Hall sensor voltage question

Yer bike busted? tell us all about it here.
Post Reply
MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Hall sensor voltage question

Post by MarkJohnston » Mar 25 2021 2:43am

Hello

I made an account just for this. I recently got a problem with my new (2500 miles) direct drive Voilamart (ebay $200) hub motor. I live in Los Angeles and I was riding two nights ago when all of a sudden I began to lose power! I couldn't pull over right away because there was a long homeless encampment stretching for two blocks. These encampments can be pretty dangerous to be by at night. Anyways I pulled over as soon as I could and found that the controllers ground (black) bullet connector had slipped out and was barely making contact. I fixed this, but that's where the trouble began.

The bike started making a very loud noise. It sounds more like a Tesla motor now when before it was whisper quiet. Also at low speeds there are lots of vibrations and it sputters a bit. Other that the vibrations at low speed and accelerating, it feels fine.

I tested the hall sensor and it tests fine. it gets 4 volts on the red and black. And the other color wires it cycles between 4.7 volts and 0 when the wheel is spin backwards.

Any idea what is going on? Is it safe to ride? I need this to get around.

User avatar
e-beach   1.21 GW

1.21 GW
Posts: 3850
Joined: Jan 10 2012 9:48pm
Location: Any Los Angeles area beach I am at. Or Santa Monica or possibly the south bay beaches.

Re: Hall sensor voltage question

Post by e-beach » Mar 25 2021 12:15pm

The first thing I would is become Mr. Obvious and take a good look at the obvious stuff.

Are all your connections are good and tight?

Is your battery fully charged and balanced?

Are your brakes properly adjusted?

Are your shifters properly adjusted?

If the problem still persists, then turn off your battery (disconnect if necessary) then disconnect all your phase wires from the controller. Keep good track of which ones connect where for reassembly. (do not let the phase wire connectors touch or it will defeat an accurate result of this test. With all the phase wires disconnected from the controller, lift the dd motor wheel off the ground and give it a good spin. It should spin freely and not judder or be hard to turn. This will tell you if the motor has a phase short or possible internal damage.

Let us know how it goes.

:D :bolt:
Favorite Quotes:
"This is L.A., sugar. There is no 'over the top." Chris Erskine
"At a certain point the entropy wins." Maria Helena Braga

Current Build: ProFlex 757 Expert full suspension. Yescomusa 36v 800w Rear DD, upgraded 10AWG solid core through axle phase wires. 15ah Headway, 1000+ cycles, 80% DOD 30A Tronsung controller.

Past: Trek 4500 Yescomusa 36v 800w front DD.
Liahona, Yescomusa 36v 800w generic front DD.
1992 Trek 800, Yescomusa 800w 36v front DD.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 25 2021 5:55pm

Okay,

I did what you said and checked everything. The motor turns freely. I don't believe there is a short with the phases.

User avatar
amberwolf   100 GW

100 GW
Posts: 32163
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Hall sensor voltage question

Post by amberwolf » Mar 25 2021 7:56pm

Does it also turn freely both forwards and backwards with the controller connected (but not powered on)?
MarkJohnston wrote:
Mar 25 2021 2:43am
The bike started making a very loud noise. It sounds more like a Tesla motor now when before it was whisper quiet. Also at low speeds there are lots of vibrations and it sputters a bit. Other that the vibrations at low speed and accelerating, it feels fine.
The usual problems that cause these symptoms are FET problems in the controller, and hall signal problems.

The latter isn't likely since you see valid signals on the halls, in the normal voltage range.

But FETs could be damaged by the RF from arcing on that battery negative wire, and may not be switching like they are supposed to (wrong timing, essentially). I don't know of a test you could do without an oscilloscope to see the waveforms to know if that is the problem (other than trying a different controller).

The Battery also has FETs in the BMS that could be damaged by RF from arcing, though that usually causes different problems than you see.

Another possibility is simply that the loose connector is damaged in a way that is still making a poor connection, so while it works it doesnt' work liek it should.

It may also be that if the controller is one of those that autoswitches between sinewave and trapezoidal mode based on certain conditions (like the Grinfineons), it might have been in sinewave mode before, and now be in trapezoidal. With the Grinfineon, this happens when the controller isn't correctly seeing the hall signals (even if they are working), and switches to sensorless mode. That *also* causes more noise and sputtering / jerking at low speeds and startups from a stop.


