Sealed Lead Acid vs Lifepo4 Experiment

[cruiserbikeman]

I have an electric tricycle that is powered by four 12 volt 12ah sealed lead acid batteries connected in series which equals 48 volts 12ah. I decided to switch to a 48 volt 15ah lifepo4 battery and use different components (controller,throttle,etc.) but use the hub motor (450 watts) that came with the tricycle. Connected everything and the motor spins like a 48 volt connected battery is supposed to. The problem with a lifepo4 battery is that even though it weighs only 16 1/2 pounds it just doesn't have the pulling power the sealed lead acid batteries have. The speed was much slower with the lifepo4 as well even though the sealed lead acid battery weighs over 40 pounds. Another thing is the motor gets hot with the lifepo4. Now I use my tricycle to carry heavy groceries (gallon milk, gallon liquid detergent, gallon bleach, 4 pound bag sugar, 5 pound ground beef, candy, 4 21 ounce bags cereal, snacks, napkins, cleanser) with the sealed lead acid batteries and the motor stays cool and is faster than if I was riding the tricycle with no groceries using the lifepo4 battery. I bet if I tried that with the lifepo4 the motor would overheat and go dead. Sealed lead acid batteries tend to get a bad rap for limited distance, but there is no contest when it comes to transporting heavy loads. So from my experience, sealed lead acid is the choice for carrying cargo while lifepo4 is good for two-wheeled bike riding.

 

[amberwolf]

Has nothing to do with the chemistry of the battery. And you also cannot draw a direct conclusion about a battery swap if you also swap any other component in the system, unless they are all exactly identical and wired identically.

The motor heating is probably because of the different controller having different limitations or methods of driving the motor, vs the original one used with the SLA,

--Or--

if the motor is brushless sensored, the motor cotnroller is probably hooked up with a "false positive" combination that causes excessive current draw.


The "pull" problem is probably because either:
--if the motor is connected incorrectly, it probably wont' have the pulling power it should because it's not being controlled correctly.
or
--the cells in the pack sag in voltage a whole lot at the probably too-high current the controller is pulling from them (the same thing that is probably causing the motor to heat up), so that you don't have as much total power available compared to the SLA (because the SLA controller isn't allowing as much current pull so they aren't sagging, and/or they are able to supply more current without as much voltage sag).
or
--the BMS in the LiFePO4 is limiting it's current output.


I'm not sure why you would change the controller and throttle and stuff if you are using the same voltage pack. If your LiFePO4 pack is 16s, then it's fully charged voltage should be about 58.4 to 59V at most. Fully charged SLA of 4x12V should be about 54.4 to 57.6, depending on your charger. The original controller probably has 63V caps in it, dunno what the FETs are, but if they are rated high enough for the SLA they *should* be rated high enough for the LiFePO4. You'd have to open it up and find the p/n on the FETs adn then a spec sheet on the web to know for sure.

However, if your newer controller works on the LIFePO4, you could easily plug the SLA into it, and see what difference you get using the *same controller, motor, throttle, etc* and the battery being the only difference.

My guess is you will see the same behavior using SLA on the new controller as you do the LiFePO4, and that if you used the LiFePO4 on the old controller/etc you'd see it works about teh same as the SLA.

If it doesn't, then there is probably something wrong with the LiFePO4 pack or it's simply not designed to high enough capabilities to run the controller and motor.

 

[cal3thousand]

Has nothing to do with the chemistry of the battery. And you also cannot draw a direct conclusion about a battery swap if you also swap any other component in the system, unless they are all exactly identical and wired identically.

The motor heating is probably because of the different controller having different limitations or methods of driving the motor, vs the original one used with the SLA,

--Or--

if the motor is brushless sensored, the motor cotnroller is probably hooked up with a "false positive" combination that causes excessive current draw.


The "pull" problem is probably because either:
--if the motor is connected incorrectly, it probably wont' have the pulling power it should because it's not being controlled correctly.
or
--the cells in the pack sag in voltage a whole lot at the probably too-high current the controller is pulling from them (the same thing that is probably causing the motor to heat up), so that you don't have as much total power available compared to the SLA (because the SLA controller isn't allowing as much current pull so they aren't sagging, and/or they are able to supply more current without as much voltage sag).
or
--the BMS in the LiFePO4 is limiting it's current output.


