Best inexpensive way to enforce a low maximum amp limit

[rocwandrer]

I want a low maximum amp limit, but i want reliability. If i run a big hub motor, with a big controller, but want max efficiency as a higher priority than max power, it seems like enforcing a low maximum current limit is an excellent way to do that.

I'm thinking of getting the BMSBattery.com 12 mosfets controller. I know people mod these for more amps... How hard/expensive is it to setup a 3 speed controller with a low amp setting, or is there a better way?

Richard

 

[itchynackers]

Cycle Analyst.

 

[ZOMGVTEK]

It's essentially free to lower or raise the current limit on most any controller.

If you want lower, remove all but 1 shunt, and grind it thinner. Or you can remove the shunts and replace it with thinner wire and see where that takes you.

Do you want a switch to be able to adjust the current limit on the fly? That can be a bit more tricky. Having a standard 3 speed switch with a low current limit isnt really necessary.

 

[rocwandrer]

It's essentially free to lower or raise the current limit on most any controller.

If you want lower, remove all but 1 shunt, and grind it thinner. Or you can remove the shunts and replace it with thinner wire and see where that takes you.

Do you want a switch to be able to adjust the current limit on the fly? That can be a bit more tricky. Having a standard 3 speed switch with a low current limit isnt really necessary.

The cycle analyst seems like a pricey option that offers a lot of features that aren't (yet?) on my priority list. I don't need a switch on the fly, I was just wondering if exploiting the circuit for the switch would be the most idiot proof way of enforcing a lower maximum current limit. There are multiple shunts? Are they parallel wired?

 

[neptronix]

shunt shave or program it.

or even more elegant: you can replace the shunt(s) with ones designed to produce lower current. ( typically smaller )

 

[Drunkskunk]

the current in the shunt is sensed by way of a small resistor. Change the value of the resistor, change the value of the current limiting. total cost, $0.10. or you could get fancy, and add a switch that changes the value of the resistor so you could have multiple current settings. you could then have safe, efficent goverend power most of the time, and an "Oh $H!T BIG DOG IS CHASSING ME" button for when you need more power.

 

[rocwandrer]

When I try to use the ebikes.ca simulator to model a 7A controller, it works just fine with high winding count, slow motors, but doesn't work at all for lower winding count, faster motors (edit: this seems to be fixed). In some cases it looks like a division by zero error, in others it just ignores my controller and gives the motor 40 amps max. Is this in some way linked to a real phenomenon? I notice the faster motor is sending way more amps to the motor at low amps than the slow wound motor. This seems to imply the faster wound motor is working the controller a lot harder at the same (low-ish) speed compared to the slow wound motor?

Does this mean, at low amps, I'm better off with a low speed wound motor from a controller reliability standpoint?

 

[biohazardman]

Lower count winds on motors work well for higher amperage and from a hot rod point of view. Yes higher wind count motors will be what you want and easier on controllers as well. So a low power motor and a low amp controller will work for you. Most want high power stuff and it is more expensive battery, motor and controller wise. Just remember though to low a speed with any kind of a load will still heat things up.
You can choose the 20A limit controller and then don't give it more than 75% throttle to stay within 7A but hit some hills or crank up the throttle and it will go higher unless you are peddling the extra it needs to stay within the limits you have made. To be certain you do not go over the limit you will need to mod the controller or as others have stated use a CA.
What I did not say was motors are designed and wound to operate most efficiently at a given speed. Operate them over or under that speed and you loose efficiency and make heat.

 

[Alan B]

Lower wind motors have VERY high starting phase currents. Small controllers will likely perform poorly or let the smoke out. For efficiency and reliability get a controller that is larger than you need, and program the current limit down to a lower value.

 

[John in CR]

Just like getting better mileage in your car, simply ride in a more efficient manner if efficiency is your priority. Judicious use of the throttle and ramping it up slowly during acceleration go a long way to greater economy, and just like with a car it's good to have more go when you need it.

It requires a certain power to ride at a certain speed, so for a given battery voltage there's a given current required unless your load is higher. No controller mod or current limiting will change what is required, and it could even have a negative effect.

Slow down, pedal more, use a lower more aerodynamic bike and riding position, and put more air in your tires. Those will all have tremendous impacts on economy.