I would guess that the most likely problem is that the controller has been damaged in some way and may need to be replaced, if it can't be "reset" to it's original operating mode.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 26 2021 12:19am

Thanks, this is informative. I tested the hall sensor connector and got 4 volts on the black and red (+ -) . I then tested the three other color wires by moving the wheel backwards and seeing if the voltage changed. I had one lead on the black negative and the positive lead I placed on each and every color and it was switching between 0 and 4.7 volts for each color. Now, I'm not sure if 4 volts is too low power for the halls. The testing manual I read said between 4 and 5 volts but other people have said if it's anything BUT five volts something is wrong with the halls.

I have a separate battery and tested it. The noise did not go away.

Moving the wheel backwards and forwards FEELS NORMAL. As another test I UNPLUGGED the hall sensor connecter from the controller. ABSOLUTELY NOTHING changed. The noise did not go away and at low torque and acceleration there is vibrations. But once it gets up to speed it feels fine though it's making a new noise that sounds like the wave pattern may have changed. Sorry I am a new to electric bikes. I didn't really know so much went into these things.

My big question is, is this thing ok to ride or could it possibly be damaged more so by riding it? Is it safe to ride? I want to be able to ride while i order a new controller.

User avatar
amberwolf   100 GW

100 GW
Posts: 32163
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Hall sensor voltage question

Post by amberwolf » Mar 26 2021 12:38am

MarkJohnston wrote:
Mar 26 2021 12:19am
Thanks, this is informative. I tested the hall sensor connector and got 4 volts on the black and red (+ -) . I then tested the three other color wires by moving the wheel backwards and seeing if the voltage changed. I had one lead on the black negative and the positive lead I placed on each and every color and it was switching between 0 and 4.7 volts for each color. Now, I'm not sure if 4 volts is too low power for the halls. The testing manual I read said between 4 and 5 volts but other people have said if it's anything BUT five volts something is wrong with the halls.
No, they're fine at those voltages as long as that is a "5v" hall system, which most are (depends on the controller, rather than hte motor). It's rare for hall sensors to be "exactly" 5v, because there's a number of things involved. Besides, they don't output any voltage anyway, they jsut ground the voltage the controller outputs ("pullup"), so as long as they toggle from something near 5v to something near ground, they're working.

(some systems use 12v+ for these signals, and some use only 3v+, most use around 5v)

Side note: Many controllers actually only output around 4.3v or so to the hall power, because they have a diode in line with that to prevent backfeed of motor voltages into the 5v main power supply in case of axle wire damage, or just induced noise into the wires from the phase wires in the motor. But they use the regular 5v line on the pullup resistors that create the 5v on the signal lines, so those voltages are almost always higher than the hall supply voltage, which confuses a lot of people. ;)

I have a separate battery and tested it. The noise did not go away.
At least it's not the BMS then, or the battery-side connectors/wiring.
Moving the wheel backwards and forwards FEELS NORMAL. As another test I UNPLUGGED the hall sensor connecter from the controller. ABSOLUTELY NOTHING changed. The noise did not go away and at low torque and acceleration there is vibrations. But once it gets up to speed it feels fine though it's making a new noise that sounds like the wave pattern may have changed. Sorry I am a new to electric bikes. I didn't really know so much went into these things.
Yeah, some of the systems can be complex to troubleshoot (easy to use when they're working, though). I've been doing this more than a decade and I am still learning new things. :oops:

If unplugging the halls makes zero change to operation, then it is almost certainly just that the controller has changed modes so it is ignoring the hall sensors (or failed so it can't read them) and is now in sensorless and/or trapezoidal mode.

It's not uncommon for trapezoidal controller operation to sound different the lower the load is on the motor, and/or the faster it is going. Sometimes there is even a particular speed at which it may make a very different sound as it crosses that speed, but doesn't make that sound at any other speed, and this can be different in sound and/or speed depending on the motor it's running. :?


My big question is, is this thing ok to ride or could it possibly be damaged more so by riding it? Is it safe to ride? I want to be able to ride while i order a new controller.
It depends on what the cause of the sound is. If it is a controller FET problem, there is a chance that they could fail shorted suddenly, which could lockup the motor. If it's a geared hubmotor, that's not a problem for riding or safety because the wheel usually freewheels rolling forward, you just wouldnt' have any motor power anymore until replacing the controller.