I'm not sure why you would change the controller and throttle and stuff if you are using the same voltage pack. If your LiFePO4 pack is 16s, then it's fully charged voltage should be about 58.4 to 59V at most. Fully charged SLA of 4x12V should be about 54.4 to 57.6, depending on your charger. The original controller probably has 63V caps in it, dunno what the FETs are, but if they are rated high enough for the SLA they *should* be rated high enough for the LiFePO4. You'd have to open it up and find the p/n on the FETs adn then a spec sheet on the web to know for sure.

However, if your newer controller works on the LIFePO4, you could easily plug the SLA into it, and see what difference you get using the *same controller, motor, throttle, etc* and the battery being the only difference.

My guess is you will see the same behavior using SLA on the new controller as you do the LiFePO4, and that if you used the LiFePO4 on the old controller/etc you'd see it works about teh same as the SLA.

If it doesn't, then there is probably something wrong with the LiFePO4 pack or it's simply not designed to high enough capabilities to run the controller and motor.

I would tend to agree here. Even the throttle could be giving him a difference. He's only using a butt-dyno and not giving any solid quantitative measures.

 

[cruiserbikeman]

Has nothing to do with the chemistry of the battery. And you also cannot draw a direct conclusion about a battery swap if you also swap any other component in the system, unless they are all exactly identical and wired identically.

The motor heating is probably because of the different controller having different limitations or methods of driving the motor, vs the original one used with the SLA,

--Or--

if the motor is brushless sensored, the motor cotnroller is probably hooked up with a "false positive" combination that causes excessive current draw.


The "pull" problem is probably because either:
--if the motor is connected incorrectly, it probably wont' have the pulling power it should because it's not being controlled correctly.
or
--the cells in the pack sag in voltage a whole lot at the probably too-high current the controller is pulling from them (the same thing that is probably causing the motor to heat up), so that you don't have as much total power available compared to the SLA (because the SLA controller isn't allowing as much current pull so they aren't sagging, and/or they are able to supply more current without as much voltage sag).
or
--the BMS in the LiFePO4 is limiting it's current output.


I'm not sure why you would change the controller and throttle and stuff if you are using the same voltage pack. If your LiFePO4 pack is 16s, then it's fully charged voltage should be about 58.4 to 59V at most. Fully charged SLA of 4x12V should be about 54.4 to 57.6, depending on your charger. The original controller probably has 63V caps in it, dunno what the FETs are, but if they are rated high enough for the SLA they *should* be rated high enough for the LiFePO4. You'd have to open it up and find the p/n on the FETs adn then a spec sheet on the web to know for sure.

However, if your newer controller works on the LIFePO4, you could easily plug the SLA into it, and see what difference you get using the *same controller, motor, throttle, etc* and the battery being the only difference.

My guess is you will see the same behavior using SLA on the new controller as you do the LiFePO4, and that if you used the LiFePO4 on the old controller/etc you'd see it works about teh same as the SLA.

If it doesn't, then there is probably something wrong with the LiFePO4 pack or it's simply not designed to high enough capabilities to run the controller and motor.

The new controller draws 22 amps while the old controller is limited to 15 amps. I wanted to change the controller because I would have to rewire the connectors for the sla battery (it uses a 3 prong plug as opposed to andersons). As I mentioned the speed the hub motor spins is the same with either battery. The difference is when you add weight the sla performs better than the lifepo4.

 

[cruiserbikeman]

The real question is why is a sla battery that is heavier and using a weaker controller (15 amps limit) able to outperform a stronger lighter lifepo4 battery that is using a stronger controller (22 amps)?

 

[spinningmagnets]

SLA are known to provide massive amps when full, thats why they are used on cars for starting. Get them half empty and they sag badly. They should only really provide half the capacity that they are advertised for. A 12-Ah SLA should really only be used for 6-Ah worth of range. The plate surface area that is available for the chemical reaction drops in a fairly straight line.

