How do you know when you’ve pierced a terminal bottom on an EVE 280Ah cell?

Maast

Compulsive Tinkerer
Joined
Oct 31, 2019
Messages
770
Location
Washington State
If it was electrolyte he poured out of the case in the video, then one or more of the pouches leaked.
I looked at the next video in the series and he took the innards out and it wasnt a pouch at all. It was two REALLY long flattened jellyrolls with aluminum and copper tabs spaced along it that were gathered up to make the tabs. The edges weren't sealed just open to the case so what he poured out was the actual operating electrolyte. He found on one of the jellyrolls that the copper anode (cathode?) wasnt covered in the LFP material and evidently shorted to something.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
The terminals consist of an aluminum block which is ultrasonically welded to the current collectors (aluminum or copper) foil inside the cells. If you go deep enough on the anode you should see copper from the weld area. On the cathode its aluminum all the way through.

From the video above it should be similar on the larger cells.
View attachment 36901

Some cells use a different fastening method for the current collector foils, but these high density aluminum cased prismatics have limited space.

I have not had an LFP cell apart, but reports are that the electrolyte has a strong chemical odor. Its a clear-ish liquid with low viscosity. If you penetrated the cell, you should be able to invert it and see/smell the liquid.

I have not seen any prismatics with separate pouches. While there are various layouts for the sandwich (rolls, flip flop, etc), they all share a common electrolyte.
So if I’m interpreting that picture correctly, the cathode consists of a folded piece of thickish aluminum foil, ultrasonically welded on the top to a round aluminum terminal (which then got tapped in our case) and connected to multiple this aluminum leads tapping out of the jelly roll on the bottom (also ultrasonically welded, or whatever).

That yellow layer looks like Kapton or something similar placed between the 2 folded layers of the cathode, probably to protect the thin aluminum leads heading to the jelly roll.

If you do punch through the bottom of the terminal, first you’ve got the thickish layer of foil, then the kapton layer to go through before you’re risking any damage to the thin aluminum leads.

Without a sealant, the greatest risk from puncturing a terminal is leaking electrolyte but in my case, since if any puncture happened, it happened when the terminal was being sealed in grubscrew+JB Weld, leakage should not be a concern.

If either some JB Weld or the stainless surface of the grubscrew became exposed to electrolyte, I suppose that could screw up the cells chemistry, but since my penetration was limited to 125 microns (1/8th turn), I’m hoping the bottom layer of the terminal was at least that thick so no foreign material got into the cell cavity...

Note that the entire cell current flows through the cross section of that folded thickish piece of aluminum foil, so in the case of the 280Ah cells charging/discharging at 1C, it’s got to be at least the equivalent of 2/0AWG in copper or 67.4mm^2 or 164% that cross section meaning at least 110.5mm^2 in aluminum.

Even if we assume half the cell width or 80mm devoted to a wide folded electrode, it’s got to be over 1mm thick...
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
Didn’t someone here destroy a cell (by shorting it or something)?

We should send them a link to that video and see if we can coax them into breakdown analysis of a 280Ah cell (or coax them into sending that cell to someone willing to make that effort in case they are not interested).
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
Small update.

First, I found a way to measure the amount of torque I’m applying through my little 3” Allen Head wrench - about 20 inch-lbs.

Do if anyone has any insight as to whether that amount of torque is sufficient to pierce the terminal bottom with a bottomed bolt or grubscrew, I’m interested.

And second, I completed another pair of terminals gluing in SA grubscrews with JB Weld today, and when the process goes smoothly, the final turn goes very slowly (1/16th turn every 1-2 minutes) and additional JB Weld oozes out of the top all the way to bottoming.

I believe that terminal I did yesterday was not oozing during that high-pressure final turn. So that could mean that 20 inch-lbs was sufficient torque to burst through the bottom of the terminal and that last 1-1/8th turn was actually driving ~1.125mm of JB Weld into the void within the folded Aluminum Terminal (above the Kapton and below the folded aluminum under the terminal.

At this point, it is what it is, but knowing that for applying JB Weld successfully to grubscrews in terminals means being able to apply 20 inch-lbs of torque translates to liquid pressure against the bottom of the terminal, I’d sure like to understand whether that amount of pressure is ‘safe’ for these soft aluminum terminals or not...

If a bolt is bottomed in a terminal and torqued until something gives, is it the thread that gives or the terminal bottom that gives? Inquiring minds want to know..,
 

Luthj

Photon Sorcerer
Joined
Jul 1, 2020
Messages
945
Its totally dependent on how much material is at the bottom of the hole. If there is more than 4 threads worth of thickness, the threads will shear, if there is less, it could break through.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
Its totally dependent on how much material is at the bottom of the hole. If there is more than 4 threads worth of thickness, the threads will shear, if there is less, it could break through.
Sorry, not understanding.