If it is a controller mode problem (switched to sensorless and/or trapezoidal or both), then there is no safety issue, it's just annoying. My bet is on this, based on the results of unplugging the hall sensors.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 26 2021 1:32am

I tested the mosfets. I unplugged the controller and pulled out my multi meter. I then switched the MM to resistance. I inserted the black probe to the negative on the power input. I then tried probing the phases to see what resistance I would get. I got 1 or 13k resistance, really high restistance. I repeated this step by switching the probes , red on red and then the black to the phases. Everything checks. I pulled apart the controller too looking for damage. Everything looks good, no hot spots, no loose wires, no burn smell. Why did this damn thing suddenly decide to switch to sensor less? It's annoying.

On a side note, the controller has been in a bag too lately to hide all the wires. It's been getting real hot and I've been thinking about making an enclosure with air holes. Also I noticed that the on/off button doesn't matter anymore. Once I plug in the battery the throttle is live, the power button only turns on/off LCD display! But this issue has been there pretty much since day 1 mile 1 on this bike

Speaking of the LCD, no error codes! It says everything is normal! :confused:

So what's going on with this thing?

User avatar
amberwolf   100 GW

100 GW
Posts: 32163
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Hall sensor voltage question

Post by amberwolf » Mar 26 2021 2:30am

I never recommend running a controller inside anything. They are designed to be run in the open air to ventilate that heat. Without that, heat builds up inside and ages or even damages components. There are much better ways to hide wiring, and the controller can be painted to match the bike color, or accessories on the bike, to help it blend in. The paint is not nearly as much of an issue as covering the case inside a box or bag, and would be unlikely to cause problems, whereas controllers do die from the heat inside bags/etc.


As for what caused the problem, it is almost certainly the RF from the arcing on the battery negative wire. This RF is broadband and high powered, so it has a lot of energy that goes right into the entire ground system of the controller, and attempts to find current paths whereever it can, and it is also probably flowing backwards thru the system, so there are a lot of things it could damage in ways that I woudln't even care to try to predict. It may only take a millisecond of this kind of thing happening to cause damage. The damage is all internal, and often microscopic, so it's unlikely you could tell if any particular part is bad or not---in this case, it's probably in the MCU chip itself.

(ESD, or the zap you get touching something sometimes, is bad enough to damage or even completely destroy electronic components. Arcing from things like that battery wire is MUCH worse).


It is possible that the display and/or the controller MCUs have some sort of default mode they go into when programming is corrrupted (which the RF could do) that causes the operation mode you see. Without access to the programming software and knowing what settings to change, it's not likely you can fix that.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 26 2021 2:48am

Okay looking like I need to purchase a new controller. So from what you are telling me, sounds like I need to build a small enclosure to hide the wires. I'll probably need to mount the controller directly to the bicycle rack with no enclosure. Ok. :D

User avatar
e-beach   1.21 GW

1.21 GW
Posts: 3850
Joined: Jan 10 2012 9:48pm
Location: Any Los Angeles area beach I am at. Or Santa Monica or possibly the south bay beaches.

Re: Hall sensor voltage question

Post by e-beach » Mar 26 2021 10:04am

I am beginning to wonder if the electrical arcing didn't just reset the controller to a default state. Any way to change your controller settings?

Also, remember that controllers are water resistant, not water proof. So mount is in a place it won't get sprayed by water while riding.

As a side bar, I have had the controller on my main bike mounted inside my frame mounted battery box for years with no problems. However, the top, bottom and sides of my battery box are sheet aluminum. My controller is mounted against the top of the inside of my battery box in a way that the top length of my box becomes a heat sync for my controller. Although, I must stipulate I never push my bikes super hard and it rarely gets as hot as it does where AW lives.

:D :bolt:
Favorite Quotes:
"This is L.A., sugar. There is no 'over the top." Chris Erskine
"At a certain point the entropy wins." Maria Helena Braga

Current Build: ProFlex 757 Expert full suspension. Yescomusa 36v 800w Rear DD, upgraded 10AWG solid core through axle phase wires. 15ah Headway, 1000+ cycles, 80% DOD 30A Tronsung controller.

Past: Trek 4500 Yescomusa 36v 800w front DD.
Liahona, Yescomusa 36v 800w generic front DD.
1992 Trek 800, Yescomusa 800w 36v front DD.

