Controller drains battery!

yuhr

100 mW
Joined
Feb 21, 2010
Messages
43
It really does:
About a year ago my “defective” battery was exchanged by the manufacturer/vendor for a new one. It turned out to be only marginally better than the original one (7 mi vs. 6 mi). The merchant informed me that “there was nothing wrong” upon testing all 12 cells of the original battery. Now I am sure the merchant was right. It took me almost a year to figure it all out: now I ride a 3-speed Worksman Cycles 20" tricycle: it is rather slow, yet has tremendous torque, so I am able to climb virtually any hill in the neighborhood using muscle power alone. So, usually I do not use the motor at all. I always turn the controller switch “OFF” at night and quite frequently did not turn it “ON” all day long (e. g. when I ride ~ 1 mi to the subway station and park the bike for all day long and later ride another mile back). In short, I use the motor only when I have to climb a very steep long hill, which is to say rather infrequently. Yet, upon getting home and checking the battery before possible recharging, I noticed that the voltage drops much more than should be expected after a short period the battery was actually used.
So, recently I did an ultimate test: I left the switch “ON” while the bike was parked at night: Next morning the battery voltage was only ~37 V! If I leave the switch “ON” for a few days, the voltage might drop below 30 V. Of course, if the switch is in “OFF” position, the battery needs to be recharged about once a month.
Summary: If the switch is “OFF” everything is normal. If the switch is “ON” the power is drained regardless the motor use and/or throttle engagement. In other words, when the motor is spinning, the battery is drained very fast, by the motor AND by the controller simultaneously!

I opened the controller - it looks and smells like a new, undamaged one would – i. e. all parts are shiny, all tiny markings are clear and legible, etc.
I wonder: is my problem unique? At the time, the merchant/manufacturer had asked me to ship the charger along with the “defective” battery. Apparently, the controller was not even suspected. Is there an explanation and can anything be done?
 
This is normal.

The controller will draw power when the power switch is turned on, anything from around 50mA to maybe 200mA, depending on the make of controller. With the power switch on all the controllers internal circuitry is powered and drawing some power.

For example, with the power turned on to the controller for 12 hours (over night), then you would drain around 10 to 15% of a typical 10Ah battery pack. Leave the controller powered on for longer and you can easily kill the battery; I've done it (accidentally) as have others here. It's one reason why I now have an auto power off circuit on my latest build. If the circuit detects that the bike hasn't drawn appreciable power for a few minutes it turns the power off automatically, to prevent the battery being drained.
 
Agreed; normal. :)

On my controllers, the Phase FETs are always connected to the Battery so that they have the least possible resistance to best current. The ON/OFF circuit simply enables/disables the MCU Voltage Regulators access to battery power; without it the brain can’t drive the motor. The leakage of battery current is through the Phase FETs and associated circuitry.

One way around the problem could be that the ON/OFF circuit drives a relay that throws a switch between Battery and the FETs, however I should think we’d want the switch buried in an oil can to suppress the arc. Not sure what the voltage spike would to do the caps; maybe if there’s a pre-charge circuit that waits ½ second before throwing the switch. Anyways – there are solutions out there that can prevent the pesky drain, it’s just that it takes more $$$ and weight to deploy correctly. Surely someone has thought this out before.

FWIW, I top-off once a week if not riding.
~KF
 
Get a wattmeter, and you can put it between the battery and the controller and see what the drain is, switch on, and switch off.

There's always also the possibility of a trickle short. Everybody who ever had a car with one raise thier hand, yep everybody. Don't drive for a week, battery dead. That would be not normal.

Most of my controllers draw about 3 watts with the controller on, and with it off, might be too low to show on the CA, but still a positive number less than a watt.

So if you check with a wattmeter, and you see 5 watts with the switch off or something like that, then it's not normal.


Another thing that is normal, is for an unused battery to drop some voltage after a charging, to it's true resting voltage. But a good 36v battery would not drop to 37v. 40 maybe, but not 37v. Just unplug it when not using the motor for a while, and see if it sitll loses that much voltage. It could be a battery issue, or just the controller draining it. Or both.
 
Thank a lot.
I just thought that the controller draws a minimal amount of energy to power its electronics, not as much or comparable to the motor. Anyway, if it's normal I can live with it.
 
dogman said:
<snip>
Most of my controllers draw about 3 watts with the controller on, and with it off, might be too low to show on the CA, but still a positive number less than a watt.
Yeah, that's my issue: One CA says it's pulling 12W idle and the other says 0W. Both CAs were purchased at the same time. Doesn't matter what controller either; I can swap - and one will read higher than the other. Must be internal to the CA. I don't worry about it too much though as long as voltage and mileage ares reported reasonably close together. :wink:

~KF
 
WOW, that's a lot of variation for a hopefully reasonably accurate device. I would have expected more like + - 2w at most.

Again, we don't know what your controller is doing is normal, till you measure it with a wattmeter. You could have a 20w short somwhere for all we know.

But it is normal for a plugged in controller to drain a battery eventually. Particularly dangerous is to ride, park the bike, leave the controller on, and then not charge the battery. A week or less and your battery is done in because it was near empty when you parked.
 
I do not own a wattmeter and actually not interested in making the diagnosis – what will I do with it? I mounted the controller under the saddle, so now I can reach the switch any time I need to. Yet, the problem remains: when I twist the throttle, the motor engages plus the controller itself drains a lot of power (being a curious guy, I wonder: where all this power goes? - the controller remains cool to touch, yet its housing has cooling fins – what for?), so the battery discharges very rapidly, in my case after ~7 mi I have to pedal to get moving.
 
Back
Top