A Brit in Sweden
- Oct 8, 2020
If 12 psi is the goal and a battery is 6" x 8" = 50 square inches, need 600 pounds.Do you think M4 threaded rod would be up the job, or is something bigger warranted?
I like compliance, more travel with relatively constant force.it got to 300lbs with a bit more than totally flattened lock washer
I am very curious to see if they actually expand at all under this pressure - and I plan to adjust the pressure once I get to a steady 3.4vI like compliance, more travel with relatively constant force.
I haven't worked with these batteries. But they are a sandwich of some materials, each having a particular TCE. They will experience a range of temperatures. If anything in the sandwich expands with heat, either the spring system takes up the travel and maintains steady pressure, or pressure increases and something inside tries to get extruded. Alternatively, under cold conditions, they shrink and the pack falls out. (or vice-versa, depending on relative TCE rate of steel rods. vs. battery sandwich)
View attachment 29216
How about doing just what you described, but with coil springs around each rod?
Suitably selected, they would be partially compressed when you reached the force desired.
One thing I really see in my yard is that PVC conduit expands, buckling into an "S" shape. Then contracts, pulling connections apart or breaking things. Much higher TCE than steel conduit and concrete wall it is mounted on.
If these are stiffer springs, and you have access to a drill press and bathroom scales... use the drill press to compress the springs a certain distance and read the load on bathroom scales?I found some springs in my garage, which might be suitable for this purpose. What would be the most effective way to test their strength and figure out what compression I need to apply to get to my 12psi (once I’ve done the sums to figure out what force is needed per spring)?
Gosh, I didn't even think about "the board on the left". You're 100% right... Springs on the rods is the plan then. I like that better anyway!
No, I have some carbon fiber cloth left over from other projects that I think would be cool to use on the ends. Mostly for cosmetics, but added strength also.
I wish we could all find that one thing we are great at ... and be able to utilize it for our work.I happen to have an ability to visualize things.
A Stanford PhD ME and professor would produce 3D drawings with Ansys, could generate exploded views.
I would look at his drawings and declare "You can't assemble those parts due to interference". "The Philips screw which holds that board to enclosure can't be installed due to order of assembly."
After he had enclosures fabricated, sure enough you could only put in two screws out of four even though they were Allen head.
I took it home, arc welded the seams, pulled out my trusty Milwaukee bandsaw and made new seams in a different place.
I then presented it to Mr. PhD and said, "Now it is practical."
PhD my @$$
One thing to keep in mind and has been said many times before in this thread and others is that generally speaking, any given spring length will correspond with a specific load. From what I understand, these cells want to expand 1-2mm from discharged to charged. Applying the 12 psi will change this expansion of course but do we know if they still grow or not when the load is applied? Any way, the point of this post is to simply the following:
If you are able to apply the proper amount of force to 1 discharged battery, when fully charged if it grows by 1mm then the load on the cell exerted by the spring will have increased. This will vary depending on the spring used of course but 1mm most likely wont have a significant increase in load on the cell. If you have 4 cells in a pack and they all grow by 1mm now the spring has been compressed by 4mm and if the pack was 8, now it has grown by 8mm (well over 1/4") This can be very significant unless you are using a rather long spring. Keeping the number of cells the springs are acting on as low as possible will provide the most predictable loads on the cells.
I think you may have missed what I was trying to say so Ill say it another way.The whole point of the springs is the recognition that the pack is going to expand and contract with SOC, temperature, etc. ..... but, by using springs, the pressure changes will be much less than if there is just fixed compression.
While 12 PSI is the sweet spot .... if we assume that pressure variations of +-3 PSI or so would be acceptable.
If 4 rods are used on a 56 Sq In cell .... Would be 168 lbs per rod.
At 9 PSI, the pressure per rod would be 126 lbs.
at 15 PSI, it would be 210 Lbs
While 3 PSI is kinda random ... I think most would agree it is acceptable and the range of pressure per rod in the above example could vary between 210 and 126 lbs.
So, there can be significant change in the pressure / rod / spring and still be in the effective range to protect cycle life of the cell.
Maybe we can get to the point that someone is actually able to measure the pressure change across a wide range of temperature and thru the full SOC.
I think you may have missed what I was trying to say so Ill say it another way.
If you put each individual cell in its own individual "clamp", springs will deliver extremely consistent pressure. Every cell you add after that reduces the consistency that a typical spring can provide. Thats it. Im not knocking springs. Im just saying that if you want the most out of your cells and if getting the most means keeping them as close to 12 psi as possible, than you should try to minimize the number of cells in each clamp as much as possible (if your using springs). If your happy with some degree of the "better than nothing" approach, then add as many cells as you are comfortable with. Im not trying to knock anything. just trying add value and a perspective I hadnt seen considered yet.
Absolutely. Using a long spring with a very low rate can substantially mitigate or almost eliminate any change in force over the distances we are talking about. Again, it was just a perspective I hadnt seen considered anywhere. Most of what I had read was just fixture + spring = good so was just trying to add a little value.Just use a longer spring with a lower k constant; same result and far easier