Electrical Musings

[FlyingCloud1937]

Sorry, the post above...misstated the Gell volts per cell as 2.4....but should have been 2.35 volts per cell.
Moderator doesn't allow changes.

Lloyd

[chala]

Originally Posted by Dockhead

But I am firmly convinced of one thing about battery bank architecture -- service bank for deep discharge loads should be a single big one.

Originally Posted by Dockhead

Trojan, 12v, 105ah each, 8 of them wired up as 4 virtual 105ah/ 24v batteries, giving (theoretically) 420 ah @ 24v nominal.

It would have been better if you had two 24V 210A banks connected in parallel. Paralleling batteries is also known as “jumping the battery” which is dear to backyard mechanics. ”Jumping the battery” is also known to be a risky business but when ”Jumping the battery” is done permanently you get an inefficient, unsafe and inaccurate configuration. Inaccurate because the voltage of a configuration in parallel is an average. Of course it is possible to read the SG of each cell and at the same time disconnect and reconnect that “field wire (blue)”, if it is blue, while underway.

If the two 24V 225A banks were not connected in parallel but independent of each other it would be possible to determine for each of them a constant current. By comparing the constant current data with the appliance that has the highest current, but not including a large inverter, it would be possible to determine how long that appliance could be powered by one bank. If suitable, the same comparison could be done with the other bank versus the rest of the load. The indicative attachment comes from an EXIDE Specifier’s Manual. The manufacturer of your batteries should be able to supply a similar document.

[Wotname]

Chala, I would be interested to know exactly what battery arrangements you have on your boat.
I apologise if you have already posted this information and I have missed it; if so perhaps link me to it .

Dockhead, please excuse the thread drift but I think it might be useful for us to know what Chala uses.

[noelex 77]

Originally Posted by chala

It would have been better if you had two 24V 210A banks connected in parallel. Paralleling batteries is also known as “jumping the battery” which is dear to backyard mechanics. ”Jumping the battery” is also known to be a risky business but when ”Jumping the battery” is done permanently you get an inefficient, unsafe and inaccurate configuration.

"jumping a "battery" is unsafe because of the tempory nature of the connections.

Originally Posted by chala

If the two 24V 225A banks were not connected in parallel but independent of each other it would be possible to determine for each of them a constant current. By comparing the constant current data with the appliance that has the highest current, but not including a large inverter, it would be possible to determine how long that appliance could be powered by one bank. If suitable, the same comparison could be done with the other bank versus the rest of the load. The indicative attachment comes from an EXIDE Specifier’s Manual. The manufacturer of your batteries should be able to supply a similar document.

The table is of limited use on a boat, but it's easy to construct a table for two batteries connected in parallel.
If you just double the duration the results will be reasonable. In practice you will get longer run times than that due to the greater efficiency of discharging the battery at a lower rate. In other words the greater efficiency of a larger bank.
Notice how the figures on the table reflect this. The battery capacity increases as you reduce the discharge current. This is one reason why a larger parellel bank is more efficient.

I do agree 2 house banks (and a seperate start) is superior , giving options for trouble shooting, running different batteries etc, but with identical batteries they should both be on effectively forming one big parallel bank for the most effient operation

[Dockhead]

Originally Posted by chala

It would have been better if you had two 24V 210A banks connected in parallel. Paralleling batteries is also known as “jumping the battery” which is dear to backyard mechanics. ”Jumping the battery” is also known to be a risky business but when ”Jumping the battery” is done permanently you get an inefficient, unsafe and inaccurate configuration. Inaccurate because the voltage of a configuration in parallel is an average. Of course it is possible to read the SG of each cell and at the same time disconnect and reconnect that “field wire (blue)”, if it is blue, while underway.

If the two 24V 225A banks were not connected in parallel but independent of each other it would be possible to determine for each of them a constant current. By comparing the constant current data with the appliance that has the highest current, but not including a large inverter, it would be possible to determine how long that appliance could be powered by one bank. If suitable, the same comparison could be done with the other bank versus the rest of the load. The indicative attachment comes from an EXIDE Specifier’s Manual. The manufacturer of your batteries should be able to supply a similar document.

Certainly, I could determine that, but why? With heavy loads, Mr. Peukert rears his ugly head at half the load, which means power in absolute terms is wasted by driving heavy loads with smaller banks. The acceptance rate of each separated bank is half that of the combined bank. What's the point?

[Maine Sail]

Originally Posted by Dockhead

Certainly, I could determine that, but why? With heavy loads, Mr. Peukert rears his ugly head at half the load, which means power in absolute terms is wasted by driving heavy loads with smaller banks. The acceptance rate of each separated bank is half that of the combined bank. What's the point?