You are saying that if there are more threads engaged that will make the total strength of the threads weaker (more likely to sheer versus strength of terminal bottom)???

I would have thought the opposite: when fewer threads engaged threads are more likely to sheer versus (fixed) strength of terminal bottom (especially since, by definition, fewer threads tapped likely means a thicker terminal bottom).

Taken to the limit, if there is only a single thread tapped, I’m pretty certain a torqued bolt would sheer off that thread before it punches through the (very thick) terminal bottom.

[EDIT: never mind, now I see how I misread, So you are saying a mm at the bottom is about as strong as a single (1mm) thread.

So less solid thickness below the last thread than total thickness of the threads could result in the terminal bottom getting pierced.

I would have thought that 1 mm of solid aluminum at the bottom would be stronger than 1 mm of thread, especially when you consider than stainless-rotating-on-aluminum contributes some cutting action that doesn’t happen at the bottom (at least not when considering piercing force of pressurized JB Weld).

But at least it’s a reference point - thanks.

Do we know the loyal length of the terminals in these EVE 280 Ah cells?]
 
Last edited:

Luthj

Photon Sorcerer
Joined
Jul 1, 2020
Messages
945
I am saying that without knowing how thick the terminal material is at the bottom of the hole, there is no way to determine the safe pressure.

If the bottom is thicker than 4 times the thread pitch, then its likely the threads will fail before the bottom does.

The threads won't fail unless you exceed the torque limit we have previously discussed.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
I am saying that without knowing how thick the terminal material is at the bottom of the hole, there is no way to determine the safe pressure.

If the bottom is thicker than 4 times the thread pitch, then its likely the threads will fail before the bottom does.

The threads won't fail unless you exceed the torque limit we have previously discussed.
I’m pretty sure that the torque needed to cause thread failure is directly proportional to how much thread in engaged.

I only had ~4 threads engaged and all but one handled 35 inch-lbs without issue.

if I had used double bus-bars and reduced the engaged threads to ~2-1/2, I’m pretty sure 35 inch-lbs would have stripped...

[PS: I edited my prior post after I understood how I had misread your earlier post, but our typing crossed in the ether...]
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
Yeah, typically 4 threads maxes the pull out strength.
Meaning at less than 4 threads you may have difficulty handling 35 inch-lbs but the benefit of a 5th thread is marginal?

Has anyone seen a cross-section of these aluminum terminals? Do we know how thick they are?
 

Luthj

Photon Sorcerer
Joined
Jul 1, 2020
Messages
945
For aluminum I think coarser threads would be better. I have not seen anyone cut a cell open. If someone wants to mail me one I'll do the deed.
 

Johan

The law of conservation of misery.
Joined
Oct 2, 2019
Messages
58
Location
Sacramento, CA, USA
Found this on youtube
Below are some approximate BLS 105Ah cell terminal dimensions derived from the same J5GURU Youtube channel.
Source: https://youtu.be/6d89tgUMEbI?t=280 (4:40min)

For M6 bolts, assuming a coarse thread, the pitch is 1mm, so a max of 5 3/4 turns would correspond to 5.75mm. It is tempting to apply theoretical analyses using e.g. a conservative 2mm aluminium bottom thickness being "sheared" out by the grub screws assuming zero thread friction due to epoxy "lubrication", infinite-strength-threads, but it is too darn complex to do on the back of an envelope. I want to see repeatable experimental results. Anyone? :)

For some more prismatic cell "teardowns": see here (not directly relevant for the problem at hand, but related).

bluey cell terminal.png
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
Below are some approximate BLS 105Ah cell terminal dimensions derived from the same J5GURU Youtube channel.
Source: https://youtu.be/6d89tgUMEbI?t=280 (4:40min)

For M6 bolts, assuming a coarse thread, the pitch is 1mm, so a max of 5 3/4 turns would correspond to 5.75mm. It is tempting to apply theoretical analyses using e.g. a conservative 2mm aluminium bottom thickness being "sheared" out by the grub screws assuming zero thread friction due to epoxy "lubrication", infinite-strength-threads, but it is too darn complex to do on the back of an envelope. I want to see repeatable experimental results. Anyone? :)

For some more prismatic cell "teardowns": see here (not directly relevant for the problem at hand, but related).

View attachment 37094
If the thinner cells have a 7.6mm thick round aluminum terminal welded onto a 2mm thick aluminum (or copper) plate, you’re think the larger cells carrying more current would be at least that thick.

I only went 1/8th turn beyond (what I thought was) bottom, but since I didn’t note when I stopped getting any oozing out of the top, hard for me to say whether I may have threaded close to 2 full turns after piercing the bottom (but I doubt it).

