My first of several stupid questions

[MelloYello]

I'm collecting the bits to assemble my first li-ion battery pack and need to order a charger. The battery will be a 36v 10s5P configuration using LG MJ1 cells. The manufacturer recommends a constant charge rate of 1.7 amps, but that's for a single cell. Do I need to increase the charging amps for the pack configuration? I don't need it to charge quickly, I'm more concerned about the long-term health of the cells, so what is the optimum charge rate I should be looking for? Thanks.

 

[DanGT86]

Yes you can increase the charge rate for more cells in parallel. So your 5p pack can be charged at 1.7 x 5 = 8.5 amps and still be at the manufacturer recommendation.

Same thing goes for discharge rates. So a cell good for 20amps discharge current would be 100amps in a 5p pack assuming your pack is built well.

Generally the lowest charging current you can live with is the best thing for the cell. But if you are well within the manufacturers recommendations then there is no reason to put up with super slow charge times.

As far as I know the best thing you can do to prolong the life of a lithium cell is to not charge it all the way. In some cases stopping at 80% capacity can almost double the cycle life of a cell.

Try to avoid storing a fully charged cell for any amount of time. I typically store mine at 75-80% max and charge to full right before a ride. They don't like being fully charged and fully discharged so unless you need to get every last bit of energy out of the lightest pack possible its better to use the middle of the range like charge to 80% and stop discharging at 20%.

 

[MelloYello]

I appreciate the information. I've been reading for weeks but some answers continue to elude me. I'll be installing a smart BMS which might allow for some adjustment of the charge/discharge parameters. My thinking may be flawed, but I thought it would be best to look for a charger that isn't rated too low yet avoid one with excess amperage and having to totally rely on the BMS to reduce the charge rate.

 

[larsb]

Bms is not intended to do anything other than balance the cells and cut the power during error/protection.

The charger and motor controller takes care of the normal use cases so there are double levels of protection

 

[e-beach]

As noted above, your charger can be an 8amp charger (if you can find one that is 8 amps).

The easy answer is to get a 5 amp charger suitable for Li-ion batteries.

For your 36v setup, look for one that charges to 42 volts.

You will need to fully charge your battery pack the first few times and allow enough time for your bms to balance them. After that you can undercharge them as stated above.

A balance charge is periodically needed. (leaving it on the charger until all your cells are balanced.) Like maybe once a week or so depending on the health of your cells.

 

[MelloYello]

I really appreciate everyone's input, obviously I have a lot to learn.

My second charging question might be too specific to my particular situation for an answer, but I'll welcome opinions and best-guesses. I'm dealing with a scooter that came with a 24v 2 SLA battery setup. After checking the rating of the caps and fets, I added a third battery to see if 36v was possible. It did very well, so I decided to switch to Li-ion. My question is this: Currently the charging is done through a 2.1 plug that's attached to the PCB. Do you think it would be acceptable to charge the 36v battery through the same port or should I use some XT connectors? Here's what I'm dealing with, the charge port is on the left edge. I tried looking for the mfr of the board (dpnp technology) for information, but they don't exist anymore.

 

[e-beach]

...Do you think it would be acceptable to charge the 36v battery through the same port or should I use some XT connectors? ....

If you keep the amperage low you should be fine. Like 2 amps low. If you want a higher amp faster charge then change connector on the pcb to something that can handle a higher amperage.

That connector might take more then 2 amps, but why take risks.

BTW, what is the amp rating on your 24v charger?

 

[MelloYello]

The 24v charger is 2 amps. This board seems like it was built to handle more than the task it was given. The caps are labeled 50v and the mosfets are IRF3205, but that doesn't tell me anything about its ability to handle more charging amps. I certainly don't want to smoke this thing, I'd rather be patient on the time it takes to charge and not worry about it.

 

[e-beach]

Over building a pcb is normal. However if you have no problem with a slow 2A charge, then 36v at 2A will not fry your board.

The other question is weather it will let you fully charge your 36v battery?

 

[MelloYello]

Over building a pcb is normal.

Speaking of that...I was surprised that the bar across the mosfets is steel. I could easily make one from aluminum with more surface area for better heat dissipation. Anyone see a downside to that?

(and my apologies to everyone for drifting off the Battery Technology topic, just taking advantage of the pic that is already posted)

 

[Chalo]

Speaking of that...I was surprised that the bar across the mosfets is steel. I could easily make one from aluminum with more surface area for better heat dissipation. Anyone see a downside to that?

It may be steel for stiffness (to prevent point loading of the MOSFET packages). If you substitute aluminum, use at least 1.4 times as thick a piece of material for equal or better stiffness.

