Thoughts about this setup?

SeRiusRod

Member
Joined
Jul 4, 2020
Messages
47
Surfing the net some time ago I stumbled on the following videos:

In the videos he describes the building of a powerwall entirely made with cheap chinese 18650x4 holders. I have those things around for cell scavenging and I know how they are so I discarded the idea and never look back. But lately, I'm continuously being told (in other communities) that this guy is a genius. I've tried to make people understand that those holders are not reliable, and to make things worse this guy is hanging four cells in parallel from one single slot.
But I'm repeatedly told that I'm wrong and this practice is gaining adepts in my country.

Perhaps I'm wrong. What do you people think about this powerwall construction?
 
Last edited:

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,512
If my count is close - this is a 14s82p battery. To me, even doing a 14s80p would be much easier to use 4x5 holders and spot-weld... as @hbpowerwall has on recent videos - its get's as close to "123..." as I 've seen for custom packs...

I don't see things as right or wrong at this level... its DIY which by definition is up the fancy of the person building the battery. Just because someone *insists* doesn't make it the best way for you! :)
 
Last edited:

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,418
Using these cell holders isn't wrong to do, just not recommended.
Reasons being:
* The metal tabs are nickel plated steel, and small and thin
* The metal tabs will over time will weaken loosing grip
* The metal tabs have higher resistance
* These holders, due to the above issues, should only be limited to a few amps for any length of time as you will begin to loose energy via heat resistance

If going with something like this, it would be good to go with, or replace with, beryllium copper tabs. They are much lower resistance, hold their strength better over time, and do not heat up as much due to higher current load.
You can usually find them on SMT cell mount devices more so than on TH (through hole) units. But you can also buy the tabs separately and install them on the holders.

The other thing you could do is solder a wire directly to the tab, instead of using the TH piece that stick out the bottom side. That way you would not be limited to the current load of a few amps, but could hit 10A or more per connection.

Also, going with beryllium copper the tabs are far less likely to corrode or rust.
 

SeRiusRod

Member
Joined
Jul 4, 2020
Messages
47
That way you would not be limited to the current load of a few amps, but could hit 10A or more per connection.
10A is not so much current for 4 cells?
I also thought about the increased resistance of a mechanical join, instead of a solid one.
But does directly soldering to the tab make sense, when a fuse will still be used there?

This is the worse example I found. Other people are doing larger, on size, powerwalls, with less cells per slot. Perhaps that's better.
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,418
Each cell slot can only handle a few amps. So if there are 4 cells in that slot, then there's the potential of doing 10A (even on laptop cells, tho not advisable for their health). And that little pin will have a hard time carrying that current.
Yes, a fuse is advisable. You can put the fuse anywhere, it doesn't have to be *on* the cells. Just somewhere near them.
Also note that cells that are connected via a nickel strip should be as identical as possible, in use, life span, capacity, IR, manufacturer, etc. The reason is because there are no fuses there.
 

SeRiusRod

Member
Joined
Jul 4, 2020
Messages
47
You can put the fuse anywhere, it doesn't have to be *on* the cells.
No, I meant that if there's a fuse, shouldn't it be "weaker" than the small tab?

I've seen some other guys using fuse wire instead of nickel strips on one side. I always use wire fusing on both sides, and I ignore if it's overkill and only needed on one side.

If you look at the 4packs markings in this this and other related videos, there are crazy capacity differences, like two amps.

Just enjoying the discussion 😉
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,418
If you put a fuse smaller than the tab, you'll be blowing your fuse constantly. Because, as I stated earlier, the small tab can only handle a few amps. If you put multiple cells in parallel in that slot, you'll most likely be pulling several amps from that slot (depending on the overall number of cells in parallel, obviously)
 

italianuser

Active member
Joined
Feb 25, 2020
Messages
130
I agree that those holders aren't the best choice, ...considering he has expensive pieces all around. I could be wrong but all the electrical parts don't actually look like a DIY to me, looks like the panels companies mount:unsure:

Those holders aren't the cheapest ones, they aren't with the spring and the wire that melts at 1Amp (I got them ones 🤪but I only use them for not-critical stuff). They cost double, or more than double, of the cheapos, they look like these ones:

https://www.banggood.com/it/4-Slots...arehouse=CN&rmmds=detail-top-buytogether-auto

or maybe this kind:

https://www.banggood.com/it/LantianRC-18650-Battery-Holder-Case-Box-for-6S-Battery-p-1602943.html
 
Last edited:

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,418
Oh yeah, if you get the actual legit Keystone quality of the holders, they up there in price. you can get them at Digikey and Mouser too.

