Are Tesla Style Fuses Enough?

Probably too late but David Smith? from the facebook group is using resettable? fuses


image_qvuunk.jpg
 
Who comes up with this nonsense?
Current is evenly split?
Is there an 18650 workers union deciding how much current each cell gets?
The current each cell gets is the inverse of its proportional internal resistance multiplied by the current from the charger.
It is NOT evenly split.
You have no business connecting huge quantities of cells in parallel if you do not understand this basic fact
 
DIYTESLAPOWERWALL said:
Who comes up with this nonsense?
Current is evenly split?
Is there an 18650 workers union deciding how much current each cell gets?
The current each cell gets is the inverse of its proportional internal resistance multiplied by the current from the charger.
It is NOT evenly split.
You have no business connecting huge quantities of cells in parallel if you do not understand this basic fact
Click to expand...

I=v/r

Its not just resistance dude, its voltage, so unless the cell is consuming loads of energy some how its getting the same current.

As the SOC changes, so does the R value and therefore the current, it 'equalises'
 
If they are in parallel then voltage becomes a constant (for that point in time) and therefore is not relevant.

As Mike found out the hard way recently, if you put 5A into 20 cells you do not get 0.25A per cell. Yes internal resistance changes with soc, which is what makes it so dangerous. If you mix capacities, then you get different R values. When you discharge to 0v as Mike did, you end up with a wide range of R values. If you try to recharge with 5A as he did, you get one or two cells that take all the current and overheat. If you don't learn the lesson, you kill cells one at a time until you are left with cells of equal capacity and similar R values. He killed 12 cells I believe and still didn't know why
 
DIYTESLAPOWERWALL said:
If they are in parallel then voltage becomes a constant (for that point in time) and therefore is not relevant.

As Mike found out the hard way recently, if you put 5A into 20 cells you do not get 0.25A per cell. Yes internal resistance changes with soc, which is what makes it so dangerous. If you mix capacities, then you get different R values. When you discharge to 0v as Mike did, you end up with a wide range of R values. If you try to recharge with 5A as he did, you get one or two cells that take all the current and overheat. If you don't learn the lesson, you kill cells one at a time until you are left with cells of equal capacity and similar R values. He killed 12 cells I believe and still didn't know why
Click to expand...

So then don't discharge your 18650s to 0v. That's a rather stupid move to begin with.
 
mike said:
DIYTESLAPOWERWALL said:
If they are in parallel then voltage becomes a constant (for that point in time) and therefore is not relevant.

As Mike found out the hard way recently, if you put 5A into 20 cells you do not get 0.25A per cell. Yes internal resistance changes with soc, which is what makes it so dangerous. If you mix capacities, then you get different R values. When you discharge to 0v as Mike did, you end up with a wide range of R values. If you try to recharge with 5A as he did, you get one or two cells that take all the current and overheat. If you don't learn the lesson, you kill cells one at a time until you are left with cells of equal capacity and similar R values. He killed 12 cells I believe and still didn't know why
Click to expand...

So then don't discharge your 18650s to 0v. That's a rather stupid move to begin with.
Click to expand...

Agreed, this is a non relevant point used to make a point that isnt relevant.

More cells in parallel = good, no two ways about it.

DIYTESLAPOWERWALL said:
If they are in parallel then voltage becomes a constant (for that point in time) and therefore is not relevant.

As Mike found out the hard way recently, if you put 5A into 20 cells you do not get 0.25A per cell. Yes internal resistance changes with soc, which is what makes it so dangerous. If you mix capacities, then you get different R values. When you discharge to 0v as Mike did, you end up with a wide range of R values. If you try to recharge with 5A as he did, you get one or two cells that take all the current and overheat. If you don't learn the lesson, you kill cells one at a time until you are left with cells of equal capacity and similar R values. He killed 12 cells I believe and still didn't know why
Click to expand...

Mike from DIY power wall in NZ ? if that is who your example is, then erm ... discussion over before you start dude, he uses the assertion methodology not the science and chemistry methodology.

I tried to help advise on bottom end balance and ended up with a lecture on why I should top end balance my offgrid packs ... i like what hes doing but dont respect some of the weird execution. execution is what he did, he executes packs with low voltage.

Pete did too, but at least he understands why and even went as far in his last video to say he would rather 0 a pack out then to over charge one.
 
Yeah I was watching the second of Mike's 0V pack videos when it dawned on me. Hang on he's charging from cells 0V using a 5A charger, oh dear.
From there onwards I was only watching to see big of a problem he would have.