Bingo.... And figuring the load a bank can support, at a given rate is easy, if you know how to do Peukert calculation and have the 10 or 20 hour Ah rating for the batteries. If you've properly, and the key term there is properly, programmed a battery monitor it will do all that math for you.

Originally Posted by chala

Paralleling batteries is also known as “jumping the battery” which is dear to backyard mechanics. ”Jumping the battery” is also known to be a risky business but when ”Jumping the battery” is done permanently you get an inefficient, unsafe and inaccurate configuration.

We've still seen zero evidence from Chala that paralleling, or as he now so eloquently refers to it as "backyard mechanics do" "jumping" a battery, is an unsafe practice. I guess the engineers at Lifeline, Deka/East Penn, Rolls, Trojan, Odyssey and the folks at the NMMA, ABYC etc. etc. are all "back yard mechanics" too because they all among many other industries support the paralleling of batteries.

Again millions upon millions of parallel banks and none of the dire end of the world "unsafe" scenarios that Chala claims will happen.

Chala where is the DATA to back up your claims?? So far I see nothing but a potentially made up story about some deaths in Europe, we asked you for a link but you gave no supporting evidence. You then attempted to extrapolate that case as the reason those deaths occurred was due to the paralleling of batteries. That could not be farther from the truth... You then spout a bunch of other stuff that makes little to no sense all with nothing to support it. Please provide us with some legitimate data..

For now I suspect, that with every battery manufacturer I know of supporting parallel banks, the "experts" or manufacturers of the actual batteries have weighed in and they actually show you how to do it....

I will say this has been amusing...

[chala]

Originally Posted by Wotname

Err... the wire size remains constant regardless of number of banks as it voltage dependant. Big wire for 12V, not so big for 36V.
Also surprised to think that heat is primary consideration in determining what wire gauge to use.

Efficiency and cost is the primary consideration.

[Wotname]

Originally Posted by chala

Efficiency and cost is the primary consideration.

Ahh... I must have misunderstood you when you posted :

Originally Posted by chala

What about smaller banks sharing the load, smaller the wire the more efficient it is at dissipating the heat.

I took this to mean "smaller the wire the more efficient it is at dissipating the heat" - my bad

[chala]

Originally Posted by Dockhead

Certainly, I could determine that, but why? With heavy loads, Mr. Peukert rears his ugly head at half the load, which means power in absolute terms is wasted by driving heavy loads with smaller banks. The acceptance rate of each separated bank is half that of the combined bank. What's the point?

Let assume the biggest load on your boat is the heating system and it draw 10.5A according to the manufacturer table for a 6G105 you can obtain 10 Hr discharge from that bank. If the draw is 17A then according to the manufacturer table only 5Hr can be discharged and so on. The capacity of a battery changes according to the rate of discharge (Peukert Equation) is taken into consideration in that manufacturer table. For example 60Min is only 45A available out of a 105A battery.

Originally Posted by Wotname

Chala, I would be interested to know exactly what battery arrangements you have on your boat.

You click Chala check the threads and you find it but remember this 12/24V is a totally different cup of tea.

[chala]

Originally Posted by Wotname

Ahh... I must have misunderstood you when you posted : I took this to mean "smaller the wire the more efficient it is at dissipating the heat" - my bad

Don’t break your head against that wall; you did not misunderstand me, the smaller the wire the more efficient it is at dissipating heat. It just has to do with the ratio between the area and the circumference.

[mbianka]

Originally Posted by Dockhead

My problem is the right and most efficient charging regime, not the amount of power consumed.

Dockhead:

I have two separate battery systems on board. I take a three legged stool approach to charging for my 10 kw 48 volt electric propulsion system. I use a combination of solar, wind and a Honda 2000 generator. I also have the ability to plug in to the grid but, I rarely find my self at a dock.
For my 12 volt house bank I have two 75 watt solar panels that pretty much keep up with my ENGEL freezer most days. If I see the battery voltage heading downward a little too much I will crank up the Honda and run it for a half hour or so to bring things up via the charger. I also converted most of my lighting on board to LED's. Best to reduce load where ever you can rather than add more battery capacity. I prefer to get most of my charging from solar or wind and only turn to the generator when really needed.

[twistedtree]

I think this series/parallel debate is much ado about nothing. Keep in mind that the internal construction of a battery consists of a bunch of PARALLEL wired plates. Higher Ah capacity batteries get that higher capacity through some combination of larger plates and/or more plates in parallel.

My point is that like it or not, you already have parallel wired cells, so get over it. Under normal operation, assuming you have things wired with proper size cables, good connections, etc, a parallel bank is no different than the parallel cells inside a single battery.

Then it becomes about failure modes and what happens. If a plate shorts inside a battery, it forces the discharge of the other parallel plates in that cell, and drops the battery voltage by 2.2V (nominal cell voltage). If that battery is paralleled with another battery, the voltage drop caused by the shorted cell will drag down the other battery as well until it's voltage has dropped by 2.2V.

