Switch or Breaker?

[geoffr]

This is what I've come up with as a breaker/switch plus contactor was much larger & more expensive...

[pbiJim]

Originally Posted by geoffr

Thanks, but my actual question is:

Where I get confused is when this is a breaker as I thought that DC circuit breakers could only accept current flowing in one direction - am I wrong?

About 15 years ago, I worked with some programmable breakers that only wanted to see power in one direction. This was on a shore side application. I have not seen those breakers used in boats.

The mechanical breakers that I normally see used in boats will trip regardless of the direction of current flow. I know that from personal experience. I have not seen any issues arising from back feeding a marine DC breaker. If someone does know of a reason that should not be done, please fill me in.

That being said, I do recall that on the little rectangular breakers that have a pair of #10-24 studs to attach wires to, often one stud is copper color & the other is silver. I have been told to attach power to the copper side & the load to the silver side. When I asked why, I was told - that's the way you are supposed to do it. I have yet to receive an acceptable answer to that question. I have yet to hear why. Perhaps there is a good reason. Perhaps there is not.

I have not yet contacted the breaker manufacturer to seek their guidance. That is probably the best place to ask the question.

[CaptainRivet]

Originally Posted by pbiJim

That being said, I do recall that on the little rectangular breakers that have a pair of #10-24 studs to attach wires to, often one stud is copper color & the other is silver. I have been told to attach power to the copper side & the load to the silver side. When I asked why, I was told - that's the way you are supposed to do it. I have yet to receive an acceptable answer to that question. I have yet to hear why. Perhaps there is a good reason. Perhaps there is not.

I have not yet contacted the breaker manufacturer to seek their guidance. That is probably the best place to ask the question.

Silver is more conductive then cooper, so having the load on the silver side presents less resistance and a better permanent contact as silver won‘t react with cooper or Alu wires (no oxide layer that effects conductivity) over time.

[CaptainRivet]

Originally Posted by geoffr

This is what I've come up with as a breaker/switch plus contactor was much larger & more expensive...

That EGIS double relay is dope and they even have a triple one. I didn’t know something like that exists and for a double 500A programmable relay 170€ is cheap compared to eg the blue system ones and it’s super compact.
My new goto ones, thanks.

[pbiJim]

Originally Posted by CaptainRivet

Silver is more conductive then cooper, so having the load on the silver side presents less resistance and a better permanent contact as silver won‘t react with cooper or Alu wires (no oxide layer that effects conductivity) over time.

The commonly available breakers I was describing have one copper color stud & one silver color stud. I believe that the copper color stud is actually copper. I do not believe that genuine metallic silver is used for the other stud, as I have never seen tarnish on them that would be consistent with real silver. I expect that it is plated copper, but the silver color plating is not actually silver.



I know that actual silver (or in some cases silver cadmium oxide) is used for the contacts inside many switches, relays & other types of devices that make & break electrical contact.

[CaptainRivet]

Originally Posted by geoffr

This is what I've come up with as a breaker/switch plus contactor was much larger & more expensive...

Regarding your 3p4S setup:
Did you measure the temps at the terminal of each of your 3 parallel cells under maximum load for 5min ?
If you installed like your drawing of cell setup the last and first cell gets much more load and wear faster then the rest of the pack. If you measure temp on each terminal I bet that first is about 3-5 degrees hotter then last cell. This shows the first contributes more then 2nd and 3rd cell of the pack even they are connected in parallel and load should be taken equally off which the temps proof is not the case.
To avoid this you need to take the load symmetrical off the pack, means you need to connect the main positive cable to middle cell 2 and main negative to middle cell 11. even better if you run 3 cables instead one, each connected to one of the 3 cells and on the other and with stacking lugs directly to the fuse and negative to shunt.
If your total load is below 50% of the single(!) cell capacity eg 140A for a 304AH cell (3x304Ah in parallel) then connecting to middle cell is enough, if above 50% then highly I recommend to run 3 cables each side, if max load is >100% then you must run 3 cables as you otherwise overload the first/last cell. This rule is for cells with 1 or 2C constant current capability cells like EVE, Lishen, Calb….Winston with 4C just double the percentages. That’s the main root cause you get runners in multi cells parallel battery pack.
Be aware that an inverter puts a 1.41x factor load on what your shunt shows you, means 100A DC shown from an inverter is actually 141A on/off multiple times a second load on the battery pack.

[CaptainRivet]

Originally Posted by pbiJim

The commonly available breakers I was describing have one copper color stud & one silver color stud. I believe that the copper color stud is actually copper. I do not believe that genuine metallic silver is used for the other stud, as I have never seen tarnish on them that would be consistent with real silver. I expect that it is plated copper, but the silver color plating is not actually silver.



I know that actual silver (or in some cases silver cadmium oxide) is used for the contacts inside many switches, relays & other types of devices that make & break electrical contact.

You are correct, it most likely silver coated cooper which is standard in high current environments like grid trafo stations or in big solar farms.
NH fuses eg using silver plated cooper and brass too. That silver plating has exactly the advantages I described above and the reason why it’s done. It doesn’t have to be of massive silver, the coating is sufficient to give you the advantages and the „core“ is cooper or brass due to costs. Silver plating additionally give you 25% higher current rating on eg busbars then a cooper only alone, tin is just corrosion protection.
On low quality one the silver color will be tin plated cooper, on high quality ones silver plated cooper.

