ASIMount-ZWO
Interesting. You are advertising the AM5N with the following information:

Let's calculate the torque for the specified C11 setup that is shown in this picture in detail (downloaded from the ZWO homepage today):
- C11 Tube: 11.28kg
- Finderscope: 0.50kg
- ASIAIR+: 0.22kg
- ASI585MC: 0.13kg
- Cables: 0.15kg
Considering the total weight of 12.28kg and a lever arm of about 28cm (from the RA-axis to the center of gravity of the C11) the calculated torque without using a counter weight is:
12.28 * 9.81 * 0.28 = 33.73Nm (this exceeds the stated maximum output torque by 12%)
Nonetheless, I already found the root cause of my RA spikes/issues. I disassembled my mount even further and found out that the breakaway torque of the RA harmonic drive is higher compared to the DEC harmonic drive. The DEC harmonic drive also runs smoother and is clearly easier to move. For me this was only recognizable when removing both motors and belts.
So I decided to interchange both harmonic drives and afterwards my RA issues were immediately gone. Now I partially have issues in DEC (what I was expecting somehow).
Please respond how to proceed. I need a replacement harmonic drive for the DEC-axis and as you can read - I am capable to change it myself. You can also respond to my support inquiry/ticket (that is still unanswered).
BR Andreas