Hub motor cuts out

My e-bike hub motor (48v) cuts out after only turning the wheel for a few seconds. I need to turn the throttle all the way off and on again to get the wheel turning again. SO, twist throttle and the motor turns the wheel for few seconds, motor stops turning wheel, twist off throttle, twist on throttle and motor turns wheel for another few seconds.
The voltage only drops down about 1v when running (reading taken via controller input). Well above the LVC. The LED unit on the handlebars shows all 4 lights lit the entire time. When the motor is turning turning it has full power.
I disconnected the brake shutoff wires.
I used another set of batteries.
Added another battery to increase initial voltage to 59.
Tried another throttle.
Tried another controller. Both were 48v controllers that are ok to 60v.
Of course checked battery and other connections (many times).
Checked wires for obvious damage.
Soldered wires that may have been a problem.
Put a new power connector on the controller.
This is my 4th bike conversion. This one I bought for my wife, but then she died of cancer, so I switched it to my bike with the same size tires. Had the problem from the minute I switched. The delay was long enough where the warranty (if there ever was one) no longer applies.

Does this sound like something inside the hub? Some kind of sensor or something? I keep going over the wires and connectors again and again hoping to find something, and I am not. I am thinking a sensor problem or something.But heck, I have been wrong before. It is how I learn.

If the phase wires and hall wires are in the wrong combination, the controller might shutdown from overcurrent or timing, etc.

If it has a "learn" wire, that can fix this problem. Otherwise you have to go thru teh various combinations to find the right one.

But the learn wire can also cause this problem, if it's left connected, depending on the controller design.

You need to get the multimeter out and pick a side to start, upstream or down stream. Since it starts up, I would focus on the controller to motor values.
The easy thing-ck continuity on phase/hall wires, PBC points to connector(s). Inspect push pins if there are white nylon connector as sometimes the pins back out. ck for breaks by twisting the cable bundle when ck'ing continuity.
Not as straightforward-I have never had a Phase Prob., but I guess I might ck. resistance value-Phase wire at connector to ground-all about the same? Values don't change when move whl and twist cable?
Halls-Power at red wire to Blk(ground). Do the Hall test, it's documented here if you do search, also covered @ Ebike CA site.
If everything ck's out, move the input side and ck all signals in for continuity/twist wire test.
Hopefully, at that time you will have seen something.

It's not clear what you have for spare parts. You have tried several controllers that are apparently good. They worked on another bike? Was the behavior the same (spin for second and stop). They are the same kind of kit or are they different? Has this motor ever worked?

I guess you have find the docs for the kit. Amberwolf asked if it has a learning circuit. I've only had one controller that had learning. I had to power it up with the learning wires connected to each other. The wheel would spin a bit. If it spun the right way, I was to disconnect the learning wires and never use then again. I did that, but the controller was only 10A and I found it too mild mannered. Wasted 50 bucks (it had an LCD too).

Oh, and my condolences on your wife. Cancer is awful. Had it run thru my family too.

Thanks guys!
I didn't know about a learner wire, even though I did 4 other bikes. I will check for it. I don't remember seeing anything like that. But worth a shot.
Yes, it ran fine on the other bike.
I did try moving the phase wires, but I need to go back and do it again.
Yep, ran exactly the same with both controllers.
Ok, I will do the tests and get back with my results.

FWIW, the problems I've seen with bad hall sensors (or broken hall wires) have prevented a sensored-only controller from being able to spin the motor at all, unless it was in just the right position to start with, and then it would keep going fine, with no load on the motor. When the hall power wire was broken, no sensors worked, and it wouldn't even attempt to spin the motor. There'd be one quick buzz but no movement, and that was it, regardless of motor position.

But some sensored-only controllers will start and then stop if they fail to detect all three hall sensors correctly, according to various posts I've seen (just not any that I've had).

If it was a sensored-or-sensorless controller, then a hall problem should simply be ignored and it would work anyway, but possibly run rougher at startup when loaded.

Amberwolf, thanks! That was my next question, would this fit in with a hall sensor problem. The motor will only start in one spot, but I still need to twist off the throttle and twist it on again to start. I will test the sensors first thing and then post with the results. It will be awhile, cleaning day - And since I can't make it through my workshop it will be a major clean.