RTLSHIP   1 kW

1 kW
Posts: 443
Joined: Jul 24 2008 11:26am
Location: S. Florida

Re: Hall sensor voltage question

Post by RTLSHIP » Mar 26 2021 10:23am

A good spot for controller is underneath the rear battery rack, if possible. That will help keep it cool and dry. Those smaller lower amp controllers will get hot under heavy riding but I can't say that's why you have this problem. You can get better mileage with a wallet sized controller.
two Aotema Brushless & Sensorless Kits
1 Free Spirit 12 Speed

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 26 2021 5:44pm

Okay, I'll keep this advice about the controller mounting in mind. Are vibrations a problem when mounting controllers? Does it need a dampener like a big peice of rubber ? Looked like there were a bunch of screws holding the thing up when I took it apart. Looked pretty solid.

The controller is kind of just the generic Chinese one. This one is 48 volts 1000 watts. It connects up to the SW900 WHICH the ebikelings one do to. I looked online and didn't see a way to reset the ESC. Anybody have any ideas?

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 26 2021 10:05pm

Hello,

I am now trying to purchase a new controller. Does anybody have any recommendations? I currently have a 48 volt 1000 watt generic Voilamart / chinese :lol: controller. It pairs with the generic Chinese SW900 LCD. however it's a little boring and I'd like to take this chance to upgrade the controller. I noticed this mixed voltsge one 48 to 96 volt one on eBay!

https://www.ebay.com/itm/372116818481

Now that's seems pretty sick. I have two 48 volt batteries so I could string them together in series for some epic WHEELIES :twisted: and some mad burnouts! :lol: I could go like 60mph and kill the battery in 10 minutes and burn my brake pads to ash! :( Haha but in all seriousness I would probably keep the voltage low at 48. Could this controller shed the heat better? It doesn't seem like it would work with this LCD. THE SW900. I also heard sometimes the hall combos don't match up well controller to controller and you have to " self learn"

RTLSHIP   1 kW

1 kW
Posts: 443
Joined: Jul 24 2008 11:26am
Location: S. Florida

Re: Hall sensor voltage question

Post by RTLSHIP » Mar 27 2021 8:09am

MarkJohnston wrote:
Mar 26 2021 5:44pm
Okay, I'll keep this advice about the controller mounting in mind. Are vibrations a problem when mounting controllers? Does it need a dampener like a big peice of rubber ? Looked like there were a bunch of screws holding the thing up when I took it apart. Looked pretty solid.

The controller is kind of just the generic Chinese one. This one is 48 volts 1000 watts. It connects up to the SW900 WHICH the ebikelings one do to. I looked online and didn't see a way to reset the ESC. Anybody have any ideas?
I've never had a controller fail (in 12 yrs) because of bumpy riding , though an aluminum rear rack once cracked because of heavy sla's. You can put rubber around it to mitigate bumps and bouces. If you put it under rear rack, you still have to be careful
with puddles & rain. As for going on 96 v controller , your hub would get very hot and the magnets could come loose. That seems pretty dangerous, too. Once you get to 30 or 35 mph, you should consider a scooter or motorcycle bike. Something more stable, and there are some legal issues to consider.
two Aotema Brushless & Sensorless Kits
1 Free Spirit 12 Speed

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 28 2021 12:53am

Here is what the controller sounds like it. I just said frock it's Saturday, let's ride! :twisted:


Listen just for the first few seconds because the noise goes away fast. There is a crazy vibration sort of noise that wasn't there before the incident.

https://youtu.be/K2pWTcsGQ7U


One of these videos shows the sort of resonance sounds a Tesla car motor makes. Can anybody confirm this noise? I googled "sensor less noise" and surprisingly only my video comes up. I guess ebikes are gonna start becoming more popular with bird, lime etc doing their thing. Maybe more info like this will be available in time. Here is the video of the resonance noise


https://youtu.be/OswZAoRpIG0



TLDR: is this a sensor less noise?

User avatar
amberwolf   100 GW

100 GW
Posts: 32163
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Hall sensor voltage question

Post by amberwolf » Mar 28 2021 2:39am

Sounds pretty much like a typical trapezoidal controller (vs sinewave). (sensored or not)

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 28 2021 5:04am

No. I swear to God it wasn't making those noises before. You really gotta listen to the longer video first. If you listen real hard for the first few seconds you can hear it.

User avatar
e-beach   1.21 GW

1.21 GW
Posts: 3850
Joined: Jan 10 2012 9:48pm
Location: Any Los Angeles area beach I am at. Or Santa Monica or possibly the south bay beaches.

Re: Hall sensor voltage question

Post by e-beach » Mar 28 2021 6:54am

amberwolf wrote:
Mar 28 2021 2:39am
Sounds pretty much like a typical trapezoidal controller (vs sinewave). (sensored or not)
+1......

