NEW 40138 cell assembly blocks from PSI

[bobmcree]

after having some problems with leaking cells in my original packaging, i was very hopeful when i heard about the new PSI built assembly parts available from andyh through his website http://www.rechargeablelithiumpower.com

the blocks have turned out to be everything i hoped for. there are a photos of the blocks on his website but i thought i would post a few more detailed ones here.

the blocks snap together into any configuration desired and space the cells for cooling and protection from cells shorting together. the seats in the blocks are about .25 mm larger than the o.d. of my cells, so they slop around a bit. there may be manufacturing tolerances that make the loose fit necessary; and i would not want to rely just on the fit of the blocks to the cells to hold the pack together. i drilled out some holes in the four corners of the block assembly and cut some 8-32 threaded shafting to length, then used locking nuts to hold the pack together with this additional frame. this makes the pack very solid.View attachment 5View attachment assembled 6 cell pack.jpg

 

[newjeff]

Glad you are assembling these Bob! The threaded rods are a great idea. Do the copper connecting strips help to stop movement of the cells along with the rod clamps? You have the threaded rods at the corners only do the plastic end caps bend in the middle or are they rigid? With more cells would you use more rods perhaps in the center of the pack?

 

[bobmcree]

Glad you are assembling these Bob! The threaded rods are a great idea. Do the copper connecting strips help to stop movement of the cells along with the rod clamps? You have the threaded rods at the corners only do the plastic end caps bend in the middle or are they rigid? With more cells would you use more rods perhaps in the center of the pack?

the blocks are very stiff, so there is virtually no bowing at all in my pack; i would use 3 rods on each side for a pack twice as wide.

i used an internal tooth lockwasher next to the cell nut, then the brass connecting strap, then an external tooth lockwasher, followed by a flat washer, then the nut which is of course loctited as is the nut closer to the cell.

when my whole pack is clamped together, the friction of the end plates (block assemblies) against the cells keeps the cells from moving even the .25mm that they could if things were more loosely connected. each of the brass current-carrying straps connects the studs together on a pair of cells so there is no relative motion at all. the nuts on the threaded rods are nylon locking type so they do not come loose but do not clamp against the block assembly. this makes the whole assembly quite stiff, but with enough give that small relative motions might still be permitted rather than the same force possibly breaking something in a stiffer system.

i think that if you made the current carrying straps from flexible braid with enough extra length to permit relative cell motion without electrode stress, it could reduce the possibility of the cells leaking. i don't know if it is necessary or not, so far most of the cells i have observed to leak significantly did so only after over-discharge below 1V. the exceptions to this were cells to which wires were attached that were permitted to cause some rotational torque on the studs. i think if you make sure the nuts on the cells are tight and use loctite on them that it may not be necessary to use flexible interconnects. one problem is of course that it is difficult to tell if a cell has leaked once you assemble it into a pack with end plates covering the cell ends. i suppose one could just drill an extra hole or two in each block that would permit visual access to the paper end caps on the cells. even fairly small amounts of leaking electrolyte can be easily seen as fading and wrinkling of the paper, but it is difficult to see enough through the slits provided in the blocks.

the larger issue of course is how detrimental is it to lose a little of the lithium perchlorate, other than it being corrosive? testing so far indicates it is not a big deal if the amount of lost liquid is small. the effect on cell longevity of losing a bit of electrolyte is unknown, at least by me so far it seems that cells that were not otherwise abused and show some slight leakage still test the same as the cells that do not show any sign of leakage. i also have cells that show no sign of leakage at all but will not take a charge. it seems the two issues are not necessarily related.

 

[newjeff]

My understanding of the leakage issues was the wrong torque used on the threaded post's first nut, not the connecting or second nut. I read somewhere there was a torque value to be used. If this first nut was too tight or too loose then leakage occurred. When connecting and tightening the BRASS(tks) connectors was there any movement on the initial nut? Is there space to insert a wrench or tool to stop any tightening or loosening of this initial or retention nut? Is this even a concern or not?