 

[amberwolf]

Some notes:

I can't do a complete analysis of the circuits on that board, especially because I can't see both sides so many traces are not visible, and a number of small transistor-style components' markings aren't visible, but the right half looks like a typical analog-type brushed motor controller.

Note that the LM339 chip that runs (is) the controller typically has a max voltage allowable of only 30v. Same for the LM393 over on the left side. How long they will last at the higher voltage, I don't know. There may be a zener-diode+resistor "regulator" (voltage limiter, really) but it's resistor is likely calculated for the original voltage, so the current to the zener may be higher than allowable for the desired output, and it may either not limit properly or it may fail and then suddenly allow full voltage to the chip when it does (which would then cause the chip to fail).

LM393

Supply Voltage - Min: 2 V
Supply Voltage - Max: 30 V
Operating Supply Current: 225 uA
Minimum Operating Temperature: 0 C
Maximum Operating Temperature: + 70 C

LM339

Vs (Max) (V) 36
Vs (Min) (V) 2
Operating temperature range (C) 0 to 70

It's hard to tell for sure, but it looks like someone has bypassed the current-limiting / circuit-protection shunt (the smaller straight bare wire on "top", toward the FETs) with a piece of black-insulated wire; if so this prevents the controller from protecting itself against overcurrents, and can cause sudden unexpected failure. Failure of brushed motor controllers usually means it is stuck on, so the motor then runs at full speed and can't be stopped except by unplugging it or the battery from the controller. (or otherwise disconnecting power from the system).


FWIW, http://4qd.co.uk has a learning section that has a fair amount of brushed motor controller info, including schematics and discussion of their older 2QD versions that are similar to this controller, if you want to learn how they typically work. A good starting page: https://www.4qd.co.uk/docs/what-is-pwm/


The left half looks like a comparator setup most likely to show charge state via the red/green LED at the left edge.

It may possibly also be to disconnect the system (via the SCR; not sure how they turn it off, or if they use it as a crowbar to blow an inline fuse somewhere (not visible)) in case of battery undervoltage/overvoltage and/or possibly for controller overcurrent conditions.

Not sure why the rectifier (biggest thing in the top left corner); perhaps the original charger is just 24VAC (the marking next to the connector sort of looks like it says this), so it just rectifies it and then feeds it into the lead-acid battery, which thru it's internal resistance/capacitance smooths this out into the necessary charging conditions.


At any rate, it's likely that none of that leftside circuitry is relevant to your 36v lithium battery (unless it includes controller-circuit protections), and so you should probably just charge that battery directly at it's own charge port so it's BMS can do it's job without interference from outside. If there *are* controller protections built into the circuitry, they probably act on the B+ or B- tabs at the edge of the board where the battery's discharge connector plugs in, anyway. (and not the charger port).

 

[amberwolf]

I'll be installing a smart BMS which might allow for some adjustment of the charge/discharge parameters. My thinking may be flawed, but I thought it would be best to look for a charger that isn't rated too low yet avoid one with excess amperage and having to totally rely on the BMS to reduce the charge rate.

A BMS doesn't do anything other than shut off input or output completely, based on whatever it's protection limits (HVC, LVC, overcurrent) are. Any overcurrent protection it may have (if any) is not likely to include the charging side, only the discharge side (to protect against excessive controller/motor load)--it expects the charger to never exceed the max charging current on it's own.

It can't limit current or voltage the way the charger does, so you need to use a charger designed and rated for the charge rate and final charge voltage that you want.

If you need to be able to change those, you can get programmable ones like the Cycle Satiator from http://ebikes.ca , or one of a few different analog-adjustable ones with knobs (or screwdriver adjustments).


A BMS that has programmable parameters will be things like HVC, LVC, balance trigger points, and possibly discharge current limits--but all of these are just hard turn on / off points, meaning when it reaches them it turns the input or output full on, or full off.

It's up to the controller to limit discharge current to a safe level for the battery, and up to the charger to limit charge current to a safe level for the battery.

 

[MelloYello]

Wow! So much information to digest, I need to read it over several times to wrap my feeble mind around it. VERY much appreciated. You are correct wolf, there is a bypass wire on the shunt. I installed it after watching several-too-many Y-tube videos telling me shunt modding was an easy way to unlock additional performance. In reality, I didn't notice any difference but didn't bother to remove it. Your description of what is possible tells me it needs to go. The 30v supply voltage of the LM393 chip has me concerned, but I did ride this thing for about 30 minutes with a third sla battery to see if 36v would kill it and it still runs. I'm inclined to go forward with this and hope for the best, realizing I might have to source a more appropriate pcb in the future. If something goes *fzt* on the board, I can deal with that, I just don't want to get into a situation where the battery pack ignites.