Sabvoton V2 (80amp) & NB Power 3k Watt - "Controller Protection" Message - Loss of Power

[NuclearDolomite]

Howdy, ES Community!

You folks were such a big help last summer while defining my bike's specs/design before finally pulling the trigger through NB Power's Chinese location. I worked with Anne over there and ordered both my 72V 20ah battery directly from them as well as the 3000watt (72v) fat tire kit, complete with Sabvoton V2 80amp controller and harnesses etc.

This 80amp Sabvoton controller came with the UKC1 display and was preprogrammed with no pedal assist. This 3000W NB Power motor came with two sets of hall sensors as well, which I thought was odd but now understand why ( i think).

The build was executed well and performance has been outstanding until last weekend. As a foreword, I do not go on long rides - maybe 6 miles each way at most. I don't use pedal assist since I opted not to have it, so this is an all-motor build; I never take it past 38mph, or level 3 out of the 5 different power levels. I get about 25-35 miles before I charge her again; she's heavy since this is a steel mongoose frame.

I was going about 34mph last week at level three and all of a sudden I lost power and a red check engine light appeared on the display with no code, just "controller protection" or something along those lines. The controller was a little warm and it was about 34 degrees out, so the motor was cool. I also use a bit of statorade but not too much, so heat is never a problem, or so I thought.

Upon restarting the bike (off then back on) and giving it some juice, even on level 2 (23mph), the bike lost power and says "controller protection," or something along those lines. I tried to ride again the next day and the same thing - three minutes into my ride, "controller protection."

I had to go to a doctor's appointment a few miles away so I unplugged the hall sensor harness and plugged in the other set of hall sensors that I don't use.

I was able to ride normally with no issues; maybe slightly less acceleration, but no real noticeable issues.

I'm sort of keen on electrical and have probes, meters, etc, but am a little scared of 72V and lithium poly batteries in general, so I haven't done any electrical diagnosis yet.

The display frequently reads a little over 3000 watts or about 2200 watts at cruising speed depending on the grade of the road, so it is possible the hall sensors simply failed and caused an overvoltage or spike in amperage in the motor, and the controller was just protecting itself? Now is isn't a real "QS" motor, and there's got to be the reason they included two sets of hall sensors, so are these things designed to fail?

OR .... could this be a controller thing? I'm finding that less likely since I have 100% functionality after swapping hall sensors.

I'm a car guy and have never had to repair 3kw electric motor before, but I also don't like the thought of just waiting for this second set of hall sensors to go out while not making any attempt to find the root cause.

For now, I've lowered my voltage max via the UKC1 display to 52V. There's a 60VH and a 60VL option, and I don't know what "VL" or "VH" is referring to (frequency? hertz?), so I chose 52V (coming down from 72V). For now, my thought is that through Ohm's law, lowering the voltage will lower the wattage the motor draws from the controller/battery, right? Essentially, R=V/I (.65 ohms) then P (watts)= Volts X Amps (I) .

Is 72V too much for this motor perhaps? Also, can I safely try the 60VL or 60VH options, or should I stay at the reduced 52V for now until I know what's going on?

 

[amberwolf]

I was going about 34mph last week at level three and all of a sudden I lost power and a red check engine light appeared on the display with no code, just "controller protection" or something along those lines. The controller was a little warm and it was about 34 degrees out, so the motor was cool. I also use a bit of statorade but not too much, so heat is never a problem, or so I thought.

If the motor never gets hot, it's unlikely to cause a hall sensor failure that way.

OR .... could this be a controller thing? I'm finding that less likely since I have 100% functionality after swapping hall sensors.

It's unlikely--if using the second set is working, then there's a problem with:
--the first set sensors
--the first set wires
--the first set connectors
etc

Or...an intermittent connector fault at motor/controller interconnect point happened to fail long enough when you saw the error, then when unplugged and replugged to the second set, "reset" the problem, such as a contact that is not seated correctly so it backs out with vibration or heat/cooling, or has spread out pins that make a good enough connection most of the time, but under specific or random conditions, it fails to connect. etc.

It could even be a phase wire connection problem, that happened to be reset by handling the wires as you were swapping the hall set plugs.

I'm a car guy and have never had to repair 3kw electric motor before, but I also don't like the thought of just waiting for this second set of hall sensors to go out while not making any attempt to find the root cause.

The main reason they fail is heat, but if that were the case, the second set would have the same problem, as it gets heated exactly the same as the other; they're both mounted inside the stator teeth of the motor for the typical hubmotor. (one on each side of the motor).

For now, I've lowered my voltage max via the UKC1 display to 52V. There's a 60VH and a 60VL option, and I don't know what "VL" or "VH" is referring to (frequency? hertz?), so I chose 52V (coming down from 72V). For now, my thought is that through Ohm's law, lowering the voltage will lower the wattage the motor draws from the controller/battery, right? Essentially, R=V/I (.65 ohms) then P (watts)= Volts X Amps (I) .

