Charging a LiFePo4 battery while using the motor?

[Fastolfe]

I apologize in advance if the answer to this question is obvious, or if it has already been discussed, or if I'm being thick, but I'd really like to understand this fully before going ahead with my project.

Okay so the question is this: can you charge a LiFePo4 battery pack while at the same time using the motor? Or in other words, will the charger be totally confused because of the voltage drops due to the use of the motor?

The reason I'm asking is because my electric bike is now fully functional, and I'm now ready to move on to the next stage of my project - which is build a trailer with a solar panel to power the bike / charge the battery on the road during my tour next summer.

I'm planning on getting a reasonably small solar panel of some kind (100-150W, not sure what kind yet), connect it to a Genasun lithium solar charge controller, and connect this controller to the battery. I believe it should charge the battery fine when the bike is not moving (provided there's enough sun of course ). But what I'm concerned about is this: if I ride and use the motor at the same time, there will be a significant voltage drop seen by the Genasun controlelr (not to mention a lot of induction noise). When I stop using the motor, the voltage should go back up. Won't that confuse the hell out of the controller?

My understanding is that lithium battery chargers are supposed to follow a CC/CV charge profile (or at least CC to get the battery pack charged to 95% of its capacity). So I suppose they constantly watch the battery's voltage, and switch from CC to CV above a certain voltage. I wonder if wild voltage variations due to the load of the motor will throw the charger "off track" so to speak, and cause it to damage the battery.

My intuition is that it should work though. I've googled quite a bit on the subject, and found of couple of pages made by people with electric bikes and solar panels who mention they use a Genasun controller and a LiFePo4 battery. So apparently there's nothing more complicated to it than connecting everything to make this work, as these people didn't seem to make a big deal out of it. Also, if the controller is indeed a CC source (at least for part of its charging cycle), it stands to reason that if the motor is pulling amps, the controller will simply supply part or all of the current it supplies to the motor instead of the battery.

Of course, there's a simple way of testing if this works: I have a 28V 8A wall charger with a bargraph display. I could just throw the bike on rollers with the battery fully charged, connect the charger, and use the motor. If I'm correct, the charger's display should drop and show it's "charging" when I use the motor, and go back up and stop charging when I let go of the throttle. But before I try it, do you think I risk damage to either the charger or the battery doing that?

 

[eee291]

there should be no problem with what you're trying to do at all

 

[dogman dan]

Agreed, the charge controller will still put out it's maximum, even when the motor use causes the battery to sag lower. While riding, the charge controller will allow whatever the panel is putting out to flow into the battery. Or when parked, and the battery is less than 90% full, the same thing will happen. You'll have all the panel + charge controller can put out going into the battery.

Only when the battery is totally full, and the bike is stationary, will the solar charge controller slow down the flow, as it should, and must.

Will the battery gain charge while you ride? Not likely, unless your panel is truly huge. But suppose your panel is putting in 80 watts, and the motor taking out 500w. The battery will last as long as it would have if your motor was pulling 420w. Think of filling a sink with the drain plug removed, it will take less flow to fill it if the drain hole is smaller. It may never fill, but you still end up with more water in the sink.

Go for it.

 

[icerider]

Fastolfe,

Please post your results when you try the experiment.

I have a similar problem with a slight twist. I would like to solar charge my current Bionx system as I ride, but on the 48 volt systems the charger is 26 volts so it appears I have an additional boost converter and control circuitry inside the battery case on the "charge" side of the battery. I am concerned that the 200W of solar panels will not consistently generate the 90W of 26 volts that the battery is expecting and when the solar output is low, that the voltage coming out of the boost converter will drop below the 26 volts that the battery charging system expects.

 

[Fastolfe]

Thanks for your replies guys. I'll try all this when I have some time.