It's true that the more parallel batteries you have, the more energy that will be discharged through the faulty, shorted plate, and the higher the risk of overheating etc. This is why God invented fuses, and is why AYBC guidelines call for fusing within a short distance from batteries.

It's also true that such a failure, if left unattended, will ultimately ruin the good battery, but a 2.2V drop in your battery bank voltage should quickly capture the attention of all but the most unobservant owners. Once you notice the problem, you can isolate the faulty battery and continue to operate at reduced capacity with the remaining battery.

In contrast, if your battery bank is made of two 6V batteries wired in series, the same shorted plate fault will once again drag down the bank voltage by 2.2V - same as the parallel case. Assuming the net Ah capacity of the batteries is the same, the same amount of energy will be discharged through the shorted plate creating the same risk of overheating etc. as the parallel configuration. If you are unobservant and continue to operate with the bad cell, you will ultimately ruin the remaining good battery, just like in a parallel setup.

The big difference is that in a series-only configuration you can't isolate the bad battery and keep on going because you are only left with 6V. This is ultimately the basis for arguments to have two battery banks, one way or another. It also argues for having an easy way to isolate/parallel those banks. This can be accomplished with paralleled house batteries, or a house bank and a starter bank.

To me, arguments that center around problems that emerge when people don't use proper cables, have poor connections, etc. are misguided since the problem is poor system construction, not a fundamentally bad approach. This would be like condemning gen sets because Fischer-Pandas are unreliable (sorry, I couldn't resist).

[Bash]

Originally Posted by Wotname

I know it is possible and therefore it has probably happened but I wonder if any of the posters here have personal experience (or even second hand accounts) of batteries failing due to internal shorts.

I had that happen years ago with a D4 gel battery, back when gels were a new thing. I had three D4s in parallel, and the middle one shorted. The other two batteries fed into it, and basically melted it. Didn't discover the problem until I got back to my marina, and had to call the fire department to assist with removing the battery. I was told later that it bubbled for 24 hours after it was removed.

Haven't used gels since.

[twistedtree]

Originally Posted by chala

Don’t break your head against that wall; you did not misunderstand me, the smaller the wire the more efficient it is at dissipating heat. It just has to do with the ratio between the area and the circumference.

I suppose this is true when you look at heat dissipation capability (surface area) relative to wire resistance (cross sectional area). I think this is what you mean by "efficiency". But I think it has little to do with anything in practice, especially relative to this topic.

The larger a wire's cross sectional area, the lower its resistance. The lower its resistance, the less heat it will dissipate per amp of current. And a fatter wire also has a larger surface area. Yes, the surface area doesn't grow as fast as the cross section, but it does grow. So not only will the fatter wire generate less heat per amp, that fatter wire is also more capable of heat dissipation.

I think that a more practical view of "efficiency" in this application is to look at the loss per amp of current. A fatter wire will both generate less loss, and dissipate that loss more effectively i.e. run cooler than a smaller wire. Coincidentally, this is what wire sizing practices are based on.

[Wotname]

Originally Posted by twistedtree

I suppose this is true when you look at heat dissipation capability (surface area) relative to wire resistance (cross sectional area). I think this is what you mean by "efficiency". But I think it has little to do with anything in practice, especially relative to this topic.

The larger a wire's cross sectional area, the lower its resistance. The lower its resistance, the less heat it will dissipate per amp of current. And a fatter wire also has a larger surface area. Yes, the surface area doesn't grow as fast as the cross section, but it does grow. So not only will the fatter wire generate less heat per amp, that fatter wire is also more capable of heat dissipation.

I think that a more practical view of "efficiency" in this application is to look at the loss per amp of current. A fatter wire will both generate less loss, and dissipate that loss more effectively i.e. run cooler than a smaller wire. Coincidentally, this is what wire sizing practices are based on.

Thank you TT.
I had taken this information as a given, especially for those with some basic electrical knowledge (and therefore who are likely to read the thread).
You have taken the time and effort to explain it in detail .

The principle design criteria for wire sizing depends to a large extent on the industry sector in question. For instance, a recreational sailboat is operating in a very low volt unstablized low power capacity environment i.e. 12V small battery banks. Wire sizing (as TT and others suggest) is usually calculated on voltage drop as the prime concern. Aviation on the other hand, operates in a stable voltage and usually high power capacity environment where the prime concern becomes safe current carrying capacity, weight and insulation quality. Structured cabling for data/telephony has other prime concerns as does say high voltage transmission lines.

All of this in inconsequential to the thread except to show that various design criteria must exist relative to the industry / end need and to reinforce the accepted view (at least IMO) that in recreational sailboats, wire is sized primarily on voltage drop considerations.

I accept that some have differing views