Modular 18650 Parallel Packs, Water Cooled Option!

[chaka]

I am getting ready to offer the battery system I use in my house designs in a more modular format. I am thinking a 1s4p pack with protection circuits good to 80-90 amps might be a good size. This would make them suitable for most 6s applications and ideal for 12s systems.

Balance function could be handled by a separate balance board and wiring harness. You could go without a balance board and use the standard balance feature on your hobby charger, build your board how you like!

Many other modular features could be made available such as voltage and temp displays, usb port/5v SBEC, power switch and last but not least an option to go water cooled! We could have the option to have a cooling system potted into the pack for those repeated high amp burst currents us adrenaline junkies demand. Pump, radiator and fan could also be made in a modular format.

The cooling system could run on 5v giving us the chance to finally make use of our UBEC's/SBEC's. System testing has shown a reduction of operating temps from 50 c down to 37 c, or 120 f to 100 f.

'This would be huge advancement for us desert dwellers. Winter is coming so I am sure I will also look into a compatible heating system for cold weather riding.

In short I would like to discuss possible pack configurations that are air shippable and compatible for all systems. Something that could be expanded upon easily without restrictions.

 

[Vulthor]

Sounds great.

Watercooling, waouh need pictures of that.

 

[LEVer]

Why not 2s4p? Nobody seems to be running odd numbers of packs here. Well, unless you have Alien/Flier 8s ESC which is actually only good for a fully charged 7s.

Water cooled? Heavy?

 

[torqueboards]

Interesting.... I think the 2S4P is nice too..

 

[chaka]

Why not 2s4p? Nobody seems to be running odd numbers of packs here. Well, unless you have Alien/Flier 8s ESC which is actually only good for a fully charged 7s.

Water cooled? Heavy?

I want to try and keep each individual component as small as possible so someone can add an extra pack for 30 bucks and gain performance and range incrementally. Top out at 12s4p and want more range? Drop the voltage down to 6s8p and add two 1s4p packs at a time.

Water cooling so far has been purely bench testing. The design of the pack is ready if anyone wants to join in the fun of setting this up on a board

Water cooled motor mounts are another item I would like to see happen and I don't see why it couldn't be implemented in a hub motor either. HobbyKing has a little water cooled motor mount add-on but our spindles are already pretty short.

 

[Vanarian]

Interesting, can you show pictures? How big is your water cooling system? What chemistry are your packs?

 

[chaka]

Interesting, can you show pictures? How big is your water cooling system? What chemistry are your packs?

I would like to keep this post directed on the concept of modular battery packs rather than me selling them but for conversation sake I am using the LG HE2 right now but I am moving on to the LG HG2. A 6s4p pack would give you over 260 Wh, a good starting point for someone new to the sport.

 

[jasonyng]

I am about to start my board I would be intersted in a 12s setup, I will send you a PM and get some details.

Jason

 

[Vanarian]

Thank you for the specs of cells. Considering you use cell format I'm really interested by this water cooling thing, if it can drop the temperature then batteries can be abused more is it ultra flat? pressurised? what BMS do you use which can support 90A with 3.7V? Is it compact? You should keep the concept going.

 

[torqueboards]

Is there a reason why you choose the HE LG2/LG HG2 cells versus Samsung 25R?

 

[chaka]

Is there a reason why you choose the HE LG2/LG HG2 cells versus Samsung 25R?

I am actually on the fence between those two models :lol: The HG2 has more capacity but only a bench test can prove that. I may have to order some and try them out too!

Thank you for the specs of cells. Considering you use cell format I'm really interested by this water cooling thing, if it can drop the temperature then batteries can be abused more is it ultra flat? pressurised? what BMS do you use which can support 90A with 3.7V? Is it compact? You should keep the concept going.

I have managed to keep the thickness down to 21mm. Going water cooled adds a a little extra width but the thickness remains the same. I potted tubing into the empty space between cells. Test so far have been with low almost passive flow and room temp water.

I use PCM circuits from bestech. If you want a few for your own builds I am always happy to do a group order.

 

[torqueboards]

The next thing is making packs flexible

 

[Nordle]

OMG! Big Need

 

[Ianhill]

Liquid cooling needs to be thought out carefully, I would say you need temp sensors dotted through the pack to know if it needs cooling or Heating you would need a cooling circuit to a radiator or heat displacement of some sort and a heat circuit (water jacket around the motor then these to circuits need to be mixed to get the correct temp or if over heating sent to the rad to cool, I would say u need a way of controlling valves and monitoring temps, a arduino would be needed to keep all the values in check and a nice piece of programing to suit.

 

[chaka]

Liquid cooling needs to be thought out carefully, I would say you need temp sensors dotted through the pack to know if it needs cooling or Heating you would need a cooling circuit to a radiator or heat displacement of some sort and a heat circuit (water jacket around the motor then these to circuits need to be mixed to get the correct temp or if over heating sent to the rad to cool, I would say u need a way of controlling valves and monitoring temps, a arduino would be needed to keep all the values in check and a nice piece of programing to suit.

No need to reinvent the wheel, monitoring fluid temp would be all that is needed. Set a thermostat to power a pump and radiator fan at near peak optimal operating temp and we are done! A motor mount cooler would likely have to be a separate cooling system because of the large temperature differential.