Best precaution is not to run cells down to 0V in the first place, failing that recharging each cell individually at very low current and seeing what passes.

DIY TESLA POWER WALL: Please stop trolling, your latest video is junk as you don't even answer the questions yourself. 7 comments all hidden, yeah.
 
I am glad that we all are able to discuss here on the forum. This place is wonderful because we get to hear everyone's experiences and learn more about this. This to me isn't my full time job, it's something I rather enjoy working with outside of work. While it is necessary to explain that resistance most definitely plays a role in the current that is given to each cell, sometimes it's best to use simplicity when doing calculations, this is why I was describing it this way. Actually, something that I learned right here on this forum was from one of Paul Kennett's video, and that is - testing for voltages and capacity is important, but it is also important to group similar resistance. When you have grouped cells that are similar resistance and capacity, these types of simplifications are okay.

Explaning the importance or relevance of something is great, there is no need to be negative to those who are here to learn.

With that being said, I think it is important to note here that we need to constantly be questioning everything about what we learn here, and constantly ask questions. This is how we evolve and how we become better. For Science!!!
 
station240 said:
DIY TESLA POWER WALL: Please stop trolling, your latest video is junk as you don't even answer the questions yourself. 7 comments all hidden, yeah.
Click to expand...

Ditto

I unsubscribed from him a while back, far to much puesdo science - I suspect he's a recent graduate from the University of Utube.
 
cstanley said:
Actually, something that I learned right here on this forum was from one of Paul Kennett's video, and that is - testing for voltages and capacity is important
Click to expand...

Keep watching Paul, he is one of the few I would advocate. Knows his science and math, ahead of the curve.
He's meticulous and for a reason, doing it right first time most of the time.

Fuses are a great example of community FUD (Fear Uncertainty and Doubt)

I can put it simply, most lithium cells in the world are unfused for a reason, they dont need it.
They have a bms that just stops over and undercharge, they are in parallel with other cells.

Open a laptop battery to see this in effect.

You dont need to fuse, you just need to manage high and low end voltage.

For DIY TESLA POWER WALL, I=v/r again mate. ohms law is all you need to feel safe.

Resistance testing isnt necessary, you can also parallel cells of different capacities just for this reason, just dont get your series modules out of sync when it comes to capacity, ie: 14s make sure all the modules are similar capacity over the series. You can make them out of any cell types, 600-2400 if you like, ohms law means the low cap cells take less charge, high cap takes more charge. The all important is a matched capacity over the series so that energy in and energy out leaves all modules at the same SOC.

Just be sure to self discharge test, and do it the right way, flatten all the cells first - dont charge them and sit them like most do.

Once a cell is discharged, it will rise to nominal voltage (3.5-3.7), and then any self discharging cell will be readily apparent as it wastes energy and the voltage drops markedly as there is little energy left in the cell to maintain that nominal voltage. Charging the cells first gives false negatives, as a cell that self discharges 5% from 100% SOC might just look like all the rest, same cell will hit 1.5v from 3.5v @ 0% capacity SOC.

kind regards
 
I agree that Paul is miles ahead in his thinking.

3nerg3, You quote ohms law like you are the only one who understands it, yet you display a complete lack of understanding in your examples. Ohms law doesn't ensure even power distribution, it has the opposite effect. That's why protection circuits are required.
As I explained to Mike (long before you mentioned it), current does NOT flow mostly to the highest capacity cells. It flows to the cells with the lowest resistance. Therefore, resistance testing is super important. It should be done at the bottom end, because that's where the high current is, and that's where you get wide variations of resistance, leading to some cells taking all the current and overheating. Thus proving that ohms law works against parallel cells. Current at the top end is very small so ohms law is less of a problem.

The scientific papers I have read on this subject show a 40% reduction in cycle life when failing to match internal resistance. Thus again proving you don't understand ohms law.
 
DIYTESLAPOWERWALL said:
The scientific papers I have read on this subject show a 40% reduction in cycle life when failing to match internal resistance. Thus again proving you don't understand ohms law.
Click to expand...

What do you consider a good threshold for internal resistance and how are you proposing we match them? I am running 1680 cells in production now. I resistance tested every single one of them and threw out anything that was > 200 milliohms. Is that a good threshold? Most were 90-150 milliohms. Additionally, how would you handle matching? We are talking milliohms here. Would you make packs of 90-100, 101-110, etc? Also, if you test one and it shows 100 milliohms, you can test it 6 more times and get 6 different readings on it, all within 25 milliohms of each other. So how do you propose the resistance is tested so accurately that you can group cells together in a way that's both accurate and eliminates the problem you're describing above?
 