LiFePO4 has a low C-rate, but if you get a pack large enough to provide your desired max current draw, they will work well, all the way down to 90% empty. Since you only list the motor as a 450W, I'm going to assume its a fast wind direct drive motor on a 26-inch wheel, so you could hopefully get over 20-MPH on 36V on flat ground. If you are running slow with a full cargo load, the motor will get hot.

You might benefit from a smaller diameter wheel (24-inch, or a 20-inch wheel with a 22-inch OD moped tire), or if you need to keep a 26-inch wheel, possibly try a low turn-count geared hub. cell_man has a 10T, and an even slower 12T that should work well for a cargo bike at 48V using a 15-Ah LiFePO4 limited to 20A by the controller.

 

[Russell]

The real question is why is a sla battery that is heavier and using a weaker controller (15 amps limit) able to outperform a stronger lighter lifepo4 battery that is using a stronger controller (22 amps)?

The real question is "what's wrong with your LiFePO4 set-up?", because what you describe should not happen. Years ago when I first switched from 36V/9Ah SLA to 36V/6Ah Lithium the smaller, lighter Lithium pack outperformed the SLA pack in every way with lighter weight, higher average voltage (resulting in better top speed), less voltage sag under load and more total capacity. Everyone else who has made the switch will tell you the same thing so as the previous posters have already said there must be a problem with your battery (weak cells, bad BMS, or ?) or your controller.

-R

 

[amberwolf]

Or, based on the motor getting hot, could also be a false positive phase/hall combo. If you have even just a small cheap multimeter with a 10A amps setting, you can put that between one battery lead and it's corresponding controller lead, and do off-ground wheel testing to see what the no-load current is. If it's more than an amp or two at full throttle, then it is probably a false positive phase/hall combo.

The motor may spin at the same *speed*, but that only means that it is getting about the same average voltage, and not really anything else.

The fact that the controller is a higher-amp controller also probably contributes to the motor heating, as it isn't going to limit the current as low as the other one. If it's 20A, then at even the "discharged" or "loaded" voltage of 48V-ish, it's still putting 960W, over twice the rated power, into that 450W hubmotor. At the higher voltage of 54V you're more likely to be at if the battery isn't sagging much, you'd have 1080W. So no surprise it is getting hotter, if it is at those power consumptions very much/often.

The older 15A controller in the same situation is only 720W to 810W or so, which isn't quite as much excess power.

It doesn't mean that the power itself is a problem, as many "500W" hubs like my 9C can handle much higher power levels for a long time, some of htem double to quadruple that, without any special cooling mods. (though they will get hot).

But how that power is delivered may be a problem in overheating the hubmotor without doing as much useful work. If the motor is not saturating at whatever phase amps the old controller can put out, then it won't get as hot when doing it's work. But if the new controller is able to put out much higher phase amps (which have little to do with battery amps, whcih is how the controller is rated/labelled), it could be pushing the motor way past saturation, so that it is making lots of heat with that current, and not doing any more work than it did with the old controller. Might even do less work, because it's getting very hot inside, very quickly, the winding resistance goes up, etc. Also if the higher phase amps cause the controller to try to limit battery amps, it may cut down the voltage being fed to the motor, which while this will reduce phase amps it may still be beyond saturation and thus wasting more of that power than it is converting into work...and it may be doing less actual work because the voltage is less (not sure about this), making you feel even less "pull" from it.

Just a thought (whcih might need rewriting to make sense).


It is also possible that this particular LiFePO4 battery (like many other batteries of all descriptions that are often "rated" high but have low-spec parts/cells) is not very powerful.


But the conclusion drawn in the OP is not correct for anyone except the poster:

So from my experience, sealed lead acid is the choice for carrying cargo while lifepo4 is good for two-wheeled bike riding.

because I have the opposite experience: SLA worked ok for my heavy-cargo hauler CrazyBike2, but each succeeding chemistry (including the relatively-low-power had-to-be-repaired no-longer-fully-capable Vpower/CammyCC LiFePO4) tried on it (and DayGlo Avenger, my medium-cargo hauler) has been better performing than the previous one, where performance is rated in several categories.


I definitely agree that the one thing that SLA can do very well is output lots of current, which can certainly produce a lot of torque...but they do it at the cost of voltage sag and reduction of lifespan. They do work for driving motors, and can work for ebikes/trikes. But by most standards they're not as good as a *good* lithium-based battery can be.