Oh geesh, I hope they aren't the second ones, the ones with the PCB's attached. What a waste a money for large scale like that. Not to mention the increased possibility of overcurrent blowing traces.
1620009221784.png
 

SeRiusRod

Member
Joined
Jul 4, 2020
Messages
47
He adds links to the holders sale in the videos description.
I thought that the th tab was like mines. That is way wider, so that's why I was thinking that with less cells connected would do. But these are cheaper than mines.
If he was to use mines, he would had wasted 500e only in holders.
 

italianuser

Active member
Joined
Feb 25, 2020
Messages
130
True, there's the link. Ye, they are the slightly more expensive ones.

But, you know what's strange about this guy. I clicked on a couple of random videos of his and he just doesn't seem right.

I noted: the way he uses the screwdriver to attach the holders, he's using a really long screwdriver (no good, the ones with extensions) and doesn't seem very confident.
The way he welds the SC25-10 (or whatever model they are) caps with the flame, at the end of the welding he moves the flame away and after many seconds he goes with the tin on the SC connector again (???), that's weird.
The way he draws a line with the pencil on the wood board where the holders stay in a very unstable manner; as if he doesn't know how to do it (???).

Maybe it's just me, the result is so professional (those control panels are wow) but he doesn't seem to know how to use the tools. But, maybe it's only my imagination! LOL

Maybe he pays someone to do the work for him. And, yes, I do like the powerwall after all!
 

SeRiusRod

Member
Joined
Jul 4, 2020
Messages
47
True, there's the link. Ye, they are the slightly more expensive ones.

But, you know what's strange about this guy. I clicked on a couple of random videos of his and he just doesn't seem right.

I noted: the way he uses the screwdriver to attach the holders, he's using a really long screwdriver (no good, the ones with extensions) and doesn't seem very confident.
The way he welds the SC25-10 (or whatever model they are) caps with the flame, at the end of the welding he moves the flame away and after many seconds he goes with the tin on the SC connector again (???), that's weird.
The way he draws a line with the pencil on the wood board where the holders stay in a very unstable manner; as if he doesn't know how to do it (???).

Maybe it's just me, the result is so professional (those control panels are wow) but he doesn't seem to know how to use the tools. But, maybe it's only my imagination! LOL

Maybe he pays someone to do the work for him. And, yes, I do like the powerwall after all!
Lol, yes that's just what i was thinking.
Also I really don't like wood used in powerwalls. And never for the structure. But are there any other options for this setup?
 

italianuser

Active member
Joined
Feb 25, 2020
Messages
130
(snip)
Also I really don't like wood used in powerwalls. And never for the structure. But are there any other options for this setup?
@hbpowerwall setup is super professional, an open setup with packs in nice view. And @Stefanseiner replicated the same structure (what a hell of a blacksmith he is:cool:) made of big metal holders (at top and bottom of each pack, a sort of a big clip / holder). @OffGridInTheCity used metal closets like the ones you find at the gym or at school (very nice setup, too).

I will be choosing the structure in the next weeks for my three 14S20P modules. I think I'll go for something similar to @OffGridInTheCity, a closed space with shelves, but on a smaller scale because I have a smaller space on a balcony with about 30cm depth available (about 1 foot).
 

italianuser

Active member
Joined
Feb 25, 2020
Messages
130
I don't speak spanish but about @hbpowerwall 's pack I understand he's saying that it's a mess if you have to change a cell LOL.

Actually, I do like the idea of having an easy job of changing a cell. What I like less is having cells attached one to another considering they release heat when under load, I *really* would be concerned of heat transmission between those cells attached with tape:unsure:
 

Wolf

Well-known member
Joined
Sep 25, 2018
Messages
1,414
Actually, I do like the idea of having an easy job of changing a cell
The idea actually, is to test the cells well and thorough enough so you don't have to replace them.
If you do a proper test and qualify the cells to begin with there should be no need to have an "easy" replacement system.
Throwing together a powerwall with cells of unknown SOH is just a disaster waiting to happen easy replacement or not.
Wolf
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,512
The idea actually, is to test the cells well and thorough enough so you don't have to replace them.
If you do a proper test and qualify the cells to begin with there should be no need to have an "easy" replacement system.
Throwing together a powerwall with cells of unknown SOH is just a disaster waiting to happen easy replacement or not.
Wolf
And its a matter of scale. 20p is one thing, 100(s) of p is another.
 

SeRiusRod

Member
Joined
Jul 4, 2020
Messages
47
I don't speak spanish but about @hbpowerwall 's pack I understand he's saying that it's a mess if you have to change a cell LOL.
Yes pretty good, but I got surprised about the despective tone he's using. Specially taking into account that he freely grabbed his video.

I'm thinking that perhaps could be possible to 3d-print a holder ring for those 4 cells and modify the holders adding a notch in the center for the ring to enter. I had success with some 3d printed filament in some exhaust splash guards I made for my motorcycle. That could allow to use nickel strip with fuse. I understand that daisy-chain 4 cells with fuse wire is simply...
 
Top