q100 week and strange sound

Hi!
I'm building a commuter bike with q100c motor and a Chinese 17 amp controller with self learning procedure. The self learning procedure was not working for me. The engine always rotated backwards, but on the throttle it was going fine. I've put everything together for a test and did a ride - it was perfect! I disassembled it and solder nicely to hide and cut excess cables and now it is not working properly The engine makes strange sound and it does not rotate with full speed. Anyone knows what I did wrong? Here is the video:

https://www.youtube.com/watch?v=F3i3qrqHIBo&feature=youtu.be

The self learning procedure is not working properly - as it was before.

It's not normal for the motor to have EVER rotated backwards, even if it would eventually straighten itself out. You have a Hall sensor related problem and the controller is having a problem Sync'ing up w/ the motor. Whether or not it's a bad connection or you have the wires mis-matched, I'm not sure. Since you mentioned it's a 17 Amp controller leads me to believe it's a genaric controller and every one I have used w/ the Q100 matched straight up, blue to blue, green to green and yellow to yellow. Never the less, some controller don't match up, like the Infineon.
If all your connection are good(have continuity), you might research the sticky thread; "Controller info links", or eventually, try all the color combos.

Triple check that no wire insulation is frayed or torn, right down to just inside the axle.
Double check that your connectors are all in tip top shape, no black marks.
Singularly check that your throttle is rocking to Tom Petty.

I'v discovered that I was doing the self-learning in wrong way. I was turning off the controller, as I had in the instruction, actually disconnecting the self-learning cable and connecting it again did the trick. Now it rotates in the same direction when using self-learning and throttle. However the problem remained. It is even visible at self-learning. The engine makes strange sound, not from the very beginning, but when it gets to some speed. Please take a look at video, it contains 3 self-learning attempts: https://youtu.be/p9ONjy6ZXZs

I've tested all the configurations of hall sensors - no difference - sound and weakness remains. Then I've disconnected the + cable of halls and the sound disappeared but controller rotates only to some speed and turns off - this look to be the same speed as sound turns on when self-learning.

Cables look fine. I have no idea what to check next

Your term " self-learning "
Are you using a internet translator website to translate ?
If so what are you trying to say in your language ?

I also have a Q100c and it makes makes that sound when I put in more than about 25% throttle, and the sound increases as I apply more throttle.
The whirring sound is not as loud as yours, however on mine it does increase with the more throttle I apply like on your videos.
I have it connected to a generic small 6 fet , 15 amp controller and using 14s battery pack. Your Q100c will run better after you have some miles on it, mine did not spin smoothly until I had 50 or more miles on it.

Motomech has allot of experience with these motors , I wonder what he says about the noise.

Also
It would be helpful to show in your video's what you are doing with the controller and twisting the throttle . So all three items in the video screen Motor/Controller/Throttle and what you are doing .
and Make the video's longer , and apply the throttle several times in the next video's.

styczoo said:

I'v discovered that I was doing the self-learning in wrong way.

The engine makes strange sound, not from the very beginning, but when it gets to some speed. Please take a look at video, it contains 3 self-learning attempts: https://youtu.be/p9ONjy6ZXZs

Click to expand...

That IS a Hall sync problem!
The next test is the Hall testing procedure;
https://www.youtube.com/watch?v=Y4O9YQG08vE&app=desktop

Testing Hall sensors;
"to test the hall sensor wires. If your multimeter had good thin tips, you should be able to slide them into the backend of the white connecftor while it is all assembled. Set your multimeter so it's set to measure voltage, somewhere in the 5 volt range. Stick the red one alongside the red wire, and the black one alongside the black wire. You should measure 5 volts. If not, there is something wrong. After that, stick the red probe into the green wire, leaving neg. probe connected to black wire, and turn the wheel very slowly. About every inch or so of wheel movement the voltage should toggle between 5v and 0v. If it doesn't toggle, there is something wrong. Then put the red to blue, leaving black connected, and move the wheel slowly to see if it toggles. Repeat for yellow.

You should always have a constant 5v coming from the controller on the red wire. You should have the Green, Yellow, and Blue wires alternate between 5v and 0v. Always leave black probe connected to the black(neg.) wire in the connector when testing the halls."

Thank you Motomech!
I did this test of hall sensors and even recorded it https://photos.app.goo.gl/EKwPCvpPaKQKnU4v5. Hall sensors look fine for me.

By self-learning I mean ability to teach the controller the order of hall sensor wires so it knows which way to rotate the motor. When you connect the learning wire controllers spins the motor in one direction. Disconnect and connect again - controller is spinning in other direction. I did it at the video, but I made a mistake and switched off the controller is wrong moment so it didn't remember the result of learning.

I'm still building this bike. I did a test, taped everything, it behaved fine - no strange sound, good speed, I even rode it once. Batteries are hidden in frame. I'll create a build thread soon..