@icerider: if your battery has the charge controller built in, that sounds like bad news for what you want to do. Me, I plan on using a Genasun MPPT solar panel controller designed to charge lithium battery packs. It's designed to "see" the pack directly, and it has its own boost converter. I doubt very much it'd work through another charger controller - not to mention the losses of yet another boost converter. I'd say your best bet would be to use a 48V solar panel controller for lithium batteries and charge the battery directly on the main power line. But I don't know the Bionx system, so don't take what I say at face value

 

[dogman dan]

I'm confused icerider, I thought the boost converter was going to put out your target voltage, as long as the panels are putting out power. Much like the charge controller would, but as a naked bulk charger.

wattage would of course depend on the panels output of the moment, but why would voltage vary if it's going through the converter?

 

[icerider]

I'm confused icerider, I thought the boost converter was going to put out your target voltage, as long as the panels are putting out power. Much like the charge controller would, but as a naked bulk charger.

wattage would of course depend on the panels output of the moment, but why would voltage vary if it's going through the converter?

With the old I2C system (no BMS), I was tied directly into the battery so the circuit was: solar panels --> boost converter --> battery (10s5p). The boost converter was a little 100W CV unit with forced cooling just set to 41.7 volts. The little CV boost converters just do the best they can, based on input and output load. When my battery was low (say 3.5V/cell) the output from the boost converter was pulled down to 36 volts but it sourced 3.4 amps of reasonably smooth 36V DC into that load so ~120 W into the battery from the 200W of solar panels. A check on the other end showed that the panels had been pulled down below their max power point voltage and were only putting out 140W, even properly oriented in AZ noonday sun. As the battery got fuller, the voltage would go up and the efficiency got a little better. The best I ever measured was 150W into the battery with the panels oriented to noonday sun. As the battery approached full, the CV function capped the voltage at 41.7 and the cell internal resistance slowed the charge rate.

Now, with the new CanBus system, the battery is 13s4p but it has a BMS wrapped around it. I still think it would PROBABLY be safe to just bulk charge the battery brick, after all, the Bionx does regen VERY happily for braking and long downhill runs. I would tie either directly into the main circuit where it exits the battery or perhaps into the power line to the motor. Then I would just set CV cutoff just a little below what the batteries see at the end of a charge from the charger. I think I should also check the max current during a normal charge and use a CC/CV boost converter to be sure I stay WITHIN the voltages and currents that the BMS expects to see during normal charging (and/or regen). I have an old battery to check that on if I get around to it. What Fastolfe is doing is directly relevant to this approach since he is also charging "around" a BMS.

The other approach would be to deliver the charging current to the charge plug, but this has an added complication too. Unlike the I2C system where the charging voltage was applied directly to the battery brick, in the 48V CanBus system, the charging voltage is 26 volts and there is a (CC/CV ??) boost converter INSIDE the battery. The advantage to Bionx is that they are using a dead standard 26 volt 90W laptop charger instead of a custom CC/CV 48V charger. Hooking into the charger port seems possibly safer -- except -- in the shade, the solar supply will drop below the expected 90W and the charging circuit in the battery will pull the CC/CV boost circuit below 26V which is also probably FINE, but it NOT something the battery designer was expecting when he designed the charging circuit to run off a laptop power supply.

I will get around to getting the solar trailer tied back onto the Terratrike Cruiser, but for the present, my solar trailer spends most of its time hooked up as a grid tied supply which will pay for the grid tied controller in about 4 years, but it also happily ties into my Trek Valencia+, giving it an effectively infinite range at 14mph on level ground on a sunny day. Cruising along in a very comfortable moderate assist generating 90W and using about 150W, I run out of legs and backside LONG before the battery runs out of power.

 

[dogman dan]

Ahh, I must have misunderstood then. Of course wattage would vary, but better efficiency in ideal conditions should be tried for. I get very lost sometimes, with my wonderful agriculture degree. I'm still pretty much at the "hmm, smoke came out, buy a new one" level of electronics understanding.

It does make sense to me though, to just charge bypassing the bms. Especially if you can set your voltage to a slight undercharge for the maximum.

I would not hesitate to charge through the bms/charge plug of any other type battery. They don't freak out if you unplug the charger and they start powering the charger backwards.