My trapezoidal controller sounds creaky at startup. That is what they sound like. I am still wondering if your controller reset on the connection of your ground wiring.
Favorite Quotes:
"This is L.A., sugar. There is no 'over the top." Chris Erskine
"At a certain point the entropy wins." Maria Helena Braga

Current Build: ProFlex 757 Expert full suspension. Yescomusa 36v 800w Rear DD, upgraded 10AWG solid core through axle phase wires. 15ah Headway, 1000+ cycles, 80% DOD 30A Tronsung controller.

Past: Trek 4500 Yescomusa 36v 800w front DD.
Liahona, Yescomusa 36v 800w generic front DD.
1992 Trek 800, Yescomusa 800w 36v front DD.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 28 2021 7:44am

Good question e beach

I have this thing called an ST LINK. You solder it into a PCB/MOBO/CIRCUIT BOARD. it allows programming through USB to a computer. Not sure where id plug it into this controller though. I'm also not a programmer. No idea how is change stuff

User avatar
e-beach   1.21 GW

1.21 GW
Posts: 3850
Joined: Jan 10 2012 9:48pm
Location: Any Los Angeles area beach I am at. Or Santa Monica or possibly the south bay beaches.

Re: Hall sensor voltage question

Post by e-beach » Mar 28 2021 10:06am

Any way to change settings through your LCD?

:D :bolt:
Favorite Quotes:
"This is L.A., sugar. There is no 'over the top." Chris Erskine
"At a certain point the entropy wins." Maria Helena Braga

Current Build: ProFlex 757 Expert full suspension. Yescomusa 36v 800w Rear DD, upgraded 10AWG solid core through axle phase wires. 15ah Headway, 1000+ cycles, 80% DOD 30A Tronsung controller.

Past: Trek 4500 Yescomusa 36v 800w front DD.
Liahona, Yescomusa 36v 800w generic front DD.
1992 Trek 800, Yescomusa 800w 36v front DD.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 28 2021 6:57pm

I didn't see a way. The only thing that can be reset is the ODO. I'm not sure if that would help.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Mar 29 2021 4:13am

Another thing too, may not be the reason behind the groaning/vibration noise at start up.

Pretty much since day 1 there has been play in the wheel. When I grab the dirty wheel (wearing gloves :roll: ) I can move it back and forth. I checked ALL spokes. Everything good there. The play is coming from the axle, but I buttoned it up tight. It doesn't rock very much.

Not sure what is going on here, this happened to me a couple times on a REGULAR bicycle and it just meant the bearings weren't right. But I'm not sure how or if this translates to ebikes.

Again I don't think this is causing the noise. Pretty sure the controller just ain't recognizing the hall sensors because I've GOT THEM UNPLUGGED and nothing changes!

But it's just an idea and id like to know why the heck the is play there and if it's a PROBLEM.

MarkJohnston   1 mW

1 mW
Posts: 13
Joined: Mar 25 2021 2:31am

Re: Hall sensor voltage question

Post by MarkJohnston » Apr 18 2021 1:38am

Hello EVERYBODY,

The new controller FIXED THE PROBLEM. I properly installed INTO THE BICYCLE RACK. It's pretty well hidden underneath the rack. Thanks AW. YOU REALLY helped me out. I had no idea that there was damn "RF" crap. Anyway I could send you $20? Thanks man.

Yeah smooooth operation now. I might've killed the last controller by keeping it in a bag.

User avatar
e-beach   1.21 GW

1.21 GW
Posts: 3850
Joined: Jan 10 2012 9:48pm
Location: Any Los Angeles area beach I am at. Or Santa Monica or possibly the south bay beaches.

Re: Hall sensor voltage question

Post by e-beach » Apr 18 2021 7:50pm

:thumb:

:D :bolt:
Favorite Quotes:
"This is L.A., sugar. There is no 'over the top." Chris Erskine
"At a certain point the entropy wins." Maria Helena Braga

Current Build: ProFlex 757 Expert full suspension. Yescomusa 36v 800w Rear DD, upgraded 10AWG solid core through axle phase wires. 15ah Headway, 1000+ cycles, 80% DOD 30A Tronsung controller.

Past: Trek 4500 Yescomusa 36v 800w front DD.
Liahona, Yescomusa 36v 800w generic front DD.
1992 Trek 800, Yescomusa 800w 36v front DD.

Post Reply