 

[Alan B]

I use a Magura throttle with a Lyen 12 FET controller and a 9C 6x10 on 72 volts. I can crawl the wheel around at speeds too low to even balance easily. The control is very fine.

Efficiency is mostly about the speed you choose. Air resistance is the major factor. Another is gradient, choose routes that are level for tremendous energy savings.

I use the Cycle Analyst so I can see where the efficiency is and how the batteries are doing. It also makes it easy to change a current limit in the field instead of having to reprogram the 12 FET controller. I can also set a speed limit and then put the throttle to maximum and let the limit take over.

 

[dogman dan]

I'd just get a lyens controller set to 20 amps. I like my 9 fet one real good for efficeint, but 72v riding. Then your other controller can be a spare, or build a less efficient bike. Limiting to 20 amps means no 40 mph riding, about 35 max. But you also use a smidgen less on stop signs.

But the reality is, if you really want efficiency, all you need is to slow down to sub 20 mph. You need a speed limiter more than an amp limiter really.

 

[nicobie]

But the reality is, if you really want efficiency, all you need is to slow down to sub 20 mph. You need a speed limiter more than an amp limiter really.

How true!

 

[d8veh]

The BMSbattery 12 FET controller already has a connector for a three speed switch. Just buy the switch and connect it. It only costs $1. It gives three power settings: low, medium and high. IIRC it only works on the throttle, not the pedal sensor.

 

[rocwandrer]

I'm looking at assist as something the motor gives me occasionally, not as something I give the motor occasionally. I own a car, but I haven't used it to get to work in 6 weeks or more.

I don't envision EVER using the throttle for acceleration. I guess I will technically be accelerating in the rare instance that I need not the extra power, but the extra traction from the front wheel to get through the dead spot and stay moving in deep snow. Even then, my experience is that once the bike is moving fast enough to have a smooth cadence, I don't need the extra grip from the front, and I expect to only want a little bit more power than I can put down myself in that rare circumstance (I usually can pick a plowed path, or the woods, if I pick the woods, it is because I'm feeling up to pedaling through the snow or whatever without assistance from the motor). To put it in perspective, I've found myself absolutely having to stop to strip off a layer of insulation in winter, then having to tramp a packed line in the snow in front of the bike in order to give me a place to get the bike moving again. Pretty far removed from the average user here, I'm guessing.

My perception from the ebikes.ca simulator is that a low amp limit reduces the total power available at any current limited rpm, does not reduce the peak rpm limit, and increases the efficiency at lower rpm a lot. Basically, without buying anything fancy or having to guess on throttle position, I can be assured that about 70% of the energy coming out of the battery pack is going towards driving the bike forward. That's what I want. So my plan is to get a motor rated for more power output than I want, and use a current limit to enforce lower power than the motor is capable of, but higher efficiency than a smaller motor would be capable of.

It seems to me that a 500w motor and 35 amp controller system limited to say 8-12 amps should:

weigh and cost only a tiny bit more than a smaller motor and controller

offer more power at low rpm than a smaller motor even with the same amp limit because of better efficiency

give more total assist power for the same batteries because of higher efficiency

offer better reliability than a smaller motor with a smaller controller set to the same amp limit

let me use the throttle to get some assistance for my pedaling power, without worrying about the optimal throttle position for efficiency (for a wider range of speeds than a smaller motor with the same amp limit)

Does that all seem reasonable, if we assume for a second that my stated goals and usage are reasonable?

 

[miuan]

You definitely want a geared motor. It won't even need to be strong, they all have a nice broad eff curve once you get to 48-60V. Using a lot of your own pedal power, do consider the weight of motor and controller. The ultimate solution would be a Q85 plus KU63 with shunt ground down to 8A and 12-15S lipo. Efficient, stealth, cheap and light!

 

[Drunkskunk]

A Q85, AKA, the Cute motor sure is cute. But it would have trouble pulling you up a wheelchair ramp.

your goal is reasonable, but your perception may be off. Decreasing the amp limit doesn't make the motor more efficent. it reduces the power available so you use less, but you have less available. If the job requires 200 watts of power, the motor will make 200 watts of output Only if it has a suffecient amount of input. If you decrease that amount, it just won't be able to do the job.

you can't make the motor run more efficently, but you can use the motor in more efficent ways. I.E., only turning the thriottle enough to get the desired amount of power out of it. You can feel working as easily as you can gauge how much power your legs are giving you.

