Alienwhoop zero FC build - slow left yaw, yaws right on high throttle

So i finally got a camera for my Alienwhoop zero build, Cockroach frame, newbeedrone gold motors, HQ Props, low mounted AKK-SG2. Flying with GNB 300mah lipos.

It seems to fly well, but yaw left is extremely slow; Maybe half the speed of yawing to the right. And if i give it high throttle, it will yaw left, For example if i catch it after a roll or dive.

Is that a sign of bad clockwise motors? I really hope not because these are new Newbeedrone motors. It has plenty of power except for that yaw issue…

Just to finish this thread off, I had played with motor rotation and frames after this, and even though i went back to the same configuration as i had originally, it now flies nicely. Same yaw left and right, and no more yaw issue on punchouts/dive catches. I wonder if maybe one of the motor plugs wasnt making good contact, or that motor needed a pack or two to burn in.

So yeah false alarm. Time to get to work on building that NRF24L01 SPI receiver. Perhaps the Frsky XM is a better longrange receiver (certainly looks the part with that massive antenna) but no telemetry is a bummer for me. I will save that RX for a Betaflight whoop (OSD)

One thing I forgot to mention @Tokyo_Dom, using an NRF24L01 means the STM32 has to perform some extra work to emulate the XN297. This is no problem with an F4 processor, but the Zero probably uses an F0 processor.

The extra work I’m referring to is the CRC calculation of the packet payload as well as some bit twiddling of the packet data. The rx_nrf24_bayang_telemetry.c source file has comments describing some timings. 81us for reading payload, then 65us for calculating CRC, then another 29us to descramble the bits.

Hopefully it will be fine, but if this ends up causing you to exceed your looptime, it is definitely possible to eliminate that layer of the code as long if you are also able to modify the multiprotocol module source code for the Bayang protocol. If this becomes necessary don’t hesitate to reach out.

I suspect you’ll be fine but wanted to make sure you know the NRF24L01 has an extra cost (in CPU time) that the XN297 does not.

Also, the antenna mod for the micro NRF24L01 does make a noticeable difference and isn’t hard to do.

On the transmitter side? There’s a whole new side of things i havent touched :slight_smile: Its Deviation (T8SG V2), so should be open source but yeah lets hope it isnt exceeding the looptime (Yes its an F0 processor - from the product page “STM32F030F4P ARM® 32-bit Cortex®-M0 CPU, frequency up to 48 MHz (can be overclocked to 64MHz)” perhaps overclocking can be done.