Jerky motion in closed loop modes with MBL1650

I am using a 14 pole outrunner motor equipped with hall sensors and an MBL 1650. The motor is running in open loop mode but it is quite noisy. In closed loop modes (torque and speed) with default controller settings, the motion is quite jerky, sometimes the motor stops and starts again.
In open loop mode, I observe strong rpm variations (between 2700 and 3600 rpm) at full power. I assume, that these speed variations are amplified in closed loop mode and cause serious problems.

The questions are: what may be the causes of the large variation in speed in open loop mode?
I checked the hall signals with an oscilloscope, they look well.
Might a slight deviation in hall sensor positioning be the reason for "unsmooth" rotation?
If so, could you provide an option to fine tune the hall sensor angle in the controller software?

Looking forward for a quick reply, a fast solution is required.
Thanks in advance!

Yo may want to try a different UVW configuration. When properly configured, the motor should run smoothly in open loop and not very noisy at all.

Griffin, thanks for the fast reply.
Actually, we tested all UVW combinations according to the manual and found the one, that worked best. So I think, UVW order is correct.
Unfortunately I have no means to measure speed with an external tool in realtime (except from an oscilloscope, that showed varying pule width of a hall signal) however, the vibrations ("unsmooth running") introduced to the system indicate, that speed is really not constant but rapidly changing.
So, which properties of the controller-motor combination can cause these speed variations? I still suspect the positioning of the three hall sensors. How sensitive is the control against not precisely positioned sensors.
And is there a way of correction by software? Unfortunately, we do have no chance to use a different motor, so any further hints are very appreciated.

Try these values and gradually tune as needed.

Start with a low I gain.
P=0
I=2
D=0

Then increase the I gain and Small P gain as needed.

I tried to tune the controller parameters according to recommendations found in the forum, but as long the speed is not stable with constant load in open loop mode, it does not solve the problem.
With low controller gains and Loop Error Detection Off, the motor at least keeps spinning in a single direction, however, vibrations are induced and speed is not stable. Moreover, the motor does not stop in Speed Control mode, if commanded 0. If it is stopped mechanically, it is vibrating around standstill. Any hints, what that might mean? Any further ideas, how to get constant running (in open mode and closed loop modes)?

Have you changed your max speed RPM value in the speed and acceleration tab when using the closed loop speed mode?

No, max. speed and accelerations were the same in open and closed loop mode.
But regarding speed, there is another issue. With max power 100, the motor runs faster than the configured max speed, in open loop mode as well as in closed loop mode?

No, they work a bit different. In the closed loop speed mode, if you set your max speed to 100, then the closed loop speed will max out at 100 and cause a sawtooth waveform basically throttling your output.

In open loop mode, changing the max speed will determine the reaction time in which the motor command given to reach it's destination will react.

From the manual, I understood that the run command is relative to the configured max. speed.
So, configuring max. speed as 2400 should yield 1200 rpm at a command (the slider in roborun+) at 500. This is not the case in our application. I do not understand how to command a specific rpm.

The max speed merely allows for the motor to travel at the max speed set in the closed loop mode. However if your motor is 1500 rom's, then the motor can't go any higher than that if you set it higher than 1500.