 

[rocwandrer]

You definitely want a geared motor. It won't even need to be strong, they all have a nice broad eff curve once you get to 48-60V. Using a lot of your own pedal power, do consider the weight of motor and controller. The ultimate solution would be a Q85 plus KU63 with shunt ground down to 8A and 12-15S lipo. Efficient, stealth, cheap and light!

I was thinking Q128 for the rear spacing that I built my front fork with, and the freewheel thread (another project down the road that has nothing to do with assist/per se). I'm worried that the KU63 wont handle the duty cycle I'm asking of it, so considering the couple dollars and oz extra for the bigger controller (provided I can work out changing the current limit...)

So at higher than intended voltage, the efficiency curve is broad, but all at higher than the original top speed, and I'm mostly concerned with lower speeds. I think for now at least, I'll stick to 36v nominal. But if you crank up the voltage and use Lipo packs, how do you handle low voltage protection?

 

[rocwandrer]

A Q85, AKA, the Cute motor sure is cute. But it would have trouble pulling you up a wheelchair ramp.

your goal is reasonable, but your perception may be off. Decreasing the amp limit doesn't make the motor more efficent. it reduces the power available so you use less, but you have less available. If the job requires 200 watts of power, the motor will make 200 watts of output Only if it has a suffecient amount of input. If you decrease that amount, it just won't be able to do the job.

you can't make the motor run more efficently, but you can use the motor in more efficent ways. I.E., only turning the thriottle enough to get the desired amount of power out of it. You can feel working as easily as you can gauge how much power your legs are giving you.

We may disagree on philosophy, but am I wrong on any of the electric motor/controller/battery/efficiency theory?

Here is my thinking:

When I am climbing a hill, don't know how fast I am going, and adjust the throttle, how am I going to regulate throttle position for max efficiency? I'd rather (for the same pedaling power) drop 2 mph than take a 10% efficiency loss and load down the battery that much harder. My thumb might disagree at the time, pissing off my legs later in the trip when the battery is out of juice

It appears to me (and if I am wrong, I sure want to know) that if the motor is calling for 25 amps and the controller is current limited to 8 amps, the controller cuts the duty cycle and gives the motor only 8 amps times 36 volts of power by giving it something like 25 amps times a 33% duty cycle. If that isn't enough to maintain speed, the bike slows down, but because of the enforced max amps, I will (for a very broad rpm range) have 36V x 8A of power input to the motor.

With the Q128 wound for a 17 mph top speed and a controller modified to have an 8 amp current limit, I'd expect that 100% throttle would give me 200w of motor output from say 7 mph all the way up to 15 mph. Because the duty cycle is like throttling back, the motor efficiency goes up compared to a higher amp limit for the same rpm and max throttle, so through that entire range, the battery is seeing around a 288w draw (8 amps at 36v). The throttle lets me use less when less is called for, but in effect I'm accepting a 200w maximum output level available at any time in exchange for more effective use of the power in the batteries. If at 7 mph the motor could give me 300W (an extra 100w!!!) but would suck 630w out of the battery to do it (an extra 342w!!!), well, I'd rather limit the output to 200w.

If I just use a smaller motor that is rated for less power to get the 200w at 7 mph (if the smaller motor can even deliver that much power turning that slow) I'm probably going to be using way more than 288w of battery power to do it.

Flaws in my logic?

 

[liveforphysics]

I want a low maximum amp limit, but i want reliability. If i run a big hub motor, with a big controller, but want max efficiency as a higher priority than max power, it seems like enforcing a low maximum current limit is an excellent way to do that.

I'm thinking of getting the BMSBattery.com 12 mosfets controller. I know people mod these for more amps... How hard/expensive is it to setup a 3 speed controller with a low amp setting, or is there a better way?

Richard

Limiting current on a large hub won't give you efficiency. You actually find maximum efficiency at the point copper loss matches all other combined losses. For most larger hubs, this peak efficiency point doesn't occur at under a KW of power or more.