 

[bobmcree]

My understanding of the leakage issues was the wrong torque used on the threaded post's first nut, not the connecting or second nut. I read somewhere there was a torque value to be used. If this first nut was too tight or too loose then leakage occurred. When connecting and tightening the BRASS(tks) connectors was there any movement on the initial nut? Is there space to insert a wrench or tool to stop any tightening or loosening of this initial or retention nut? Is this even a concern or not?

it is of course the first nut closest to the cell that retains the pressure on the nylon or teflon washer that seals the port in the cells. at least on the first packs i worked with there was no loc-tite on these nuts, and we knew nothing about the possibility the cells might leak if we tried to actually attach wires to the nice heavy lugs on the cells. imagine that, seeing a 1/4" threaded stud on a battery and imagining it could be ok to attach wires to it? i did not swing these cells around in a circle over my head by the wires. i was just not careful enough, apparently. in any case i have pointed out several times that only the cells that were destroyed by a malfunctioning vms leaked excessively, and that this brought my attention to the fact that several other cells that had not been so abused had leaked slightly but still seemed to be fine, except for a couple that seemed to have been damaged in the process.

from my experience with this chemistry, the cells do not normally vent unless they are severely abused or the nuts on the cells come loose. i believe the washer/nut combination on these cells is adequate to prevent the cell bursting in event pressure does build up, and that just keeping the nut tightened properly should prevent other leakage. i still have not heard anything official from the folks selling these cells about the recommended torque on the nuts, but i think just tightening them snug and using loctite should be enough to keep them from leaking under most conditions.

it is not easy to use a pair of wrenches to tighten the nuts on the cells once the new blocks are installed. i think with the combination of inner and outer tooth washers and a flat washer on each terminal i can get them tight enough to make good contact, and loc-tite will keep them that way. this assembly system will be great in the event a cell needs to be replaced, or even when all the cells need to be replaced with the new 14 ah type which i understand has actually been seen and tested and is a real product available ???

 

[newjeff]

Here is the torque document I was talking about Bob.

 

[Ypedal]

I have reservations about the loctite .. would it not get in the way of a good eletrical contact ?

My plan was to apply a tiny bead of hot melt glue ( I have 5 grades of the stuff.. the industrial grade is fierce and resiliant ) but it can be removed if need be..

There are certainly many ways to go about this.. but the loctite seems less than ideal to me..

 

[Patriot]

I've been reading this thread and others pertaining to issues with the PSI cells, including some issues with certain dealers. I have no opinion on those issues, so I'll stay away from that.

Anyway, I have always been quite impressed with the known performance of the PSI chemistry.

However, I have always been unimpressed by the physical engineering of them. The posts being as skinny, easily stripped, broken, or causing damage to the cells themselves I think is their main drawback. That's one of the reasons I sacrificed a little performance, but went with more durable design engineering in the newer Headway 38120S cells. Granted, they don't give the juice like a PSI, but for ebiking, I think they are more than adequate. Even for a motorcycle, I think they would be more than sufficient. For real cars? Well, maybe not good enough. PSI definitely has the quality power production stats in that area.

However, if I may suggest, and if anyone from PSI is reading this..

If they can adopt a more durably engineered cell, like Headway, yet with the existing power available in the current PSI, then I think they'll have a serious winner. Needing all the extra plastic blocks is nice for custom configuration, but it does get a bit frustrating when one needs all the extra parts to make a custom pack. The blocks are a very good idea though for the cells being as delicate as they are. Also they will maintain a garaunteed spacing between them, to prevent excessive heat build up, which is also a benny. But for most small applicartions, I think many prefer being able to just strap together a small pack. without needing all the extra stuff.

Either way, Bob, I hope these new blocks really work out for you. I'd hate to be on the recieving end of your disconcertion.

 

[MitchJi]

Hi,

Anyway, I have always been quite impressed with the known performance of the PSI chemistry.

However, I have always been unimpressed by the physical engineering of them. The posts being as skinny, easily stripped, broken, or causing damage to the cells themselves I think is their main drawback. That's one of the reasons I sacrificed a little performance, but went with more durable design engineering in the newer Headway 38120S cells. Granted, they don't give the juice like a PSI, but for ebiking, I think they are more than adequate. Even for a motorcycle, I think they would be more than sufficient. For real cars? Well, maybe not good enough. PSI definitely has the quality power production stats in that area.

However, if I may suggest, and if anyone from PSI is reading this..

If they can adopt a more durably engineered cell, like Headway, yet with the existing power available in the current PSI, then I think they'll have a serious winner. Needing all the extra plastic blocks is nice for custom configuration, but it does get a bit frustrating when one needs all the extra parts to make a custom pack. The blocks are a very good idea though for the cells being as delicate as they are. Also they will maintain a guaranteed spacing between them, to prevent excessive heat build up, which is also a benny. But for most small applicartions, I think many prefer being able to just strap together a small pack. without needing all the extra stuff.

