Potting Batteries, Thoughts and Experiments

Nice!
Can LVC, HVC and balance V be adjusted?
Is BT to monitor pack on the fly?
Can several BMS's monitored simultaneously?
Can they be connected and charged/discharged in series? (how many?)
Will you sell them alone, without battery pack? :)
 
YfZQwQs.png


2 spare "bricks" OK inside cabin. so check-in your mobility device without battery. then carry two of justin's bricks. sounds like a plan 8)
 
justin_le said:
Alex W said:
Justin: is there any news on productizing these? A friend is working on a bike that these would be ideal for. I noticed that the non potted small packs are out of stock with no return date on your site.

Hey Alex, yeah we decided to do a fully home-brew BMS route rather than trusting a 3rd party BMS supplier for this project and have just dialed in the final touches on that. Here's a bunch of the panelized boards right after firmware programming this afternoon. You'll notice five RBG addressable LEDs and an push button, plus if you look really closely you'll spot a bluetooth antennae
View attachment 1


We're hopeful to get all the rest of the manufacturing ironed out to release before the end of the year. We've been sitting on 500 10s x 1p packs made up of LG-MG1 cells for a while that are just waiting to get built up into these stackable modules, and I'm keen as anyone to get them in people's hands and ebike projects!

Will this BMS be available separately? are the limits programmable?
 
Overclocker said:
2 spare "bricks" OK inside cabin. so check-in your mobility device without battery.

2 Spare bricks?? It says pretty clearly on this Cathay Pacific guideline that you can can have up to 20 pieces of spare small lithium battery in your carry-on baggage! That's great actually, on all the other airlines documents I only ever saw the upper limit of 2 packs mentioned for the 100-160Wh batteries, but not a count on precisely how many of the <100 Wh batteries could be allowed.

CathayPacificLimit.jpg

That makes me feel a lot more at ease when I'm going through checkout with 4 or 5 of these in my back pack.
 
justin_le said:
We're hopeful to get all the rest of the manufacturing ironed out to release before the end of the year. We've been sitting on 500 10s x 1p packs made up of LG-MG1 cells for a while that are just waiting to get built up into these stackable modules, and I'm keen as anyone to get them in people's hands and ebike projects!
Looking forward to another leap forward by eBikes.ca. Well done!
 
Overclocker said:
oh you're using MG1 which are just under 10Wh (i just looked it up). so 10s within the 100Wh limit.

Yeah, that's why the 48V (13s) option while we considered it isn't nearly as attractive, since you'd be limited to using <2.1 Ah cells and not being able to take advantage of the cutting edge stuff that has both amazing energy and power density. But with 10S, you can get the best of everything, coming it at 98-99 Wh with top of the line cells. It really hits a sweet spot with the regulations as they are.
 
justin_le said:
Overclocker said:
oh you're using MG1 which are just under 10Wh (i just looked it up). so 10s within the 100Wh limit.

Yeah, that's why the 48V (13s) option while we considered it isn't nearly as attractive, since you'd be limited to using <2.1 Ah cells and not being able to take advantage of the cutting edge stuff

Couldn't you just spec the pack to use say 70-80% DOD then label the pack capacity as <100Wh's. Make it a default setting in the BMS to only charge to 3.9v or so. Then for those customers that are not often flying can increase the charge limit to get the cells a fuller charge. After all we all know watt hours depends on the cells charge voltage.

Nice looking BMS, looks like it has a lot going on... :shock:
 
justin_le said:
Overclocker said:
oh you're using MG1 which are just under 10Wh (i just looked it up). so 10s within the 100Wh limit.

Yeah, that's why the 48V (13s) option while we considered it isn't nearly as attractive, since you'd be limited to using <2.1 Ah cells and not being able to take advantage of the cutting edge stuff that has both amazing energy and power density. But with 10S, you can get the best of everything, coming it at 98-99 Wh with top of the line cells. It really hits a sweet spot with the regulations as they are.

As cells continue to improve, you'll go over 100 Wh with a 10s pack pretty soon. I'd be interested in 7s1p or 2s4p packs. Those could be easily assembled into 14s packs, but are still under 100 Wh, even with the next generation of cells.
2s packs would be the most flexible in terms of voltage, but you could only do 4p or 8p easily.
You could also make a "LiGo add on pack" or something that was 3 or 4s and 2 or 3p to get to 48v.
 
