digital servo refresh rates

[markb50]

I got a trex760x top combo for Christmas and doing fbl setup with ikon fbl unit. These servos 820m and 825m have refresh of 333, are there any advantages or disadvantages of setting refresh rate lower say 200. I fly mild 3d and am more interested in longevity then super speed. I have seen a comment here and there about it but I can't seem to find any info about it on vender sites. Any info would be appreciated.

Thanks

[Andy from Sandy]

I cannot find it at present but Mikado has a document that discusses the pros and cons of refresh rate.

They use 166Hz for the cyclic servos, either way it is not user adjustable. This is Mikado's control loop and others will be doing something different no doubt.

[ssschmidt4]

On my Trex 700x top combo I ran the cyclic servos on 200 and the tail at 333 (Ikon2). I have MANY flights on the heli with no problems. I now have a Vbar Neo and I am not sure what the cyclic refresh rate is as you can not change it in the settings. Anyways if it is different than 200, I can not tell a difference.

[Andy from Sandy]

I now have a Vbar Neo and I am not sure what the cyclic refresh rate is

Just read the post before your one.

[motoxxxman]

Originally Posted by markb50

I got a trex760x top combo for Christmas and doing fbl setup with ikon fbl unit. These servos 820m and 825m have refresh of 333, are there any advantages or disadvantages of setting refresh rate lower say 200. I fly mild 3d and am more interested in longevity then super speed. I have seen a comment here and there about it but I can't seem to find any info about it on vender sites. Any info would be appreciated.

Thanks

The servos are rated at 333hz, but align/beastx recommend running 200hz on the cyclic servos. And 333hz on the tail servo.
Higher frequency supposedly is more precise, but can also cause response issues and overheating especially when running high BEC voltage. Dropping cyclic to 200hz you will never notice a difference in performance, but the servos will last longer and less likely to have wobble issues.

[markb50]

Thanks for the info

[scorlett]

I've been flying that setup (w/CGY760r) for a couple of months. Let me know what you think.


As of last month:




Sam

[markb50]

Nice!! I just maidened mine 2 days ago flew great.

[Atomic Skull]

It depends.


In theory a lower refresh rate may prolong the life of your servo that depends on the design of the servo and the FBL. If the servo always runs the motor at say 500hz or whatever no matter the speed of the control signal it's not going to make much difference though it still might if the FBL also does not filter the sensor input from the yaw channel and passes through every tiny vibration to the servo control loop as a control change. Most FBL units have some sort of filter on sensor input to avoid overworking the servos due to vibration being seen by the control loop. The idea is that the filter will average out tiny changes in attitude from vibration and make them invisible to the PID loop while passing though large changes from actual movement.

If the motor control rate varies with the speed of the control signal (i.e. matches it) then it will probably increase the motor life to lower the control signal speed at the expense of performance. But if you can't tell a difference between e.g. 250hz and 333hz then you weren't gaining anything from running it at 333hz anyway except a shorter motor life.

Even if the motor runs at a constant refresh rate using a lower frequency control signal might increase the lifespan of other components on the logic board but I doubt it makes much of a difference. also keep in mind that MOST digital servos run the motor at a constant refresh rate no matter the frequency of the control signal. Also some digital servos have a maximum control signal frequency that is less than 333hz. the BMS127WVs I have on my Oxy2 have a max control signal frequency of 250hz according to Blue Bird's specs and there are a few digital servos out there that can only run at 40hz-60hz. (basically they can only be run directly off a receiver PWM signal and not in any sort of "fast" mode)

[ticedoff8]

These servos 820m and 825m have refresh of 333, are there any advantages or disadvantages of setting refresh rate lower say 200.

The "refresh rate" is how many times per second the iKON2 will adjust the servo's position signal.
On the brushless or coreless servos, when the signal is off, the servo arm will be free to move under a load (like a swashplate).
On a slower refresh, the servo arm will move farther than with a higher refresh.
The question is: Will you notice it?
The higher refresh basically updates the servo position 333 times per second. That can cause heat and draws more power.
Most people will not notice the difference between 150Hz and 333Hz.
The key is to never run a servos rated at a lower refresh rate at a higher refresh rate.
I set all my brushless servos to 200Hz.

[Atomic Skull]

Originally Posted by ticedoff8

The "refresh rate" is how many times per second the iKON2 will adjust the servo's position signal.
On the brushless or coreless servos, when the signal is off, the servo arm will be free to move under a load (like a swashplate).
On a slower refresh, the servo arm will move farther than with a higher refresh.

No that's not how digital servos work the motor refresh rate is not coupled to the control signal like it is with analog servos. Digital servos use a PID control loop (just like the PID in a tail gyro) and refresh the motor at a fixed rate. The pulse width of the PWM control signal is only used to determine what position to move the servo arm to. The microprocessor running the PID is what actually refreshes the motor and it does this at a constant rate. Some servos use a high enough refresh rate that the whine they make under load is out of the range of human hearing which is why some servos whine and some don't.

[ticedoff8]

Wikipedia:

When these servos are commanded to move, they will move to the position and hold that position. If an external force pushes against the servo while the servo is holding a position, the servo will resist from moving out of that position. The maximal amount of force the servo can exert is the torque rating of the servo. Servos will only hold their position for their timeout duration, though; the position pulse must be repeated, usually within 20ms (50Hz), to instruct the servo to stay in position.

So, this may explain why I was thinking that the "off-time" of the digital servo would allow for micro-movement of the arm until the next refresh came in.
But, I cannot find anything that implies there is a PID gain loop inside the servo.

[Atomic Skull]

Originally Posted by ticedoff8

Wikipedia:
So, this may explain why I was thinking that the "off-time" of the digital servo would allow for micro-movement of the arm until the next refresh came in.
But, I cannot find anything that implies there is a PID gain loop inside the servo.

I recall Extrapilot said that digital servos use PID control and a simple microprocessor. That's what makes them "digital" and not analog.