I posted some similar comments on another thread and was basically told there are no problems (not by Bob). It seems obvious that PSI could and should improve in the cells in this area.

I think the blocks are a great idea even if the didn't help resolve the weak/leaking stud issues. They make it very easy to fasten together the round cells.

I think the spacing is also good if the studs are fastened together with solid strips. Otherwise expansion caused by temperature changes would put stress on the studs.

 

[thomas]

Currently building packs for my motorcycle heres how I have mounted a 6mm bolt on the blocks for 16mm2 cable connections. The cell blocks slide very tightly into 2mm alu alloy boxes. I need to do a cover with cable reliefs and some spacers that will hold the blocks down.

 

[bobmcree]

Here is the torque document I was talking about Bob.

thanks jeff. the date indicates that this document became available about a month before we purchased our packs from the US distributor. since this is a PSI document, and they made the cells we purchased, it would have been nice if it did not take well over a year for the document to be brought to our attention. A lot of headaches might have been avoided.

i still doubt that proper torque on the nuts will prevent the cells from leaking if the vms permits them to discharge to below 1V. this of course should never be permitted to happen, but of course anything that CAN happen, WILL happen. It looks to me like the nut/washer system is probably adequate to keep the liquid in the cells under all normal conditions, and that they do provide an adequate mechanism to keep the cells from bursting if pressure develops when the cells are used outside the design parameters or in the event of some unforseen internal failure.

we all live and learn. the slots in the psi blocks indicate that they are providing for the possibility of the cells leaking, and the blocks do serve to protect the studs and minimize stress on them. i think the plates gary made for his packs should serve essentially the same purpose.

the cost of the blocks and hardware to assemble a 12 cell pack was about the cost of a single cell. it seems like money well spent.

 

[bobmcree]

I have reservations about the loctite .. would it not get in the way of a good eletrical contact ?

My plan was to apply a tiny bead of hot melt glue ( I have 5 grades of the stuff.. the industrial grade is fierce and resiliant ) but it can be removed if need be..

There are certainly many ways to go about this.. but the loctite seems less than ideal to me..

i am relying on internal and external toothed washers to bite into the brass nut and the interconnect straps. i am only adding the loctite after the nut has been tightened. i have not observed in the past that loctite can work its way into an existing electrical connection and degrade it, but there might be a case for not using loctite on the first nut on the cells. i think the lockwashers make the argument against loctite on the outermost nut less convincing, and since it will keep the assembly tight a case could be made that the interior nut does not need loctite.

in the case of the cathode terminal, it is hollow aluminum and i have seen significant corrosion from a very small amount of leakage of electrolyte, so the loctite might serve to protect the connection from this corrosion rather than to degrade it. it might even be a good idea to coat the terminals with dielectric grease to keep corrosion to a minimum. this grease is non conductive, it just keeps corrosive elements away from electrical connections.

 

[AndyH]

Bob,

I'm very happy that the assembly blocks are doing what you want them to do!

I have a customer-returned cell that should have a 'mildly impaired seal situation' - how can I make this leak? I've cycled the cell twice on the bench to 2.1V LVC at 1C with no liquid or smell. Now that my CBA amplifier is back, I can take the cell to 12C if necessary. Thoughts or suggestions?

Thank you!
Andy

View attachment LEAK_TEST2.jpg

 

[bobmcree]

Bob,

I'm very happy that the assembly blocks are doing what you want them to do!

I have a customer-returned cell that should have a 'mildly impaired seal situation' - how can I make this leak? I've cycled the cell twice on the bench to 2.1V LVC at 1C with no liquid or smell. Now that my CBA amplifier is back, I can take the cell to 12C if necessary. Thoughts or suggestions?

Thank you!
Andy

hi andy, the brass nut holds the nylon washer below it in contact with the vent port which seems to be just a space around the electrode. the fact that the stud has snapped off does not necessarily mean the cell will leak. As long as the brass nut is kept tight and the stud is not stressed by a connecting wire or other forces i have not seen cells leak. It is now clear that if we had simply been given the available information from the manufacturer at the time we bought the cells that most of the leakage could have been prevented. I would recommend a micro torque wrench to check the tightness of the nuts on the cells and to tighten any additional nuts used for interconnection. Your customer is fortunate to have dealt with you; the studs we broke off were not covered by warranty.

