eee291
100 kW
They also have the option on AliExpress now.
https://www.aliexpress.com/item/1005003835665401.html
https://www.aliexpress.com/item/1005003835665401.html
NEW QS 165 MOTOR
- Thread starter illegalbike
- Start date
owhite
100 W
- Joined
- Aug 3, 2020
- Messages
- 285
Be advised people that I have an encoder version of the QS165 and I'm getting line noise in the connection between my motor controller and the encoder. This could very well be associated with my weirdo controller and the encoder, but it also definitely looks like the line into the encoder is making the problem worse.
This is mostly showing up when I put an oscilloscope on the lines going to the encoder to track down the source of the noise. Right now the noise on the lines are so bad the controller is not able to get correct counts from the motor -- so no joy on running it in sensor mode.
IllegalBike has put together a hall version for the motor and obviously I will consider something similar.
This is mostly showing up when I put an oscilloscope on the lines going to the encoder to track down the source of the noise. Right now the noise on the lines are so bad the controller is not able to get correct counts from the motor -- so no joy on running it in sensor mode.
IllegalBike has put together a hall version for the motor and obviously I will consider something similar.
magudaman
10 kW
I getting ready to purchase the qs165 for my project and keep coming across two version, a 72v and 60v. Is there actually two different winds to this motor?
owhite
100 W
- Joined
- Aug 3, 2020
- Messages
- 285
owhite said:
EDIT: Ignore this post most of the noise was coming from my controller.
Turns out the noise in the line of my QS165 was coming from the motor itself. It was obvious something was wrong because when I was hooking the QS up to a VESC, the readings for encoder output in VESC-tool were a mess.
To find the source I powered up the encoder with GND and 5V, and the put an oscilloscope on one of the encoder outputs. If you turn the motor by hand, the line looks like what you see here:
However, when I put the motor on two different VESC-based controllers the signals were a mess:
This was easily corrected, and I wouldnt be surprised if this was just a problem with my motor. What I discovered is that there was some kind of strange electrical connection between the motor chassis and the encoder. Here's a picture of the PCB holding the encoder:
The noise stopped when I loosened the three screws and pulled the pcb away from the motor, and surprisingly the line noise did not return when I mounted the PCB back on the QS. There's some kind of adhesive that's used to hold the PCB in place -- maybe that's conducting electrical noise. No idea, but it's fixed now.
This is also good news because VESC controllers work with these encoders.
EDIT: Ignore this post most of the noise was coming from my controller.
SlowCo
1 MW
owhite said:
Great that you were able to solve the noise issue.
And if VESC controllers work with the encoder on this motor then the new Flipsky 75200 could be a nice match: https://flipsky.net/collections/ele...for-e-foil-fighting-robot-surfboard-agv-robot
magudaman
10 kW
owhite said:owhite said:
To find the source I powered up the encoder with GND and 5V, and the put an oscilloscope on one of the encoder outputs. If you turn the motor by hand, the line looks like what you see here:
However, when I put the motor on two different VESC-based controllers the signals were a mess:
This was easily corrected, and I wouldnt be surprised if this was just a problem with my motor. What I discovered is that there was some kind of strange electrical connection between the motor chassis and the encoder. Here's a picture of the PCB holding the encoder:
The noise stopped when I loosened the three screws and pulled the pcb away from the motor, and surprisingly the line noise did not return when I mounted the PCB back on the QS. There's some kind of adhesive that's used to hold the PCB in place -- maybe that's conducting electrical noise. No idea, but it's fixed now.
This is also good news because VESC controllers work with these encoders.
Really appreciate you doing this testing. I have my QS165 and a Flipsky 75200 coming hopefully next week. I will report what happens with mine, really hoping I don’t have similar issues.
Turns out the noise in the line of my QS165 was coming from the motor itself. It was obvious something was wrong because when I was hooking the QS up to a VESC, the readings for encoder output in VESC-tool were a mess.
owhite
100 W
- Joined
- Aug 3, 2020
- Messages
- 285
magudaman said:
I would not count on the QS165 encoder working with your controller. I'm still having noise issues which is shown in the video below.
Here are some settings that you definitely should use in vesc-tool:
Motor settings-->general-->general tab:
Motor type: FOC
Invert motor direction: whatever
Sensor port mode: ABI encoder
ABI encoder counts: 4096
Motor settings-->FOC-->general:
Sensor mode: Sensorless
Then, use RL and lambda to get your parameters.
To "calibrate" the encoder
Motor settings-->FOC-->Encoder
There is a little play button to perform encoder detection. The purpose of this is to get Offset and Ratio. Hit the play button, then zip over to:
Data analysis-->Realtime Data-->(Right panel) Rotor position, and make sure you turn on the "RT" button on right to start collecting real time data. Now you should be able to see your VESC collecting data. Are you seeing a lot of spikes? Is the signal stair stepping up or down? If the detection ends go back to
Motor settings-->FOC-->Encoder and take a look at your results. A good sign is if it says your Ratio: 7. See this video for more information.
[youtube]fUQKr0zElaU[/youtube].
And hey, dont panic you may have better luck than I did.
magudaman
10 kW
Well crap, I guess I could always run it sensorless. Motor is showing up Monday.
owhite
100 W
- Joined
- Aug 3, 2020
- Messages
- 285
magudaman said:
Let me emphasize:
- this will be highly dependent on your controller
- VESCs in particular have circuitry that you may need to remove or augment to work
- presumably the QS controllers work just fine
- my collective skills on de-noising could fit in a fruitfly's navel
magudaman
10 kW
Got all my stuff! Flipsky 75200 and my Qs165 with encoder. Hoping to get all setup this weekend to test everything on the bench!