It started making this sound after I've shorten the cables and solder it

Maybe your self learning controller has some limitations. I was reading about the Bafang G310, which is another small motor that has a high gear ration like the Q100H.
https://www.ebikes.ca/product-info/geared.html

Justin Le writes ..."The G310 motor have a large 11:1 gear reduction and 16 magnet rotor producing 88 effective pole pairs, which is higher than the 40-50 effective poles in most other geared motors. In many situations this puts the electrical commutation frequency (eRPM) at cruising speeds that is outside the range of sensorless controllers."

Perhaps your controller is treating the Q100H as a sensorless motor? Or maybe it's just not fast enough.

By the way, I had a Q100H on my folding 20" bike that had a similar problem, except it was quiet and spun well to max rpm on the test stand. When I tried to ride it, it groaned and made the same sound as your video.

It was a bad 20A sine wave controller. I moved the controller around to different ebikes and the problem followed it.

docw009 said:

Maybe your self learning controller has some limitations. I was reading about the Bafang G310, which is another small motor that has a high gear ration like the Q100H.
https://www.ebikes.ca/product-info/geared.html

Justin Le writes ..."The G310 motor have a large 11:1 gear reduction and 16 magnet rotor producing 88 effective pole pairs, which is higher than the 40-50 effective poles in most other geared motors. In many situations this puts the electrical commutation frequency (eRPM) at cruising speeds that is outside the range of sensorless controllers."

Perhaps your controller is treating the Q100H as a sensorless motor? Or maybe it's just not fast enough.

Click to expand...

Yes, I think you are right on Doc. Although some have been able to run the Q100 sensorless, I never have had any luck. Perhaps they were using a 328 speed range and I was using the faster spinning 201 or 260. At any rate, it is quite possible that the compound gearing spins the Q100 so fast that the self-learning module can't read the pulses.

docw009 said:

It was a bad 20A sine wave controller. I moved the controller around to different ebikes and the problem followed it.

Click to expand...

Yup more then likely its a bad controller. Could be anything in there, so best is to order another and then open that one up and see if you can see anything.

I've connected another controller and problem remained. I might have touched hall and phase pins while putting nut over motor wire while installing wheel. Instructions how to change halls in q100 are welcome.

A very difficult job. The Halls are mounted in a "stand-off" position. The 3 long leads are like long legs and I don't think it is possible to position them correctly by hand. I think they must be installed with a jig at the factory.

If it is too difficult to replace all the hall sensors then you can just buy a bare hub motor , the same motor you have but without the wheel , and just switch out the core of the motor.
And buy a controller from the same place as the motor .

I have not yet made a wrench to take off the cover off my Q100c , but perhaps motomech has . Motomech can you show everyone how you made to a tool to take out the core of a Q100c/Q100h , Q128c/Q128h ?

I've taken several Q100's apart w/out a special tool. Buy a hex-head bolt and some flat washers and install it snug in one of the cover screw holes. Then strike it shapely with a drift(punch)to loosen the cover in a counter-wise direction. This works on frt. motors and un-used rears. Once a rear motor has been used, the force of the cassette/free wheel tightens the cover too much. Then a tool like this one must be used;
https://endless-sphere.com/forums/viewtopic.php?f=3&t=13998&hilit=cute+tool&start=75
Here is a nicer one, probably made by a machine shop;
https://endless-sphere.com/forums/viewtopic.php?f=3&t=62056&p=926939&hilit=cute+tool#p926939

My Rear has tighten so much it will only be opened up when I can make a tool.
styczoo, one of the two tools linked to by motomech is what you will probably need.

motomech said:

Once a rear motor has been used, the force of the cassette/free wheel tightens the cover too much. Then a tool like this one must be used;

Here is a nicer one, probably made by a machine shop;
https://endless-sphere.com/forums/viewtopic.php?f=3&t=62056&p=926939&hilit=cute+tool#p926939

Click to expand...

Thank you all! This is exactly what I'll do!

I've tested with another motor and the problem disappeared. For that test I used 328 rpm motor (I'm going to use 201rpm version in my build) and there was a roaring sound when the wheel rotated about 45 km/h and it was not accelerating anymore. I guess it should accelerate further according to simulator. The 201 version did not give such sound at full speed, what is causing it?

My controller, a Chinese one, have 3 speed switch 80/100/120%. I'm not sure how this works, does it limit the throttle, amps? When no wires are connected I guess it is the 100% mode. In 80% mode there is no sound I've mentioned before. Video in mode 100%, no 3 speed selection wires connected, sound like starting jet
https://www.youtube.com/watch?v=0KAGwGwm35I

I was planning to use 201 rpm version, but I did a test on 28" bike, 328 rpm, 52v 17amp and was positively surprised. On straight I was easily going 45 km/h, my speed was limited by the sound above. Uphill performance was also very good. The start of my commute hill is flat so I've got into speed, then I was able to hold 30-22 km/h on a 7-14% steep hill. Note that I'm fit and was pedaling. I'm no longer so sure that 201rpm is the right choice for me

The 328 one only if you are a very fit strong leg rider , the 328 does not do well when riding into a strong headwind so keep that in mind. When going up hills , especially with a 328 version keep the speed up to at least 20 kph, better to keep it 24 kph or higher, so you better have strong legs and lungs.

styczoo said:

I'm no longer so sure that 201rpm is the right choice for me

Click to expand...