I heard there was a bit of trouble shipping the formula E batterys around due to the 100wh shipping rule, There's good innovation happening there, specially with battery management and form factor.
 
redilast said:
Couldn't you just spec the pack to use say 70-80% DOD then label the pack capacity as <100Wh's. Make it a default setting in the BMS to only charge to 3.9v or so. Then for those customers that are not often flying can increase the charge limit to get the cells a fuller charge. After all we all know watt hours depends on the cells charge voltage.

It's a temping and obvious loophole, but given that the regulations are really about grams of lithium, and then they say "as an alternate to grams lithium you can use this formula based on nominal watt-hours", I don't think that would stand up to much scrutiny. The potential for damage from a cell doesn't change just because the charge voltage cutoff on the BMS. But yeah given that all the authorities have to test from the pack is the available watt-hours, you can easily set the BMS to limit that to <100 Wh even if the cells are intrinsically capable of much more than this. The 4.2V/cell reference for 100% SOC is somewhat arbitrary.

redilast said:
As cells continue to improve, you'll go over 100 Wh with a 10s pack pretty soon. I'd be interested in 7s1p or 2s4p packs. Those could be easily assembled into 14s packs, but are still under 100 Wh, even with the next generation of cells.

When the cells get close to ~3.8-4.0 Ah in the coming years then I think a 7s-1p layout will be in the cards for sure. There would be a pretty decent market outside of ebikes for 24V bricks like this, and then it's trivial for ebikers to series them to 14s.

You could also make a "LiGo add on pack" or something that was 3 or 4s and 2 or 3p to get to 48v.

That too! It's too bad that 15s isn't more common, since a 5s2p pack would let us use our same mold tooling and cells. That in series with a standard LiGo gives a 55V nominal setup, which at 63V hot off the charger is still within the range of 48V rated motor controllers. Anyways, distracting thoughts for now, as we need to get the standard 36V 2.7Ah unit fully fleshed out and on the market before entertaining more variants.
 
Apologies if this is going off topic but are your BMS going to be for sale separately (and are they able to go up to 13S)?

It's a bit of a problem at the moment with obtaining BMS from China direct... either buy cheap and test (and find out its a hand grenade!) or pay through the nose and hope for better quality... the testing really is the valuable part with these things and I've been hunting round for somewhere more reliable than my other sources.
 
Lurkin said:
Apologies if this is going off topic but are your BMS going to be for sale separately (and are they able to go up to 13S)?

It's off topic but I'll answer it publicly since you are far from the first to ask/wonder about this. The answer is no. The direct retail sale of bare BMS boards is the very last thing I would want to get involved in, and once you've witnessed the support side of it you'd understand this is something no sane person would entertain. But if it does prove reliable and rock solid in our LiGo modules we may look at B2B arrangements with other upstanding oem's who are making neat products which could take advantage of these BMS features.

It's a bit of a problem at the moment with obtaining BMS from China direct... either buy cheap and test (and find out its a hand grenade!) or pay through the nose and hope for better quality... the testing really is the valuable part with these things and I've been hunting round for somewhere more reliable than my other sources.

It's been like that for years and I don't see that changing any time in the future I'm afraid, which is unfortunate for the few competent people out their making DIY batteries. It's just that there are sooo many more people doing DIY packs who simply shouldn't be, and they spoil it.
 
Hi Justin, thanks for contributing important information. As a customer I'd buy several modular potted 1P groups and/or BMS. I'd also be keen on a 'developer' style release of the BMS where support is not available.
 
Rube said:
Hi Justin, thanks for contributing important information. As a customer I'd buy several modular potted 1P groups

Well, that is now possible! After all the inevitable last minute show stoppers to come up again and again we're now live with the devices in full beta production mode, most of the details are here:
http://www.ebikes.ca/product-info/ligo-batteries.html

We went through a few design iterations for the pack profile and in the end settled on the simple rectangular layout on the right
LiGo Design Evolution.jpg

The technique to do the glass fiber reinforced epoxy potting around the cells also took a few iterations to get right, here's how the final mold looks
LiGo Tooling Sample.jpg

We have 5 of these molds CND'd so are currently limited to making 5 packs a day or 10 packs in a long day, but we'll hope to step that up to higher volumes soon.