75200 was $241 usd shipped with 10 day delivery
Qs165 was $411 usd shipped with 9 day delivery
75200 was $241 usd shipped with 10 day delivery
Qs165 was $411 usd shipped with 9 day delivery
magudaman
10 kW
owhite said:
Well good news. Mine seems to detect with no issues so far. I used the parameters you mentioned and ran the encoder detection. Came up with 80.4 offset and ratio of 7. The real time data looks pretty flawless and smooth. Still seems a bit glitchy on startup on the bench but I need to triple check all my settings. I can’t seem to access field weakening on my controller and have a couple other functions I would like to read about some more. But at least I can go ahead and order my frame.
owhite said:
Perhaps try adding a layer of Kapton tape between the PCB and the motor housing, and using nylon / plastic screws to secure it? (or plastic "tophat" washers between the screws and the PCB so it insulates the screw entirely from the PCB and the entire hole thru the PCB (so ti can't possibly touch any layers).
If it doesn't change anything, then it's not electrical noise conducted directly into the PCB, and is more likely to be induced noise in the wiring from encoder to controller. In that case, shielding (braided copper, foil, etc) the entire set of encoder wires from right at the PCB to at least just outside the motor, and preferably all the way to the controller PCB, grounding only at the controller PCB itself, should isolate it from induced noise.
Can somebody share his knowledge on how to connect a QS165 encoder to a non-fardriver controller? Iam currently struggling with their pinout definition 
The newest firmware for ASI BAC controllers supports encoder and I have to connect:
- Encoder Sine V source (Sine voltage output from encoder)
- Encoder Cos V source(Cos voltage output from encoder)
- VCC
- Ground
Or is the QS165 not an encoder with sine/cosine wave per mechanical revolution and thus won´t work?
Thanks in advance!
The newest firmware for ASI BAC controllers supports encoder and I have to connect:
- Encoder Sine V source (Sine voltage output from encoder)
- Encoder Cos V source(Cos voltage output from encoder)
- VCC
- Ground
Or is the QS165 not an encoder with sine/cosine wave per mechanical revolution and thus won´t work?
Thanks in advance!
magudaman
10 kW
c.wagner said:Can somebody share his knowledge on how to connect a QS165 encoder to a non-fardriver controller? Iam currently struggling with their pinout definition
QS165 encoder pinout.png
The newest firmware for ASI BAC controllers supports encoder and I have to connect:
- Encoder Sine V source (Sine voltage output from encoder)
- Encoder Cos V source(Cos voltage output from encoder)
- VCC
- Ground
Or is the QS165 not an encoder with sine/cosine wave per mechanical revolution and thus won´t work?
Thanks in advance!
It has an abi incremental encoder not a sin cos encoder. You could always move over to VSEC controller, it is natively supported.
owhite
100 W
- Joined
- Aug 3, 2020
- Messages
- 285
I generated some motor parameters for the QS165. These might be valuable to someone else so here goes:
Two methods were used to collect the data. One was using the motor detect function in vesc-tool, while the motor was connected to a 75100 ubox vesc motor controller from Spintend. The other was with test equipment. The values are very close with one exception -- resistance. The resistance of the QS165 is relatively low and the discrepancy here is probably due to added resistance of the MOSFETs in the ubox. The MOSFET resistance would not be as significant for motors with a higher resistance.
Two methods were used to collect the data. One was using the motor detect function in vesc-tool, while the motor was connected to a 75100 ubox vesc motor controller from Spintend. The other was with test equipment. The values are very close with one exception -- resistance. The resistance of the QS165 is relatively low and the discrepancy here is probably due to added resistance of the MOSFETs in the ubox. The MOSFET resistance would not be as significant for motors with a higher resistance.
Hi, I juste got an vesc controller and was wooundering if you could help me on how to connect the encoder to the vesc because the name of the pin on the connector of the motor does not match the input on the controller.magudaman said:c.wagner said:Can somebody share his knowledge on how to connect a QS165 encoder to a non-fardriver controller? Iam currently struggling with their pinout definition
QS165 encoder pinout.png
The newest firmware for ASI BAC controllers supports encoder and I have to connect:
- Encoder Sine V source (Sine voltage output from encoder)
- Encoder Cos V source(Cos voltage output from encoder)
- VCC
- Ground
Or is the QS165 not an encoder with sine/cosine wave per mechanical revolution and thus won´t work?
Thanks in advance!
It has an abi incremental encoder not a sin cos encoder. You could always move over to VSEC controller, it is natively supported.
owhite
100 W
- Joined
- Aug 3, 2020
- Messages
- 285
Progress on this frame.
[youtube]v7N6URsDw-4[/youtube]
controller is a spintend ubox 75100 with 18s batteries. VESC settings are 60 phase amps, sensorless. The vesc-tool showed it was reading the encoder fine but it was choppy when I had in encoder mode -- probably because of bad wiring. I rode it a bit but I took it easy so I didnt blow out the controller -- a better controller is coming.
[youtube]v7N6URsDw-4[/youtube]
controller is a spintend ubox 75100 with 18s batteries. VESC settings are 60 phase amps, sensorless. The vesc-tool showed it was reading the encoder fine but it was choppy when I had in encoder mode -- probably because of bad wiring. I rode it a bit but I took it easy so I didnt blow out the controller -- a better controller is coming.
illegalbike
10 W
- Joined
- Jan 28, 2021
- Messages
- 94
larsb
1 MW
That’s a nice development, the price is 2-3x that of the others though and they stepped it up to be rated at 10kW on paper. Probably correct rating but the 138 motor variants also output those numbers. Seems QS motors are starting to decrease the bang for the buck!
romelec
100 W
It's not yet on their website, and there is "No stock, CNC made available, if interest, please contact seller." in the aliexpress listing, so I guess it is a pre production price.
Edit: confirmed from QS
Edit: confirmed from QS