If you actually care about efficiency over power, get a small geared hub and power it properly.

If you just want to reduce battery current for any controller, increase the resistance of the current measuring shunt. IE, if it has two shunt bars, snip one. If you want further reduced current, snip one and dremel sand it thinner.

However, if you want efficiency, size your motor for the task and run it properly. Starving an excessively large motor just makes for lots of core-loss waste.

 

[rocwandrer]

I want a low maximum amp limit, but i want reliability. If i run a big hub motor, with a big controller, but want max efficiency as a higher priority than max power, it seems like enforcing a low maximum current limit is an excellent way to do that.

I'm thinking of getting the BMSBattery.com 12 mosfets controller. I know people mod these for more amps... How hard/expensive is it to setup a 3 speed controller with a low amp setting, or is there a better way?

Richard

Limiting current on a large hub won't give you efficiency. You actually find maximum efficiency at the point copper loss matches all other combined losses. For most larger hubs, this peak efficiency point doesn't occur at under a KW of power or more.

If you actually care about efficiency over power, get a small geared hub and power it properly.

If you just want to reduce battery current for any controller, increase the resistance of the current measuring shunt. IE, if it has two shunt bars, snip one. If you want further reduced current, snip one and dremel sand it thinner.

However, if you want efficiency, size your motor for the task and run it properly. Starving an excessively large motor just makes for lots of core-loss waste.

But I have no interest in peak electrical efficiency. I have an interest in having a certain level of assist over a broad rpm range, and getting decent efficiency at each rpm in that range.

Are you saying the smaller motor, rated for say 250w peak, and giving ~75% net efficiency at 80% of the no load speed and perhaps 100w output at that speed, can give me 70% or so efficiency and 200w at 40% of the no load speed?

Are you saying the 500w rated motor throttled back (either with the throttle, or with the current limit on the controller) can't give 70% net efficiency at 40% of no load speed and still put out 200w of output power?

You see my point? What's the error in this thinking?

 

[ZOMGVTEK]

Just for some sort of reference... 26Wh/Mi is fairly typical for my bike in eco mode. 1000W Golden Motor on 24S LiPo, speed limited to 38% throttle, 22 mile trip to work. The identical roundtrip route, with my co-workers Q100 350W 6S kit, nets about 24-25Wh/Mi. My bike cooks around at 16-17 MPH average in these conditions, and his more efficient geared Q100 goes closer to 18-19 MPH average. Something like 18 on the flat for my bike, 21-22 on flat for the Q100. My 'big' DD hub, driven by 24S LiPo, is less efficient, going slightly slower, than his tiny little geared 350W hub. This is obviously with my controller speed limited. Its set to 60A, and will draw about 3-4kW going up a steep hill, whereas his bike stops and you need to pedal.

Most of the difference could simply be that one steep hill that the Q100 requires additional power to go up any real incline, forcing you to pedal. My bike has no chain, so I never ever pedal. Using my bike how I prefer to, is something like 45-50Wh/Mi, and a bit over 30 MPH average speed. Obviously this is wasteful use, but its roughly double the energy required to do the same thing, all to go twice as fast.

 

[amberwolf]

Since essentially all you are looking for is to conserve power, and you are going to limit the total amount of power you want out of the motor anyway, you would be best to pick a motor designed to only ever output that much to start with, at the RPM you typically expect to be at when needing that max power (meaning, up a hill, probably, since you're not wanting it for startup acceleration). It's more efficient (AFAIK) than using a much larger motor to far less than it's potential. (it is certainly *safer* to use a larger motor under it's potential, in that you can never overheat it, but not likley more efficient).

For that 7A you previously mentioned, you won't need more than a 6FET controller at most; a 12FET is much larger and likely more money than you need. The 6FET will start with a lower current limit anyway, making it even easier to modify downward to 7A. It can be done physically by removing shunts, or by programming it. If there are three shunts, well, many 6FETs are only 20A controllers, so removing two shunts would give you about 6.7A limit right there. If only one shunt, you could either unsolder one end and solder two more shunts in *series* with it, or shave out 2/3 of it's thickness. Programming would be easier than any of that.