I was a little worried about how messy the wires might end up becoming when people do larger pack assemblies using these as building blocks. For just 3 or 4 modules with a folding ebike kit dealing with a few parallel connections is no big deal, but it could get unweildy with big stacks. Fortunately with the genius of anderson powerpole connectors it can actually be made pretty clean, with just a single master plug of stacked powerpoles to do all of the parallel joining. Here is an 8 block (800 watt-hours) all stacked side by side
LiGo 8 pack in block.jpg
And then the same block rearranged 2 long and 4 wide, which is just the perfect size to hang in a bicycle frame bag.

As it is shown it is 36V 22Ah, but could just as easily reconfigure for 72V 11Ah. Each pack has 2 pairs of wires coming out of it, and so after all the parallel connections are done you end up with two spare unused anderson plugs, one of which typically connects to the controller, and the other one can be used as a charge port or to run lights or other accessories.
 
Cool. Any idea on the price point?

Also just curious about how long it takes to assemble one battery pack module? Including the time to weld the cells together and attach the BMS, wiring, potting. Like labor time, not epoxy curing time.
 
redilast said:
Also just curious about how long it takes to assemble one battery pack module? Including the time to weld the cells together and attach the BMS, wiring, potting. Like labor time, not epoxy curing time.

That's a question I should probably have a more exact answer to by now. It used to be quite long but I think we've got things dialed in to less than an hour each at this stage, though since a lot of the steps are done in batches at different times rather than seeing a single pack go from start to finish it's hard to say exactly. We had the cell tabbing into the base 10s x 1p packs done by an LG subcontractor so it's BMS, wiring, programming, light pipe bonding, fiberwrapping, potting, demolding, and testing.

Here are a few pics of the current process. There's a bit of origami at first with woven fiberglass cloth to surround all sides of the pack but also not result in loose glass strands that can prevent a seal in the mold plates


Prepped in Mold.jpg

We went from hand mixing and vacuum degassing the resin to running a pneumatic powered dispenser with static mix tube and shaved off a fair bit of time and mess with that
Dispensing into LiGo.jpg

Ideally the two plates of the mold pop off without taking chunks of text with them. We found it works best to remove the side plates when the resin is cool, then put it in the oven to soften the resin again before pressing it out of the aluminum center piece so that corners don't chip and flake off.
LiGo Demolding.jpg

Since we are doing batches of 5 at a time it gives a good sampling number to make a single change in the process and then see how that affects the finished result in a statistically meaningful way. Just figuring out the optimum type and quantity of mold release to have the sides remove from the mold easily but without looking too glossy involved like a dozen iterations, just compare the difference on the left and the right
Effects of Mold Release.jpg

Also, for those wondering why the alignment button holes are twice as dense than the buttons themselves, it's so that the packs can be stacked on top of each other with a stagger so that there are more options for fitting them inside the triangular frame cavity of a bike
LiGo Staggered.jpg

Cool. Any idea on the price point?
They are listed on our web store site or those interested in that, but I'd like to keep this discussion as a technology and development thread
 
Fantastic work.

Have you given much consideration into adapting the packs to be USB-C compatible?


My guess is someone in Europe will develop 20V 100Wh pack for some sort of last mile EV that docks two in parallel.
You could un-dock the modules into singe packs to charge a laptop or phone then parallel them for 200W discharge (ideal for EN250W law)
 
justin_le said:
Overclocker said:
oh you're using MG1 which are just under 10Wh (i just looked it up). so 10s within the 100Wh limit.

Yeah, that's why the 48V (13s) option while we considered it isn't nearly as attractive, since you'd be limited to using <2.1 Ah cells and not being able to take advantage of the cutting edge stuff that has both amazing energy and power density. But with 10S, you can get the best of everything, coming it at 98-99 Wh with top of the line cells. It really hits a sweet spot with the regulations as they are.

They may be sort of old 2012 tech 18650s (still, awsome quality), but the Sony VTC4 are 7.4Wh rated (2000 mAh), so it's actually possible to make a 48V 13S1P LiGo pack with them and still be under 100Wh (actually 96Wh). And what's awesome is that each cell is rated to up to 30A drain, and can easily take 15Amps continous without heating much.....

