... what really happens is the charger never reaches it's residual current number and continues to charge at 14.2 volts until the absorption charge timer expires. Or... You pick a number based on what you think your loads might be and your batteries are sometimes undercharged.

I would argue that although your logic and theory are 100% correct, in real applications there is enough "slop" in the system that the cable lengths are not as important as one might theorize.

I totally agree for LA. Holding the voltage high for a long absorb time, followed by a float voltage that continues to apply current is just fine, if not preferable for LA. The full batteries will over charge a but, but that's fine. And in the mean time the laggards will catch up.


Less so for LFP. A good charge profile brings them up to 13.8 to 14v and only holds absorb for 10 minutes or so. Some batteries call for more so the balancers can run, but it's a far cry from the hours that LA get. And I don't know any LFP profiles that use return amps.


Another contributing factor is that across the middle range of SOC, LFP voltage is very flat. It's hard to tell the difference between 40 and 70% SOC. So operating in that range, current differences will driven by the conductor impedance differences more than SOC of different batteries. This will exacerbate drift in SOC. Difference in SOC really only show at the end of the charge cycle when the voltage pops up. That's when current flow will be more influenced by differences in cell SOC, but it's for a much shorter period of time vs LA.


All this said, lots of batteries are built with internal cells that are paralleled, not just whole batteries that are paralleled, so it's a solvable problem.

 

[mvweebles]

Here is an article on battery interconnections that I saved a while back. Not sure if it is the same one that a couple of posts above mentioned was from an unlikely source?

SmartGauge Electronics - Interconnecting multiple batteries to form one larger bank

Very similar. I extracted a graphic showing the batteries daisy-chained so positive and negative are at opposite ends of the paralleled batteries. Note the difference in amp-load between the four batteries - two at ends are just under 27-amps and two inside ones are just over 23-amps. Pretty sizeable difference. He shows two balanced configurations, including the hub/spoke to a bus bar.

Thanks for this -

Peter

 

[STB]

For whatever it is worth, I'm with ksanders on this one....

Tolerances are important in engineering.

In this case, I see bigger differences from the quality of the connection at the terminals than from the wire length, especially over time.

In any case, if you are worried about it, upsize the wire gauge, total resistance will go down, and the absolute difference in wire resistance will go down proportionally.