Same with motors--you don't need much of a motor to only pull 7A max. The 9C 2807 on CrazyBike2 takes about 7A to hold me at 20MPH, IIRC, without pedalling, with a nominally 48V LiFePO4 pack (actual 52-56V in the most usable range, IIRC). But that's a lot of motor for a bike that's heavy/big and no pedalling, whcih takes 60A+ to accelrate from a stop. You don't have nearly those requirements, so you don't need that much of a motor. Either a small DD or geared hub would do for you fine, in the 200-250W range max.


If you already have a Cycle Analyst, you could use it's current-limiting features rather than those of the controller, so that you can change them "on the fly", for experimentation and dialing things in while out riding, rather than havng to come back to the workbench to change the controller settings (electronically or programmatically).


I don't think you're going to get any more power at low RPM with the larger motor/controller than a smaller set, because your amp limit is the same. If your *wind* is different, then yes, you might, but it doesn't necessarily have anything to do with it being a "bigger" motor or controller (being capapble of higher total power). It probably would not be more efficient, either, but rather less so, as Liveforphysics mentioned (core losses/etc.).


As for being "assured" that 70% (or any particular number) of the power from the pack makes it to the ground, I'm not sure it's possible to guarantee that, with just a current limit. It might be more complex than that, but I am not a math person so I'll have to leave that to someone else to help with. It doesnt' "feel" like it would work that way, though.



One thing also to think about, is that for best efficiency, you want to pick a motor that is designed to operate at the RPM (and voltage) you want to be applying most of your power at. This also means it's BEMF will be optimal for that RPM and voltage to help negate that current, ensuring less power is simply wasted as heat. (I think...again not the math guy)

 

[Russell]

If you want efficiency simply use the motor sparingly, use a light touch on the throttle and keep your speed down. This will work whatever set-up you choose though I would suggest a small geared motor to keep the weight penalty down and because they spin freely without the drag of a direct drive. If you're buying a new motor and if speed isn't important then get a slower wound motor since they make it easier to modulate the throttle for low speeds.

As others have mentioned a simple way to adjust the current limit on your existing controller is to add or remove solder from the shunt. I tweaked a "22A" controller down to 13A and up to 27A by this simple and free method. The thing is though it never made much of an impact in overall battery life for me since most of the time I follow the rule above; "use the motor sparingly, use a light touch on the throttle and keep your speed down".

If you are going with a new controller then opt for the 3-speed switch which allows you to set three different current limits. You can get controllers with this feature quite cheaply from China.

-R

 

[dogman dan]

Yeah, you are clearly not getting it. Slow on the flats is going to be efficient. Hills take what they take, less watts is going to actually kill your efficiency on the hills if it is less than the laws of physics say is required. x weight lifted y height = z watts.

Say you ride 20 mph. Take a huge motor and a 10 amp controller. At 36v, you have 360w right? That just happens to be about the average power needed to go 20 mph. See if you can guess what amount of power a smaller motor and 20 amp controller would use. You got it, 10 amps. So on the flat, efficency is going to be pretty much entirely decided by your choice of speed, regardless of the motor and controller choice.

Now lets look at the hill problem. As I stated above, if you have less watts than needed you slow down too much. Slow down too much for the particular motor winding, and you will make heat with most of your avaliable watts. This slows you down more, making more heat, which slows you down more which makes you more heat which slows you down more which makes more heat..........

So on the hill, your extremely low amp controller is not going to just limit your amps, it can actually cause you to melt your motor.

As others have said, you need a gearmotor. You need a certain wattage to get up steeper hills. And idealy you'd pick a motor from a slower speed range. What size motor you need depends on the degree of slope plus the length of any hills you need to climb. 5% grades are handled ok by the small gearmotors. Above that, you want more powerful ones like bmc or Mac. You want to have 1000 w avalialbe for steeper hills, but you'll still use lots less if it's flat terrain.

I just use throttle all the time, and my most efficeint bikes are all using slower winding motors. They get less hot on steep hills, and more importantly, don't allow me to waste energy going faster than 30 mph. Some have 20 mph max speed, like my longtail cargo bike. But it will still climb a steep hill fast enough to keep cool, which saves me tons of wh.