Scooterman is right, 328 Q100's are too high geared for probably 90% of riders and their routes. In a big wheel, they really never get up into the strongest part of the powerband. Another way to put it is, when something starts to slow the motor down (headwinds, hills, etc.), they are already down into the rpm range and don't have to far to fall before trouble starts.
The "trouble" is, when your system "sees" a low rpm/high load situation, the controller tries to "feed" the motor every Amp it can and this overheats the Phase wires and the connectors can melt. Then the phase wires themselves and eventually the windings can cook.
There are some things you can do;
1)First and foremost, NEVER let the hill climbing speed fall below half of the "over the road" top speed! If, when climbing, you feel like the bike is going to come to a stop no matter how hard you pedal, Jump off first! Letting a small hub motor "chugy chugy" to a stop is almost sure death.
2)Keep the max Amp rating of your controller low. 17A is good, don't add more.
3)Replace the Kit's bullet-type connectors on the phase wires, they are heat-sinks. Use quality connectors, or better yet, solder them.
Armed w/ this info, your system might just last a while.

The above is the reason that for the large majority of Q100 builds, the mid-speed 260 is the best speed range. It's just the best compromise, and to tell the truth, I've used both the 201 and the 260, and I think the 260 climbs just as well on the same control/batt and has the advantage of being a couple of mph faster.
But, unfortunately, the 260 is not avail. as a "C" model.
A couple of years ago, I opened up some motors to see if I could put 260 gears in a C motor. In the end, I decided I could make it work, but I would have to use two motors to make one. Not really worth it.

Yes I know that all, I've been reading a lot before deciding which version to buy. Bought 201, but got 328 for testing and I'm surprised how good it performs in my case.
What is worrying me is the sound at max speed which is 44 km/h 328 rpm's hmm... I guess with my voltage the unloaded wheel should spin at 68 km/h according to
https://www.ebikes.ca/tools/simulator.html?bopen=false&motor=MCUTE36V_328&batt=cust_52_0.2_9&wheel=700c&hp=150&black=grade&cont=cust_17_34_0.03_V&mass=95&frame=mountain&k=1&autothrot=false&throt=100&grade=0

So to avoid overheating I should go at least 38 km/h full throttle? Right?

Sorry, but without actually hearing the sound it's making, it's hard for me to comment on what it might be. As always w/ something like that, the first step is to determine if the noise is mechanical or electrical related.
Your sim. plot well represents the problem w/ a Q100 328 in a big whl.
(I changed a couple of things, removed the rider input and changed to mph).
First off, the top system speed of 25 mph is what I got on 44V w/ a 24" wheel.
Notice how much of the power curve is beyond the reach of the system. This is a power (Current) limited system, rather than a rpm limited system. What it wants is more Amps. Plug in the Grin 25 Amper and the top speed jumps up to 28 mph! It now is seeing enough Amps that the system has become rpm limited (note that the load point sits right on top the red mountain of the power curve).
The problem is, you are now pushing over 1300 Watts thru the diminutive Q100C. This is "melt things" territory. Also note that the efficiency is rather poor, at 74 %.
The saving Grace for the strong rider is, that w/ the 25 mph max speed point, the power curve extends beyond it in more or less a plateau and it is possible for a strong rider to add 2, maybe even 3 mph w/ his legs, combining his input w/ the motor power just below the surface, so to speak.
One of my first 2WD builds was two 328's where I tried to make use of the "out of reach power curve", the second motor acting like a pr. of strong legs. It had good jump at start and by the time it got into the 20's mph range, the two motors sync'ed up and helped pull each other up to a scary 28 to 29 mph top speed. But it was lazy in the 5 to 20 mph mid range and since it was heavy (2WD) and I rode it like a mini-bike, I didn't like that. Since then, my 2WD's are always dual 260's. They only top out at 25 mph, but they pull strong thru the entire power range.
I suspect, that if you changed nothing other than putting the motor in a 24" wheel, the bike would climb stronger, be more efficient and faster as well (the sim confirms that; 25.5 mph & 77 % efficiency) . But I understand that, for several reasons, that is something you don't want to do.

Video with sound: https://www.youtube.com/watch?v= I've tested again and the free spinnig wheel speed is actually 58 km/h at batterie charged to 54v. I'm not going that fast anyway