 

[cruiserbikeman]

You might benefit from a smaller diameter wheel (24-inch, or a 20-inch wheel with a 22-inch moped tire), or if you need to keep a 26-inch wheel, possibly try a low turn-count geared hub. cell_man has a 10T, and an even slower 12T that should work well for a cargo bike at 48V using a 15-Ah LiFePO4 limited to 20A by the controller.

The bike uses 22-inch wheels.

 

[cruiserbikeman]

I bought the electric tricycle (Model EW-88) from an online electric bike retailer. It was already fully assembled. It's good for what it does which is haul cargo.

 

[The fingers]

I have only about a year/2500 miles on my kit with SLAs and I like them so far. I am somewhat low tech but am learning much here. For my first time battery replacement I'll probably put 15AH ub12150 to replace the 12AH ub12120 and continue my current riding/charging habits. Thanks for the information about your experience for when the next pack needs replacement, as I am planning to add a cargo trailer in the future.

 

[dnmun]

i thought lifepo4 was cheaper than lead now. you can buy the big pouch packs for $.50/kWh.

how much do you pay for a 1kWh lead acid pack? how much does it weigh? i assume you mean SLA, not deep discharge marine battery.

 

[The fingers]

Lowest drop in like for like replacement of the stock batteries for the pack is around $60 delivered.

 

[spinningmagnets]

So...you are using a different controller, but the same motor. Are you certain that the plugs on the new controller are hooking up correctly? How many pins on the motor plug? (I'm hoping the number of pins will show if it uses hall-sensors, or is sensorless). Try what Amberwolf said about checking no-load current with the hub-wheel in the air. 48V in a small wheel should pull pretty hard.

Pic shows a front hub, description details a rear wheel hub.

 

[Rassy]

I thought SLA was great to, back in 2007 before using LiFePO4.

The only question I have is your reluctance to try an apples to apples test as has been suggested by others. At this point you are not only potentially mis-leading others, but also making the wrong conclusions for yourself.

As far as "changing" the connectors you could just make up a simple jumper that allows you to plug in the other battery. Someone on the forum recently pushed his incorrectly wired controller/motor phase and halls until other damage was done, so you need to figure out why your system isn't working correctly.

Let us know what was going on when you find out. Good luck.

 

[iamsofunny]

I've used 2 SLA packs and 2 Lifepo4 packs and the lifepo4 delivers much more torque, climbing ability and speed.

 

[The fingers]

The reason I would consider upgrading would be to increase range without adding more weight.

 

[dnmun]

The reason I would consider upgrading would be to increase range without adding more weight.

how much do you pay? you said 12V 15Ah for $60? is that really correct? $60/180= $.33/watt hour and add cost of charger?

but you can buy lifepo4 packs which last 10 times as long and produce twice the power at a higher voltage, for $.50/watt hour and the charger is included.

how can reality be ignored?

maybe back in the days when lithium ion power was expensive and lead cheap. now it is just the opposite, lead is expensive and lithium ion is cheap.

 

[The fingers]

If I save money in the process, great!

 

[dogman dan]

Well, there is several possibilities here, given that several things were changed at the same time.

The first thing I'd want to try, is going back to the original controller and throttle, and see what happens there. Maybe your wires are not in the right order on the new controller, ( assuming it's brushless, is it?) or maybe something is wacky with the new throttle, and you only get 75%.

The other possibility is that it IS the battery. The problem may not be that lifepo4 sucks, but that you bought a cheapo from and ebay shark, and it's sagging like grannies tits. Lower voltage because half the cells are dead meat and you are flogging the hell out of it causing it to sag almost to the lvc, or perhaps past it if the bms is equally crap.

Keep experimenting. But set yourself up to change just one variable at a time. I'll be interested to hear how the bike runs with the current controller and throttle, but the old lead running it too.

Put a voltmeter on it, so you can see if the battery voltage sags like mad when you use the throttle. More sag than 2-3 v from 15 ah lifepo4 aint normal. Find out if your charger works. If your charger only puts out 50v, there's your problem, or one of multiple problems anyway.