Make a 13S2P pack with two 96Wh-13S LiGo batteries and with those 192 Wh (4.0Ah) you can run your BBSHD for 10-13 km (with an averaging 15-20Wh/km riding style).
I think it would even be reasonable to expect 20 km range with those 192Wh when using PAS level1/5 (around 300W usage) on the BBSHD (I get 25-30km/h constant speeds on PAS 1/5).

BUT i have to admit.... 14S would be even much better (more punch with the BBSHD) !!!



Sony US18650 VT4 datasheet : https://cdn.shopify.com/s/files/1/0674/3651/files/Sony_VTC4.pdf?838
 
Matador said:
Make a 13S2P pack with two 96Wh-13S LiGo batteries and those 4.0Ah you can run your BBSHD for 10-15 km.
BUT i have to admit.... 14S would be even much better !!!
Sony US18650 VT4 datasheet : https://cdn.shopify.com/s/files/1/0674/3651/files/Sony_VTC4.pdf?838

Well actually, with 13S2P and a 30A BBSHD controller, cells would see a max of 15A so more like 1900 mAh/cell (3.8Ah pack in 2P)...

VTC4.jpg
Sony US18650VTC4 (Rated 30A, 2100 mAh).jpg
vtc4 graph2.jpg
MB1haVz.png
KhGMgrT.png
 
Thinking a pair would be best solution for winter bike. Share trips and batteries fit in parka pockets. Don't qualify for testing, but it's not a common use. I future a pair once the wrinkles are gone will be a great solution. Actually my 36v would be fine year round with small packs for grocery runs. Small town...
 
That looks like a great product. I love it. Looking forward to seeing what comes of the blue tooth functionality.

This would be perfect for me - I've been hoping and waiting for a pack like this for years - safer and easier than LiPo bricks! At my power needs, a pair of these would get me about 12 miles full throttle - great as a range extender for my primary battery. I could probably get 20 miles out of it at 60% power for pedal assist. And for just 2kg weight. Nice...
 
Matador said:
They may be sort of old 2012 tech 18650s (still, awsome quality), but the Sony VTC4 are 7.4Wh rated (2000 mAh), so it's actually possible to make a 48V 13S1P LiGo pack with them and still be under 100Wh (actually 96Wh). And what's awesome is that each cell is rated to up to 30A drain, and can easily take 15Amps continous without heating much.....

Hey Matador, trust that it is on the radar and you are spot on in the cell type we'd want to use for this. In a perfect world, we'd have the 36V 2.75Ah modules as we do now, plus a 48V 13s 2.0 Ah high power module as you describe, plus a 7s 4.5Ah module based on he 20700 cells. All packs would be just under 99 Wh to take most advantage of this size window for shipping, and there would be solutions for 7s (25V), 10s (36V) 13s (48V) 14s (50V) and 20s (72V) systems. We have our work cut out to get the 36V modules into a commercial success which needs to happen first, and then it's good to know that after that there is a natural path to extend this into a full product family so that it's useful to a wider range of setups.

chas58 said:
This would be perfect for me - I've been hoping and waiting for a pack like this for years - safer and easier than LiPo bricks! At my power needs, a pair of these would get me about 12 miles full throttle - great as a range extender for my primary battery.

For that 2 packs is great put in parallel with the primary battery. For use as its own battery, we'd generally recommend at least 3 LiGo's in parallel so that the average discharge current is under 2C, just to ensure that you still get good cycle life from the cells. That's still only 1800 grams and fits easily in many bicycle seat bags.

Anyways, for those who didn't see it yet we published a video the other week which gives an intro and summary of the features present in the first release of these packs:
[youtube]zjaUeh3jbFM[/youtube]

One of the main features which we had to take out of the initial release firmware is wireless pairing of all the modules in a pack, so that you can press the button on just one of the batteries and it will turn all the LiGo's on or off in unison. Those who saw us with these at the Interbike show last year got to see this in action, but it introduced a number of convoluted bugs that we still need some more time to work out.

chas58 said:
Looking forward to seeing what comes of the blue tooth functionality.

Well I'll be keen to share and show once it's ready! Thanks for your interest guys. The feedback so far has been pretty good, and partly as a result of that I've added the option of selecting XT60 or XT90 connectors rather than Anderson plugs on the store site for those don't quite share my full love and enthusiasm for anderson powerpoles:
LiGo Connector Choice.jpg
 
Back
Top