[CaptainRivet]

Originally Posted by CaptainRivet

Regarding your 3p4S setup:
Did you measure the temps at the terminal of each of your 3 parallel cells under maximum load for 5min ?
If you installed like your drawing of cell setup the last and first cell gets much more load and wear faster then the rest of the pack. If you measure temp on each terminal I bet that first is about 3-5 degrees hotter then last cell. This shows the first contributes more then 2nd and 3rd cell of the pack even they are connected in parallel and load should be taken equally off which the temps proof is not the case.
To avoid this you need to take the load symmetrical off the pack, means you need to connect the main positive cable to middle cell 2 and main negative to middle cell 11. even better if you run 3 cables instead one, each connected to one of the 3 cells and on the other and with stacking lugs directly to the fuse and negative to shunt.
If your total load is below 50% of the single(!) cell capacity eg 140A for a 304AH cell (3x304Ah in parallel) then connecting to middle cell is enough, if above 50% then highly I recommend to run 3 cables each side, if max load is >100% then you must run 3 cables as you otherwise overload the first/last cell. This rule is for cells with 1 or 2C constant current capability cells like EVE, Lishen, Calb….Winston with 4C just double the percentages. That’s the main root cause you get runners in multi cells parallel battery pack.
Be aware that an inverter puts a 1.41x factor load on what your shunt shows you, means 100A DC shown from an inverter is actually 141A on/off multiple times a second load on the battery pack.

Forgot If you have even number of parallel cells eg 2 or 4 or 6P then connect each in the middle of the busbar between 2 cells so this way 2 cells are closest and share the load. 4S till 100% of single cell capacity directly in the middle of the 4p busbar, till 150% 2 cables each in the middle of cell 1 and 2 plus 3 and 4. above 200% connect 4 cables each to one terminal of parallel cells. This is called symmetrical offloading.

[GordMay]

Originally Posted by pbiJim

About 15 years ago, I worked with some programmable breakers that only wanted to see power in one direction. This was on a shore side application. I have not seen those breakers used in boats.

The mechanical breakers that I normally see used in boats will trip regardless of the direction of current flow. I know that from personal experience. I have not seen any issues arising from back feeding a marine DC breaker. If someone does know of a reason that should not be done, please fill me in...

A non-polarity DC circuit breaker does not have a specific polarity designation, meaning it can handle current flow in either direction, regardless of whether the current is positive or negative.

In contrast, a polarity-sensitive DC circuit breaker is designed to handle current flow in only one direction.

If polarized breakers are wired incorrectly, and turned off under load, the circuit breakers might not be able to extinguish the arc, and the circuit breaker will burn out.

Polarized DC circuit breakers use a small magnet, to direct the arc away from the contacts, and up into the arc de-ionization chamber [arc chute]. If the direction of current flow, through the unit is reversed, then the magnet directs the arc away from the arc chute, and into the mechanism of the unit, thus destroying it.

[geoffr]

Originally Posted by CaptainRivet

Regarding your 3p4S setup:
If you installed like your drawing of cell setup the last and first cell gets much more load and wear faster then the rest of the pack. If you measure temp on each terminal I bet that first is about 3-5 degrees hotter then last cell. This shows the first contributes more then 2nd and 3rd cell of the pack even they are connected in parallel and load should be taken equally off which the temps proof is not the case.

I don't understand the logic behind your argument above. From what you are suggesting this would apply within the "bank" connections as well.

The flow of electrons through any possible path in the bank will have the same resistance (assuming each battery is the same (they will be close enough)). Since the path resistance is the same, then the current through each battery will be the same, so the temperature should be the same. Where does the "extra" heat in your argument come from?

[CaptainRivet]

Originally Posted by geoffr

The flow of electrons through any possible path in the bank will have the same resistance (assuming each battery is the same (they will be close enough)). Since the path resistance is the same, then the current through each battery will be the same, so the temperature should be the same. Where does the "extra" heat in your argument come from?

thats the theory and half truth. and with already 21 till 23mohm they are not the sameanymore.



means how you and i did configure exactly the same like you (just 4P4S with all having 14mohm) you have 3 parallel cells with positive inside and opposite is the other pack with 3 negative inside and Cell 1 (positive) and 12(negative) battery is closest to the terminal, cell 2 and 11 already one cell width further away and so on. so distance and therfor resistance is not equal!
if you put then a high load on you will see that (room temp is 20 degrees) that cell 1 and 12 are at 28 Degrees, cell 2 and 11 at 26 degrees, cell 3 and 11 and 25 degrees. take an IR temp gun and check yourself.

So what actually happen is that the cell pair first gets or delivers the most current (that why its hotter) then the next pair a bit less and so on. Voila you have imbalnce...why because distance and therfor resistance is different to each cell and all current goes through cell 1 and 16 which can also overload this cell. if not overloaded cell 1 and 16 defintily wear faster creating more probblem down the road and you get the 130-150mv variance Jedi states.
typical pitfall of multiple parallel setup, i know and still went into same trap.

[geoffr]

Can you send a diagram of how it "should" be wired as I'm still not following. I haven't decided on which cells yet, so could be 4P,3P (I doubt 2P).

In the diagram below my understanding would be that the resistance between the Red+ input wire through C1,C2 or C3 would be the same (other than the difference between the batteries themselves) to the C4 +. Are you saying that is would not be?

My understanding is the following:
If the Resistance of the Bars are B for the distance between Cells (eg C1-C2):
Red+ -> C4+ = R(C1) + (B*3) ohms For path through C1
Red+ -> C4+ = R(C2) + (B + B*2) ohms For path through C2
Red+ -> C4+ = R(C3) + (B*2 + B) ohms For path through C3

If the resistance is the same, then the current is the same, then the heat is the same...