Version 2.1:

Major governor algorithm improvements:

◦Performance was greatly enhanced by adding dynamic factor into the regulation loop.

◦Governor is not dependent on the RPM Sensor anymore.

◦In case of RPM Sensor malfunction, governor will still work (with lower performance).

◦RPM Sensor failure reported in the Log.

◦RPM data noise detection feature is more precise. It can distinguish a real noise in a better manner.

◦Head speed can be changed in any situation with no performance influance.


I have version 2.1 installed in two helis and both showed rpm sensor data issues in their logs:

00:00 Calibration Finished
00:10 Good Health Message
00:20 RPM Sensor data are lost
00:30 Governor was Engaged
00:40 RPM Sensor data are too noisy
00:50 RPM Sensor data are too noisy
01:00 RPM Sensor data are too noisy
01:10 RPM Sensor data are too noisy

The "Sensor data are too noisy" continued for the whole flight.

The first heli is a Gaui X3 with a regular Spirit, HobbyWing 50A V3, and Align 460MX. The second heli is a Protos 500 with a Spirit Pro, YGE 90LV, and MSH/Scorpion 0880kv. I added a small ferrite ring to the end of the sensor cable of both escs, and a capacitor to the power pins of both Spirits. On the Gaui I also separated the sensor wire from the power wires and ran the power wires down one side of the frame and the sensor wire down the other side.

While bench testing the rpm can be seen jumping around. It was constantly and quickly changing rpms so it was difficult to see the range of rpms but it was definitely more than 20 or 25 rpm. I have also set the rpm sensor filter to 8. On the Gaui I also experimented with various timing and PWM frequencies with no effect on the rpm sensor errors.

In flight the rpms seem to be holding. However I don't have any previous experience with governors so I am not the best judge of governor performance. I don't have the skill fo 3D but doing pitch pumps, loops, flips, and rolls, the head speed seemed to hold. I have been flying with 6 - 5 and also 7 -5 for Governor Response - Holding Performance. I have also tried various values for both with no effect on the rpm sensor noise issue.

So with the RPM sensor data warning messages that I am getting, do I have a problem and should I stop using the Spirit governor? Also how does this relate to the Version 2.1 algorithm improvements:

◦Governor is not dependent on the RPM Sensor anymore.

◦In case of RPM Sensor malfunction, governor will still work (with lower performance).


I have also read all previous posts that I could find regarding this rpm sensor noise issue and it seems that there are helis with no issues and helis with issues but no solutions. Unfortunately both of my helis with governors have issues and I am at a loss on how to resolve the issues ... so frustrating.

I know it sounds simplistic... but if it flies fine, it holds, you like it... why should you care about somr warnings in a log file? Messed up priorities?

On second thoughts... get some telemetry so you can verify the RPM behaviour in flight...

_________________
Oxy 2 ♦ X-Vert ♦ Warp 360 ♦ Blade 450X ♦ Blade 450X Nightflyer ♦ Goblin 500 (RIP) ♦ 2 X mCPX v2 ♦ night fly mQX ♦ LadyMQX ♦ nano QX ♦ nano QX FPV ♦ Inductrix ♦ DX8 G1

I'm getting noisy signal on my HW130hv but it flies well and the RPM hold well as shown by my telemetry. So if it flies well just use it

No, it's not messed up priorities. The warnings were put into the Spirit by the designers for a reason I would think so I would like to know their significance, and I would also like to know a bit about the governor not relying on the rpm sensor anymore. There are probably other people who are curious too.

I have however flown my helis as is with the warnings.

Hello,

the problem could be caused also by very high governor gains, especially Governor Response value.
If the value is too high, motor can perform erratic changes so there will be noise created just by this.
The RPM may jitter just because it is trying to hold it very precisely. So you can try to decrease the value and see what is difference.

If sensor data are lost, the governor is working in the way that as it was programmed with V throttle curve and mixes from cyclic.

If you are seeing a noise warnings, it does not mean governor will work bad, but for sure it might be better.
When updating from 2.0 governor gains can be just too high, so you can try to decrease it.

If RPM Sensor filter does not help, I recommend to set it as low as possible.

_________________
Spirit System developer

Thank you for the information, Tomas.

I have only flown my helis with the Version 2.1 governor. I was setting up the Version 2.0 governor when Version 2.1 was released so I updated the units before flying.

I did try lower response values to start out with during bench tests and the warnings were there. I also tried a variety of values in the field. However, I will try a variety of values of response and holding performance in the field again. Unfortunately I do not have telemetry in my helis so it will be assessed by sound and feel.

After following Tomas' advice, I no longer get sensor data errors on both helis. Previously I had continuous errors for the whole flight.

I lowered RPM Sensor filter from 8 to 6 and I ended up with Governor Response of 5 and Holding Performance of 3 for both helis.

I did get continuous errors one time when I went airborne before the headspeed had stabilized. I have the Spoolup rate set to fast and it takes about 10 to 15 seconds to reach the governed headspeed on the X3. I may try Very fast Spoolup rate because the Fast spoolup rate on the X3 seems too slow for me.