The exception seems to be a batch of cells I have that were discharged down to below 0.5v when the original factory vms failed. This was also the very first pack sold so it might be different from some others. Many of these cells were found to be hissing and leaking just sitting there dead on the shelf, and when i slightly loosened the nut on one that was not leaking, it released quite a bit of gas and liquid. When these batteries are fully charged there is virtually no free lithium present at all, but as they discharge this is no longer true. I believe when the cells are discharged this deeply that the free lithium ions are reacting and i think it is probably oxygen gas that is being liberated, but i am not sure about this. The leaking electrolyte is very corrosive to the aluminum cathode studs.

Of course the cells should never be discharged this deeply and hopefully this particular problem will not occur. About half of this batch of cells will no longer accept a charge at all, but the other half still seem to be fine. Even with significant leakage the performance of the cells at least in the short term seems fine.

Your photo illustrates a problem that several users have had with the cells, and i believe it is a major problem with this design. Of course no system is perfect, and if proper precautions are taken the problem can be minimized. In virtually any system i have used in the past where a threaded stud is used to attach a wire to a power source, the threaded stud and nut is just used to hold the lug down against a metal base that makes the real connection. In these cells it is the threaded stud itself that IS the connection, and it can be seen in your photo that the cathode stud is hollow aluminum with another threaded interior section. That thin hollow aluminum threaded tube is carrying all the current. This just does not seem optimal to me. Leakage of a small amount of the lithium perchlorate electrolyte may not damage the cell, but might corrode and weaken these studs and cause early failure of cells that otherwise might have lasted much longer. There is some anti-oxidizing liquid available for power distribution connections that use aluminum wiring; it might help to put some of this on the cathode studs.

 

[AndyH]

hi andy, the brass nut holds the nylon washer below it in contact with the vent port which seems to be just a space around the electrode. the fact that the stud has snapped off does not necessarily mean the cell will leak. As long as the brass nut is kept tight and the stud is not stressed by a connecting wire or other forces i have not seen cells leak. It is now clear that if we had simply been given the available information from the manufacturer at the time we bought the cells that most of the leakage could have been prevented. I would recommend a micro torque wrench to check the tightness of the nuts on the cells and to tighten any additional nuts used for interconnection. Your customer is fortunate to have dealt with you; the studs we broke off were not covered by warranty.

Hi Bob,

Right - broken studs are not covered by PSI.

I've been cycling this cell without the nut - exactly as seen in the picture. I'll keep cycling it. I want to see if it will leak with NO nut on the terminal at up to the 10C constant rate with a 2.1V LVC. Then I'll disable the LVC cutoff and take the cell down to nothing.

This inexpensive torque driver from Harbor Freight is doing a great job - it was about the price of one cell. http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=65397 Besides, using a torque driver made in Taiwan on cells made it Taiwan just seems like the right thing to do.

Except for material, these terminals aren't much different than disk brake bleeder screws. I've destroyed enough of these and other small bolts over the years that I've grown a great respect for torque specs and keeping torque wrenches calibrated. Of the 250 or so cells that have moved thru me in the past two years, I've only replaced two - and both were 'operator error' with what was thought to be a 'properly calibrated hand' but found to be a bit off. I no longer want to waste my 'creative time' trying to figure out how to remove what's left of a broken bolt - especially when I need to use the vehicle I'm working on to get me to the store to get replacement bolts.

I like your idea of using an anti-ox paste on the terminals!

One point on torque - torque specs are given for clean, dry, un-lubricated fasteners. What what we really want when we tighten the nut on these cells is the proper clamping force. Lubrication - including anti-seize and locktite - lowers friction in the fastener. Lowered friction results in higher clamping forces with the same torque. This can lead to over-stressed, stretched, or broken fasteners. Any lubricant forces calculating a reduced torque value - usually 30% lower than dry. For our 6.9 in lb/.78nm terminals, that suggests something in the area of 4.8 in lb/.55nm.

Broken Bolt Case Study: http://www.fera.org.uk/cs5.htm
Affect of Lubrication on Torque: http://www.vfbolts.com/torque_value.htm
Useful Bicycle Torques: http://www2.bsn.de/Cycling/Torques.html

The attached docs are the cell 'care and feeding' guidance I have.

Andy

 

[newjeff]

Isn't it amazing that PSI supplied the same doc's to other users, but Bob's supplier failed to include these as it turns out very important documents for you. You would think that someone designing a BMS for a supplier/broker would need to have this information!