Well, good and bad news. Yes, even though it didn't act like it there was a broken wire or two going to the hall sensors. Easy fix. Take hub apart. Cut wires. Rerun wires through shaft. Solder and heat shrink wires. Secure wires so they wouldn't rub inside the hub.
Ok, that all went as planned.
Part of this was an experiment to see if aluminum forks would work with the front hub motor. Well, one of the dropouts cracked before I even had a chance to ride the bike. LOL. Well, since no-one in my area carries forks that work I am off to find a steel fork from a used bike, cheap new bike, or from one of the internet stores. Thanks for your help.

There's a number of threads about front hubmotors and torque arms, including the Torque Arm Picture Thread.

I'd recommend adding some to any motor, front or rear, for most bikes, because the way hubmotors use the axle flats to transfer torque is not a good one and applies way more force to dropouts than they were ever designed to take, in directions they are not designed at all to take it.

A steel fork may not survive any better than the one you already have, without torque arms.

(other hubs like IGHs and coaster brakes transfer their torque via long arms, as do certain hubmotors like the Heinzmann).

Thanks, but all my bikes have torque arms on them. The dropout broke before I had a chance to ride the bike. It was probably cracked at some point in the past. Just a sign that this particular fork is probably too brittle.

It was probably not already broken or too brittle. They're simply not designed to take *any* force in the directions hubmotor axles put on them. It does happen that they break from non-hubmotor use but it's not very common, except when torque is applied to the faces or the sides by hubmotor axles, or similar events, like extreme overtightening of regular axle nuts, or incorrect installation of regular bike wheels and such.


If the torque arms dont' fit *perfectly* to the axle and are not secured and mounted in a way that correctly prevents any rotation of the axle, then torque is transmitted to the dropouts. That's the main thing that breaks them. Few torque arms are perfect fits, and many get installed in ways that don't prevent rotation.

A motor axle can break a dropout even with the bike upside down off-ground, with relatively gentle throttle application, and no load on the wheel.


THe second thing that breaks them is that many forks have "lawyer lips", and a recess for the nut/washer. That's made for a smaller diameter nut/washer than that which comes with hubmotors, cuz the axles on bike wheels are smaller. So unless you fill the recess with a "C washer" or similar, so that pressure is not put on the lip or other part of the dropout face, the side pressure of simply tightening the nuts can break the dropouts.


The torque arm threads have details on these things, as do the various "spun my axle" and "broke my fork" and "broke my dropout" etc threads. Also a thread by Justin_LE about testing forks and dropouts to see exactly why they fail.

I'm confused, I read back thru your posts to see what kind of bike we are talking about(no info) and I see you apparently had recently more or less the same problem w/ a rear hub motor?
AW is correct, things don't break of thier own accord and your cavalier attitude could be hazardous to your health. Sorry if this sounds harsh, but having the wheel come off a 20 plus mph ebike can cause serious harm.
There is no reason a hub motor on the kind of power you are running can't be safely mounted on a suspension fork(they are alloy, not alum.), I have been doing it for 8 years.
But, as AW points out, everything has to be perfect. The slot on a genaric T. A., made to fit both a 10 m/m and a 12 m/m axle may be too large to secure a motor w/ a 10 m/m shaft, the slotted plate should fit w/out excess slop. The arm must be bolted or at least sit flush on the fork leg w/ max. contact area. If hose clamps are used, the T.A. must be installed in a way that the motor torque pushes against the fork leg.
If you are in any doubt about these issues, spend the extra $$$ and pick a T.A. from Ebike CA to suit your needs
When a T.A. failure happens, there are usually rub marks or something that a close inspection will reveal(Yes, it has happened to me when I repaired a flat in a work area w/poor lite and I didn't get everything perfectly lined up. Fortunately, it was at the start and only one side snapped off).
I shopped around and found a quality budget fork for a little over $200;
https://www.amazon.com/gp/product/B07197D9LW/ref=oh_aui_detailpage_o06_s00?ie=UTF8&psc=1
It seems to me, that going to a ridgid fork on a 20 mph plus ebike is a major step backwards. If you want to stay w/ a sus. fork, you just need to match the spec.s of an aftermarket w/ the original when shopping around

I don't know what in the heck you are talking about. Does no-one have reading comprehension skills? Nothing you are saying has anything to do with what I wrote. Of course you can put a hub motor on an alloy fork. But you can't on an aluminum fork that is already cracked. Please reread what was written.