At this point, that cell is what it is, but the lessons for me are:

-use JB Weld for gluing terminals within the first hour (despite 4 hour set time). For me, this means only 1 battery / 2 terminals per batch.

-use enough JB Weld to get oozing out of the top (I use a matchstick-head sized dollop at the bottom of the terminal and another matchstick-head sized dollop on the cupped end of the grubscrew.

-thread grubscrew until resistance increases significantly, by which time you should start to see oozing out of the top thread within the next half-turn or so

-the final couple turns generally go very slowly - 1/8 turn every 1/2 minutes - but when in this phase, monitor torque required as well as continued oozing and if torque required reduces (it gets easier to turn) and there are no signs of further oozing, the terminal bottom may be in the process of giving out and it’s time to quit turning and call it a day.
 

codfish

Solar Enthusiast
Joined
Feb 27, 2020
Messages
72
While JB is weld is a great product, how about penetrating threadlocker instead.
Finger tighten, and add a drop of this. I've never used it, but it sure sounds simpler.

 

blutow

Solar Addict
Joined
Dec 20, 2020
Messages
321
While JB is weld is a great product, how about penetrating threadlocker instead.
Finger tighten, and add a drop of this. I've never used it, but it sure sounds simpler.

My take on it:

-thread locker will basically glue the threads in place and might prevent shear forces by limiting movement.

-JB weld may offer some additional structural strength by filling loose spaces between the thread interface

A bad analogy probably:

-thread locker is like spraying your foot with adhesive before you shove it in a boot. It dries and your foot and leg kind of sticks to the boot to prevent twisting and pull out

-JB weld is like putting some concrete in the bottom of the shoe and letting is squeeze up through your toes and around your ankle when you shove your foot in. Pretty solid once it sets. The danger is that you may not be able to get your foot all the way in because of the hydraulic effect

I think either will help prevent pull out, but JB weld is likely the stronger solution if you can make it work
 

codfish

Solar Enthusiast
Joined
Feb 27, 2020
Messages
72
My take on it:

-thread locker will basically glue the threads in place and might prevent shear forces by limiting movement.

-JB weld may offer some additional structural strength by filling loose spaces between the thread interface

A bad analogy probably:

-thread locker is like spraying your foot with adhesive before you shove it in a boot. It dries and your foot and leg kind of sticks to the boot to prevent twisting and pull out

-JB weld is like putting some concrete in the bottom of the shoe and letting is squeeze up through your toes and around your ankle when you shove your foot in. Pretty solid once it sets. The danger is that you may not be able to get your foot all the way in because of the hydraulic effect

I think either will help prevent pull out, but JB weld is likely the stronger solution if you can make it work
I agree in principle, but if a threadlocker is "permanent" and requires heating to 500F to release, then it seems like to must be doing a pretty good job of filling all those little spaces and bonding to each surface. I wonder what a dried drop of it feels like compared to JB Weld.

Seems like a bench top experiment would answer the question. piece of aluminum, drill and tap a few holes. Each one gets either JB Weld, red, green, blue threadlocker and a grub screw. wait till morning and evaluate.
 

Luthj

Photon Sorcerer
Joined
Jul 1, 2020
Messages
945
For the shear thicknesses we are talking about, a permanent threadlocker will probably perform similarly to a epoxy. At least for threads which aren't damaged.
 

fafrd

Photon Sorcerer
Joined
Aug 11, 2020
Messages
2,192
For the shear thicknesses we are talking about, a permanent threadlocker will probably perform similarly to a epoxy. At least for threads which aren't damaged.
That’s my guess as well.

I’m pretty confident JB Weld will recover my partially-stripped terminal more effectively than Loctite Red would have.

And I’m reasonably confident JB Weld will also do a superior job glueing grubscrews into my worn threads (not stripped, but upper 4 threads clearly worn from ~4 insertions/removals of stainless bolts, likely due to galling).

But from everything I’ve read about Loctite Red and experienced with JB Weld, if I ever find myself glueing grubscrews into pristine / unused aluminum threads, I believe Loctite Red should perform as well as JB Weld and sounds waaay easier to install, so I’d go that route in a heartbeat...
 

blutow

Solar Addict
Joined
Dec 20, 2020
Messages
321
I have not personally messed with these threads/cells yet, but my prior experience with aluminum is that it can be prone to sloppy machining resulting is loose tolerance. Thread locker is good at preventing a bolt from turning in the hole, but most flavors are not good at filling gaps and providing additional structure where needed. When my cells arrive, I plan to assess the tolerance and make a call. My preference would be thread locker, but I'll consider an epoxy approach if things are really loose. My batteries are going in a van, so I don't want any questions on torque or reliabilty
 
Top