If you were actually lowering the voltage, by providing a different battery of that voltage, then that could be true.

But all you're doing by changing the display setting is changing the HVC / LVC and/or the battery meter response. It doesn't change the voltage the controller or motor sees or uses.

To reduce power to the motor, you'd need to reprogram the controller's current limit to lower that, instead.

Reducing power will reduce acceleration and hill climbing ability, and top speed wherever the speed you want takes more power than you then have.

Is 72V too much for this motor perhaps?

You can probably use two or three hundred volts on the motor itself without a problem.

If you have overheating of a motor, it's usually because of too high a load on it; you'd need a bigger motor to handle that power. There are other possible causes, like a wrong phase/hall combination (false positive, it works but heats quickly), or a motor that's wound for too high a speed compared to what it's being used under heavy load at, etc.

 

[NuclearDolomite]

@amberwolf I'm still on the clock for work (recruiting for quality & mfg. engineers) but took it out for a spin to grab a coffee on 52V thinking that reduced the voltage, lol... Was astonished by my battery going from 50% earlier to 100% and quite worried that I've really gone it and done it now. Sat back down at my desk and read just part of your response, and literally answered one of my worries, lol. You rock amberwolf- so grateful!

For shits and gigs, I plugged the "malfunctioning" set of hall sensors back into controller... battery reading 71.2 voltage at 40% or so capacity just now.

Had no issues whatsoever.

I'm off work in 90 minutes. Will give you a better esponse that addresses more of your suggestions here in a bit.

Time to cold call some engineering managers!

 

[NuclearDolomite]

Howdy, @amberwolf !

Just about to finish up here at my desk for the desk and figure I'd kill a few minutes by providing a more in-depth response so I can figure out what my next steps should be.

1.) I need to go back and change the battery meter setting to 72V since, as you stated, changing that does nothing.
2.) Motor barely even gets warm most of the time with that statorade in there. I drilled the pin hole for statorade toward center of motor and used tape to ensure I don't drill more than a rabbits hair deep. It's also sealed with clear silicon, motor never opened. Did this upside down so no shaving went into motor. Thermal performance boost after that.
3.) I don't go off jumps or hop curbs. It's adult built owned and driven. Plus my battery bracket isn't made to take that.
4.)

Or...an intermittent connector fault at motor/controller interconnect point happened to fail long enough when you saw the error, then when unplugged and replugged to the second set, "reset" the problem, such as a contact that is not seated correctly so it backs out with vibration or heat/cooling, or has spread out pins that make a good enough connection most of the time, but under specific or random conditions, it fails to connect. etc.

It could even be a phase wire connection problem, that happened to be reset by handling the wires as you were swapping the hall set plugs.

This is perhaps where my reasoning gets me due to my inability to reproduce this issue on either set of hall sensors yesterday.

Even if this isn't the issue, this bike is WAY too fast for me. It seriously wants to do like 55mph, so I only want my max speed to be like 39.

This controller came preprogrammed, and I have the dongle that came with it. I'd really like to limit the wattage to perhaps 2000 watts or something so I can get more life out of the components and make the bike a bit more manageable.

If I'm a newby to programming these Sabvoton's, do you recommend a good place to start? I think the less I push these Chinese components, the less likely they are to fail, right? Like, there's a big quality difference between what I have (nb power QS knockoff) and let's say a controller made by Grin and a real QS or Grin motor.

 

[amberwolf]

2.) Motor barely even gets warm most of the time with that statorade in there. I drilled the pin hole for statorade toward center of motor and used tape to ensure I don't drill more than a rabbits hair deep. It's also sealed with clear silicon, motor never opened. Did this upside down so no shaving went into motor. Thermal performance boost after that.

If the motor casing doesn't get warm, especially between the spokes, then you don't even need the statorade. All it does is transfer the heat from the stator to the magnets and thence to the area between the spoke flanges, for a DD hubmotor. (for geared hubmotors it doesn't do anything useful, because there's an extra air gap it can't bridge).

This is perhaps where my reasoning gets me due to my inability to reproduce this issue on either set of hall sensors yesterday.

That points to an intermittent connection problem. Wires and connectors are the most common failures in any system that has them.

They're hard to track down, but there is often visual indication of a problem. For high current connections there's often heating damage nearby or at the connection (discoloration, melting, distortion, etc). For low current connections there's not, but you can sometimes see insulation damage or distortion near a connection fault, where mechanical damage to the wiring occured. Can happen in use or during manufacture or installation.

Wire damage can happen literally anywhere along it's length on any wire, but is more common at connection points and wherever they enter or exit something. (most especially at the axle exit from a hubmotor)

If the connectors are not molded around the contacts/wires, you can look at the actual contact-wire interface to see if it's correctly crimped (or soldered for that type), and that it doesn't have broken conductors at the entrance to the back of the contact.
Youc an also check the contacts to see if they are mating correctly, or bent, spread, etc.

