Photobug said:
I do not know the details of the business but also assume during busy season most of the bikes are out every day.
If you want to develop a system that will actually do the work they need, you will need to find out the details of their usage patterns, and of each of their bike's batteries and chargers. Without that info, you're just building a general solution that may or may not do the job they need, potentially wasting their money and time at best, or making a dangerous hazard at worst, depending on the solution you build and how they end up using it. (they may not use it the way you intended...and that is where you have to build safeties in they can't accidentally bypass, or just plain build it so they can't do it wrong).
If I sound a bit harsh, it's because batteries can be serious fire hazards when treated the wrong way, and it doesn't take more than one wrong charging to damage one to catastrophic levels. Since the end-user (the business's employees doing the charging) usually doesn't really know a lot about all this, any solution you make will almost certainly need to be foolproof and failsafe, or you could be on the receiving end of a lawsuit if it burns their business down. We'd all prefer to avoid that.
I probably didn't remember to cover all of the potential gotchas, but I tried to do so. Hopefully someone else will add anything I missed.
*********************************************************
Overnight Bank Capacity
So would need to develop a battery storage situation. As much as would like to do a LIPO, I don't think it is in the budget or at least not a big enough battery bank to support charging the entire fleet each night.
There's no need for something as power-dense as LiPo (like RC Lipo, for instance), especially since those usually have a pretty short cycle life.
For a battery bank on a solar setup, you're looking more for high capacity and long lifespan. Since lead tends to work better when kept near full charge, FLA battery banks (like car batteries, as many in parallel as needed to store the number of Wh required to charge the bike batteries) would probably work better for this purpose.
If budget won't allow that big a battery bank, but they do need to do most of their charging when the solar isn't active, then you should figure out which bikes are the lowest Wh batteries, and setup a bank that will fit in that budget that will charge however many of those bikes it can. Then mark those bikes as the "solar charged" bikes, so the employees only charge *them* on the system overnight, so it doesn't get depleted and not charge the ones on it enough to be used the next day. (since the overnight system is presumably not monitored by anyone, and simply assumed to be charged to full by the next day, as that is usually a safe assumption in these cases for AC-powered charging systems, and they'll apply the same assumptions to the solar).
I could calculate the total batteries in the fleet and amp draw of the chargers but do not know the flow of the bikes, and I don't know that they know either. Are 90% of the bikes out each day and do all the batteries come back with the batteries drained? They may know the bikes out percentage daily but likely have no idea what capacity of the batteries are used daily.
Does teh grant require that the solar system do all the work of charging the bikes? Or only some of the bikes? or??
To build a system that even *can* do the work, you need to at least guesstimate the total Wh of the batteries in the bikes that have to be solar-charged; the OEM batteries may even have this printed on them. (if not, you can calculate a guesstimate by multiplying the V x the Ah)
Assume worst case, that every bike comes back empty every day, and has to be fully recharged overnight to be ready for the next day.
************************************************************
Monitoring.
Part of this grant they have is to record and keep track of the amount of energy and savings from the solar setup to charge the e-bike batteries.
Then they will need to get some watt meters to monitor charging; enough to monitor every bike that is being solar-charged.
This can be done at the bike level, but it's easier to do it at the AC input to the charger, with something like a Kilawatt meter. If the meter can handle the current draw from all the chargers operating at once, you can use just one, but it's likely you'll need more than one to do this. Since you don't have to monitor each individual bike, just the total power in Wh being drawn from the system (to compare kWhr costs to the regular AC system), all that has to be done is to reset all the meters each charging session, and total the numbers from them all, so however many meters it takes to support the total current draw from all the chargers at maximum current draw simultaneously.
They should be able to assume that (if using inverter on the solar) that the watthours used on it are the same as that of regular AC; it may be different (worse) on the solar inverter if the inverter is not a good sinewave like the regular AC; powerfactor of the chargers could be worse and efficiency would drop.
If you don't use an inverter, you'll need to use a DC-type wattmeter instead, between the individual current-limited DC outputs of the solar system and each battery, because you cannot wire all the batteries in parallel, even if they were the same voltage types, for multiple reasons--each battery *must* have it's own separate DC-DC from the solar source; this is going to cost (probably a lot) more than the battery-inverter system. Unless they make a meter you can use with the variable voltage / variable current DC of the solar panels themselves, that can handle the worst-case simultaneous current draw of all the separate MPPTs, (or you use separate panels and meters for each MPPT, etc) and the meters will not interfere with the MPPT operation, there won't be a way to use a single or grouped set of meters to measure Wh, you have to do it at each battery's charge input.
**************************************************
DC charging
I like the idea of avoiding an inverter as it is just one more expense and weight if this thing becomes mobile.
It would be nice, but since you are not dealing with a single battery voltage or type, then to do a non-inverter system you are going to be building not one system for all the batteries, but what amounts to a complete system for each and every battery they have.
If you were doing this for just one bike, (not one at a time, but just one, period), using an inverterless straight DC system is relatively easy. But doing it for a bunch of different bikes with different batteries is not.
Even doing it for a bunch of the same bike that have to be charged at the same time is not.
First, you can't just apply a DC voltage to a lithium battery to charge it.