 

[craneplaneguy]

All the MMPT controllers I've dealt with in off grid solar systems, take higher voltage, some up to 600 VDC, and then reduce it DOWN to the battery voltage. Doing so allows the array to be much further array from the battery while using much smaller cabling, so when I first read your post I was ready to say no way, until I clicked on the link to the Genesun and saw it boosts UP not down, very cool! The sink analogy is perfect, Dogman. One thing though, a MPPT controller, whether boosting up or down in voltage, is of great benefit ALL the time, even when, especially when, the system voltage is well below the regulating point. They constantly juggle the incoming voltage and current in a way that boosts the efficiency of the PV module, letting it run in it's sweet spot, rather then lugging down, to use a mechanical analogy.

I was living off grid for 28 years when the first generation of MPPT controllers came out, and it was almost too good to be true, the same 1200 watt array once wired through the MPPT (in my case 24 VDC array/24 VDC battery bank, so no boost either way) immediately starting producing 10 to as much as 25% more current into the battery! Cold weather, and a low state of charge on the battery produced the highest gains, and that was great because that was exactly when I needed the extra power the most. Thanks for bringing these MPPT voltage UP boosting controllers to my attention, and I agree, it should work just fine for you. One thing though, when I buy a pallet of modules for a large grid tie installation for a customer, my primary criteria is price per watt, overall efficiency doesn't matter much if it's cheaper to just buy another module or two. The only time it makes sense to go with a more expensive per watt module is in outer space (the most efficient PV modules "in the world", out of it actually, are used by NASA) or in a case like yours with limited room. So expect to pay a premium price for a highest efficiency panel, they are out there, I just never deal with them as we have plenty of room in Idaho for a few more panels if it keeps the price per watt down. Here's a pic of a recent 3780 watt array I fabbed up for a grid tie job, 12 modules @ 315 watt each! Voltage coming out was over 400 VDC, before the grid tie inverter knocks it down to 240 VAC. Including the rack it weighed about 1300 lbs and now sits on a 8" pipe along with two other arrays, THIS would charge your bike!

 

[Fastolfe]

[...]One thing though, when I buy a pallet of modules for a large grid tie installation for a customer, my primary criteria is price per watt, overall efficiency doesn't matter much if it's cheaper to just buy another module or two. The only time it makes sense to go with a more expensive per watt module is in outer space (the most efficient PV modules "in the world", out of it actually, are used by NASA) or in a case like yours with limited room. So expect to pay a premium price for a highest efficiency panel, they are out there, I just never deal with them as we have plenty of room in Idaho for a few more panels if it keeps the price per watt down.[...]

On a solar vehicle, the main concerns are efficiency and weight. Whether I stick the panel on top of my velomobile (which I probably won't do for various reasons) or I make a dedicated trailer, I'm limited to a surface of 23 sq.ft by law (or twice that if I do both). So I have to make the most of that surface. With that to work with, I can put together a panel with monocrystalline cells that would theoretically put out 350+Wp. But it would weigh over 40 lbs. If I use flexible panels, it would only put out 250 Wp, but it'd only weigh 5 or 6 lbs. As for price, well, it'll be expensive no matter what.

 

[craneplaneguy]

Yes indeed, rereading my earlier post it sounds like I'm advocating cheaper per watt modules for your setup. What I meant to say is what you just more succinctly wrote. Price be damned, top efficiency is what you need for your limited size " array rack". I see Panasonic is getting some buzz on a new module with 22% efficiency, doesn't seem to be on the market yet though. Let us know what you come up with.

 

[douglashart]

ICERIDER , I saw your thread a rear ago and I was all grins , never got around to thanking you for, what was a breakthrough for me . Just finished completing my bike build . now a solar trailer! ill give a call and thank you personally when I'm out touring the desert

 

[icerider]

ICERIDER , I saw your thread a rear ago and I was all grins , never got around to thanking you for, what was a breakthrough for me . Just finished completing my bike build . now a solar trailer! ill give a call and thank you personally when I'm out touring the desert

Thank you. I had an absolute ball doing the solar cruiser project. I got a lot of good starting-point ideas from ES and then just tried stuff until it worked.