Even if this isn't the issue, this bike is WAY too fast for me. It seriously wants to do like 55mph, so I only want my max speed to be like 39.

You can just set a speed limit, if your controller setup program offers that. If not, you can install a device (like the ebikes.ca Cycle Analyst) that does this by reading your wheel speed and then rolling back throttle input to the controller if the limit is exceeded.

The catch with any speed limit is that if you are riding in traffic, there are certain situations where being able to accelerate out of the way of a much larger vehicle is very handy. It is often not possible to brake and avoid these situations, as traffic behind you may just run you over before they can react, or you may lose control and crash and *then* get run over, etc. But sometimes just a quarter-second of VROOM will get you out of the way of the truck that decides to change lanes into the spot where you already are, etc.

The best way to manage speed is just learn control of your throttle hand.

If that's not possible for whatever reason, you could set the speed limit, with an override designed so that if you suddenly slam on full throttle it'll do it anyway utnil you roll it back yourself. Or use an override button, but remembering to press that in a rare emergency is unlikely to happen, unless you first train it into yourself so it's as much a reflex as braking. Personally, I would never make an emergency control that operates differently than the "natural" training a rider already has for that function; it's just too likely to not be used in time (or at all).

(that kind of thing has caused numerous aircraft crashes and incidents)

This controller came preprogrammed, and I have the dongle that came with it. I'd really like to limit the wattage to perhaps 2000 watts or something so I can get more life out of the components and make the bike a bit more manageable.

If you don't use the power anyway, then there's no need to limit it. If you don't have a wattmeter, you might consider installing one (the Cycle Analyst does that, too, but it's pretty expensive just for that function; $20 gets a useful one off amazon)

If it accelerates too quickly, or wheelies, then limiting power is one way to manage that.

But if you limit the power, then like limiting speed, you remove options you might need if you ride in traffic (or various other conditions).

If I'm a newby to programming these Sabvoton's, do you recommend a good place to start?

That I dont' have any info on, but there are lots of posts and threads about setting them up. I can say that most of the apps for most controllers are complete garbage and if they even work to program them, they are so badly setup and labelled (not even counting poor translations) that it's tough to have any idea of what settings you're really changing in there.

Some like the Phaserunner setup program are better, but I'm not sure a PR could do the power levels you're after. Depends on the actual power you need to do the job the bike has to do for you. (the ebikes.ca simulator can help you figure that out).

I think the less I push these Chinese components, the less likely they are to fail, right?

Yes, generally--but the battery is probably the only one to worry about if you are not seeing anything above ambient temperatures on the motor, controller, etc.

Like, there's a big quality difference between what I have (nb power QS knockoff) and let's say a controller made by Grin and a real QS or Grin motor.

QS motors are the best-constructed *available* hubmotors I've personally encountered (they also have non-hubmotors that look pretty good from what's been posted here). The old Ultramotors are as good, but they're not available new and not intended for the power levels you're after. (I use a pair on my SB Cruiser heavy-cargo trike, but that peaks around 2kw or so for a few seconds at startup from a stop, spread between the two motors, driven by a pair of Phaserunners, falling off to the 400-500w each they're actually intended for).

There are threads around here discussing the NB Power motors; it's unclear if they make them themselves or if they buy them from QS with NB labelling.

QS is also a "chinese component", and realistically so are just about every other part you can buy...the differences are in the QC done on the parts, and the design behind them, as well as the manufacturing processes and parts sources/materials required by whoever is having them made and whehter they enforce them.


Unless you're overheating motors, or physically abusing them such that axle breakage is a risk, even some of the crappiest hubmotors will still last "forever", at least until you get bored with a build and start redoing it.

Wheel builds...that's a different story. Many (probably most) hubmotor wheels are built using spokes too thick for the bicycle rims they're on, so some eventually have spoke breakage, rim failure or other related issues, depending on their usage and any fixes that get done to them. For non-bicycle-rim builds, that's not normally an issue (but it can stil be a crappy wheel build).

Controllers, well, if it has the features you want, isn't getting hot, etc., then it's good enough.

Batteries...now those, they're often so bad that common blue brick batteries often get called all sorts of names. :/ Even ones in cases are not any better. Most of them either have crappy cells, poor builds, badly designed BMS, or other issues. If yours isn't giving any issues, then nothing to worry about until it does. But even if it's a good battery, I'd still always use it at less than it's max capabilities (meaning, get one that can do as much more than you need as possible), because as it ages it will naturally become less capable, and if it starts out being pushed to it's limits then very shortly it's being pushed past them as it ages, and eventually is constantly being pushed too hard and starts to really fail.