You must have the specific maximum voltage that specific battery charges to, and not higher or lower. You must have the specific maximum *current limited* current that specific battery requires (or lower, but not higher). If you don't do this, you may damage the batteries or even cause a fire. (the fire might not happen the first time you do it--it may take cumulative damage to do it, and it could ignite during any usage, at any time, not just during charging, once the damage occurs).
So you can't just use an MPPT with output high enough to charge the highest voltage battery, and run with it. If you have controls on that MPPT that the person doing the charging can set to the specific for each battery, *and you trust them to always always always do this correctly* (which isn't going to happen, someone will make a mistake eventually and damage a battery or cause a fire), then you could use a single "programmable" MPPT to charge *one battery at a time*.
The current limiting is extremely important. The battery doesn't do this itself; the charger has to. When a battery is empty, without current limiting in the charger (that matches what the original charger limited to), quite a high current can flow when applying a voltage to it. If it's not enough to damage the charging voltage source itself, it'll probably be enough to damage the battery, or the charging wiring, possibly even enough to start a fire during the charge process. (if not, enough to damage the cells to potentially cause one later).
Some batteries are capable of being charged at a somewhat higher current than the charger they come with is designed for; sometimes cost is the only reason a slightly higher current charger is not used. But sometimes it's because they calculated out (or tested) what current the cells can take as they age, which will drop over time, and they picked a charger that will "never" exceed that, so the packs never get damaged by the charging process itself. So unless you analyze every pack being charged for it's internal parts (not just every brand, but each individual pack because they may not all be made of the same cells, etc), you're much much safer to assume the manufacturer knew what they were doing and use only the maximum current that the charger that came with the pack is limited to, for charging that pack.
**************************************************************************
Second, you cannot wire multiple batteries to the same one simultaneously. Only one battery can be connected to one charger or MPPT at a time.
Each battery will come back at a different state of charge, which is a differnet voltage level. Connecting them together means wiring them in parallel at the charge port. Doing so with different voltages means current will flow between them. If this current exceeds that which the BMS in the battery can handle (especially since the charge port is not designed to handle reverse current), it will damage the FETs. That usualy means they get stuck on. It's a silent failure, so you won't even know this has happened, most of the time. When that happens, the BMS can no longer ever prevent overcharge of a battery; as it gets out of balance over time, high cells will get higher, and low cells will get lower, until a high cell eventually is at the overcharge point where damage or fire occurs.
It's even possible that, since the current isn't limited except by resistance (which isn't much) when this occurs, you could have a nearly full battery wired in parallel with a nearly empty one (or worse, a bunch of nearly full and just one nearly empty, if wiring many in parallel), and so much current flows from the full to the empty that it heats the empty cells so much they just burst into flame right then and there. If the cells don't, the wiring could.
****************************************************
Why an inverter is "better" in this case
Even though an inverter wastes power...it means you can use the original chargers, and the end user only has to plug them in, just like they always do. Any mistakes they could make they could make without your system, and therefore are also not your fault.
The inverter just provides regular AC power. The chargers do all the work they originally did, and the batteries are at no more risk of damage or fire than they were without the system you will design / provide.
You don't have to build a system that considers each individual battery's needs, and design it to ensure the end-user can't screw up and plug the wrong one into the wrong charging port so they don't burn their business down or set a bike on fire under a rider later.
You don't have to build Wh monitoring into each charging setup; you can just do it at the inverter's AC output. The end-user doesn't have to total and reset each battery's monitor, they just record and reset one or a few AC-side meters.
*******************************************
Important safety considerations of DC out MPPTs (or any other source)
The MPT-72 seems like a nice option.
Does it have user-settable voltage and current limits? If not, you can't use it to charge the batteries safely.
To be safe, it must have a current limit that does not exceed that of the charger for the specific battery it is being used to charge.
To be safe, it must have a voltage limit that matches that of the charger for the specific battery it is being used to charge.
If it's not user-settable, you need one for each separate battery, one which has already been set by you or is factory-set, to those limits, and it needs to be setup so it can't be connected to the wrong battery.
If it *is* user-settable, then each person that will ever do the charging will need to be trained on how to do that, and they must be trustable to always do it right, every time.
Information from companies:
When I looked at the commercial e-bikes it looked as though there were only two prongs on the battery so maybe no communication or control from the charger, but I will reach out to the battery companies to get their input on the potential to charge without and inverter.
It's likely that most don't have charger communication, but there are some systems that do. They almost always have extra pins on the charger port than the two needed for current flow, so if there are no extra pins, it is a safe bet that they don't have charger communication.
For OEM bike companies, they will almost certainly tell you the warranty is void if you use anything other than the original charger, and that nothing except the original charger is safe / allowed to be used / will work on them, depending on what their lawyers tell them to say. If they do say anything else, you'd really need the information to come from their engineering department to trust it, as anyone else may not know what the real information is, and could just be going by marketing info or ad copy. That's pretty common, unfortunately.
In most cases, your best source of information on what specs to use for charging is right there on the charger that came with a particular battery. Don't use anything that exceeds those numbers, and it should work like the original charger, as long as it operates in the same way as the charger.
It is possible the whole project will need to be scaled to show proof of concept on a smaller scale, as in only build the system to charge a portion of the batteries but size at least the MPPT to grow to accommodate more panels and batteries in the future.
That's probably a good way to do it, if the budget is limited. Build a foolproof small-scale system, then as budget allows, duplicate that system as many times as necessary to do the work required.