All I know is that pack manufacturers match internal resistance. How closely? I have no idea. I bet they have very good test equipment that is highly accurate and only requires a single test. They also have millions of cells to match with, so that must make it easier
 
DIYTESLAPOWERWALL said:
All I know is that pack manufacturers match internal resistance. How closely? I have no idea. I bet they have very good test equipment that is highly accurate and only requires a single test. They also have millions of cells to match with, so that must make it easier
Click to expand...

Oh I'm sure they sit there all day testing cherry picking cells.
 
Have you ever bought cells directly from the manufacturer?
I have. Twice they delayed because they were not happy with the matching

Sarcasm is only successful if you have the evidence on your side and the knowledge to present it logically. Otherwise you look like a fool
 
All of these cases have valid points. I think I speak for most of us when I say, we don't have the means or really the want to purchase expensive equipment to test all of our cells. I think we acknowledge and assume that this definitely can be done better with the right resources, and we accept the risk that's involved (what ever that risk may be).

I do however enjoy talking about it (both sides) because it's always good to keep stuff this stuff in mind.
 
DIYTESLAPOWERWALL said:
Have you ever bought cells directly from the manufacturer?
I have. Twice they delayed because they were not happy with the matching

Sarcasm is only successful if you have the evidence on your side and the knowledge to present it logically. Otherwise you look like a fool
Click to expand...

No I have not. What are your thoughts on resistance tolerances and how to best group cells together, in what ranges? I'm interested in learning.
 
mike said:
DIYTESLAPOWERWALL said:
Have you ever bought cells directly from the manufacturer?
I have. Twice they delayed because they were not happy with the matching

Sarcasm is only successful if you have the evidence on your side and the knowledge to present it logically. Otherwise you look like a fool
Click to expand...

No I have not. What are your thoughts on resistance tolerances and how to best group cells together, in what ranges? I'm interested in learning.
Click to expand...

He got his math backward in his last example, dont read into it too much.
Most manufacturers that delay shipping do so due to supply, not cell matching.

Hes just projecting his ideals onto that situation.

You dont need to match resistances, as that takes care of itself, a low capacity cell will rarely take more current unless it has something else wrong with it, where it tries to use the energy for something outside of useful chemistry.

The only time cells do interesting things is when they have a short, higher resistance leads to lower current flow. What is interesting is that higher resistance can occur due to higher SOC or thermal issues due to current being used for bad chemistry. Making the assumption that its the latter is safe, but usually wrong.

cstanley said:
I do however enjoy talking about it (both sides) because it's always good to keep stuff this stuff in mind.
Click to expand...

BUT, sometimes its like saying that we should talk about theology right alone side science, as its discussing both sides.
I disagree with that sentiment.

Especially when ppl directly harm themselves with sudo types of thinking based on other people's assertions rather than fact.

Take a lead acid battery, as it gets older, the resistance naturally increases at all stages of the SOC

What does this mean, it charges 'faster' r values increase, current decreases, we have all seen this affect, the battery appears charged, gets to high SOC voltage, and current of charger reduces.

Lithium is no different. Its not magically changing the electrons into monsters in the bottle waiting to explode.

kind regards
 
pid=\ said:

cstanley said:
I do however enjoy talking about it (both sides) because it's always good to keep stuff this stuff in mind.
Click to expand...

BUT, sometimes its like saying that we should talk about theology right alone side science, as its discussing both sides.
I disagree with that sentiment.

Especially when ppl directly harm themselves with sudo types of thinking based on other people's assertions rather than fact.

Take a lead acid battery, as it gets older, the resistance naturally increases at all stages of the SOC

What does this mean, it charges 'faster' r values increase, current decreases, we have all seen this affect, the battery appears charged, gets to high SOC voltage, and current of charger reduces.

Lithium is no different. Its not magically changing the electrons into monsters in the bottle waiting to explode.

kind regards
Click to expand...


I just don't think people will harm themselves by deciding to believe whomever they choose for whatever their reasons are in this particular argument. Grouping like resistance with cells will do no harm. Whether or not it matters is what you are arguing. I haven't commented on that. I am just saying, if it were a topic of safety then yes, but just because some people choose to fuse each cell, or match resistance isn't going to cause harm. If it makes whomever feel safer, then who cares?
 
You must log in or register to reply here.
Back
Top