If you wanted a really nice system you could pump continuously and have an electric bypass valve bring the radiator/cooler into play when cooling is needed. This would equalize battery temps at all times making our weakest links a little stronger.

One hurdle is a good manifold system to route the coolant. We could plumb the cooling line in series, one line through the whole system for simplicity but I think plumbing each pack on its own loop would have better results. If each pack is on its own loop we could increase the overall flow of the system and better equalize the distribution of the coolant.

 

[Ianhill]

My old turbo car had a problem like this cold water enter the block by the time it was half way through it would have warmed up so a thermal camera showed a increase in temp across the pistons similar would happen with the battery's in a simple 1 route system but they would have to be under massive stress for unequal cooling to show it self

 

[onloop]

For this concept to be feasible The size of the water Cooling system needs to be smaller (& cheaper/lighter) than the area required to add more parallel cells.

Does the water cooler system need power? How much amps drawn? What voltage? Does it need a buck?

Also. I think it would be easier to design a drive train system that doesn't cause the batteries to over heat in the first place. Lower amp draw = less heat

If you cannot make a more efficient drive train maybe a simple Airflow intake could be considered? Even perhaps an aluminium block/heat sink/enclosure that had semi-circle cutouts that each cell where seated into.

At the end of the day I think your recent focus on batteries overheating is a by-product of your existing drive train system asking too much from your power system.

Keep it simple is my design method.

 

[whitepony]

cooling is what car manufacturers do as well (for example with non conducting oil) - not sure though how important that is for skateboards though ... with my 10S4P battery I went for a 2000m sprint at max speed (48.4km/h) and the battery wasnt warm AT ALL. I wouldnt know why I should need cooling with my 40 cells and Im not sure why you would with your crazy 80cells.

 

[Vanarian]

I definitely see benefits for some cell models with really high discharge rates, depending on situation the power you need is not constant (i.e. a top speed run does draw power during acceleration but you need less power to just maintain the top speed you've reached; on the contrary making a sprint climb of a hill would definitely need the maximum continously, thus maybe causing more warmth of the packs).

There is room somewhere depending on what you use :wink:

I also may second what Onloop said, the more efficient your ssetup is, the less heat you'll need to fight. But it doesn't mean you shouldn't try this system either.

 

[whitepony]

right, i tend to forget about hills cause there arent any around where i live ;p

 

[Ianhill]

Your not thinking about the charge cycle when most heat and failure occur that is why a cooling jacket for the battery's is need intense charging, the Tesla can charge to 80% in 30mins creating a lot of heat

 

[chaka]

I would like to remind my brothers and sisters that my goal for electric skateboarding is high speed racing. It will require near constant acceleration and unimaginable watt usage.

I few folks, most likely the same person has mentioned I run really tall gearing and that it is inefficient. I have done several days of watt usage testing and guess what, it isn't. I only pull 300 watts at 30 kph, 400 watts in a full tuck at 40 kph, 50kph......Don't know since I don't like looking straight down at that speed. If I stand straight up at 40kph I pull about 600 to 800 watts depending on wind speed.

What my gear ratio does allow is greater torque to be applied to my wheels, more teeth on the belt for less belt slippage and faster acceleration and better braking. If i use the gear ratio to its full potential yes it is very inefficient for extended range just like flogging your car uses more fuel to cover the same distance than driving with a conservative foot.

My latest iteration of my board design now has a temp gauge flush mounted topside so naturally I am moving to the next step in temperature monitoring, regulation. It may turn out that a cooling system is only needed in very hot climates for extended riding

cooling is what car manufacturers do as well (for example with non conducting oil) - not sure though how important that is for skateboards though ... with my 10S4P battery I went for a 2000m sprint at max speed (48.4km/h) and the battery wasnt warm AT ALL. I wouldnt know why I should need cooling with my 40 cells and Im not sure why you would with your crazy 80cells.

Tesla uses glycol/water in there cooling system. Nissan uses air :lol: Try doing non stop sprints till depth of discharge thats when the magic happens. However, your ambient temp is likely low enough for your pack since you technically do have an air cooled system. My packs are fully enclosed so heat tends to build up and dissipate and dissipate slowly, A great thing for cold climates but not so great for blasting around in the heat of the desert.

Back to the subject of a modular 1s4p pack with protection circuit. I am thinking this would be a great way for riders to gradually switch out to an 18650 format since they are completely compatible with prismatic lipo's and you wouldn't have to fuss with a BMS. 4p seems to be the lower limit for moderate/light use in a 6s system so that is my reasoning for not going lower. Yes, I consider cruising at 40-50kph light use. Yet you would likely pop the over current protection if you tried to accelerate to aggressively.

P.S. whitepony, if you don't already have a full face helmet I hope you get one soon https://youtu.be/dIbyYH_yW-Q

 

[chaka]

I did some testing and found out these protection circuits pop at around 60 amps. Looks like the only option for now is a 1s2p pack using these circuits. I am looking into a 2s4p circuit but it may cause some balancing issues or may not, only testing will flush them out.

I may have to try reconfiguring the 1s circuit to higher specs since I would like the option of not having a balancing circuit if we choose.