"P is for acceleration, D is for braking" is just a simplified way of saying it to get started. The D term usually resists gyro changes, while the P term depends on the instantaneous error between setpoint and gyro: when the body falls behind the command, P pushes it to catch up, and when the body rushes past the command, P also brakes in reverse. The video uses clips with the same number of P and D in the black box to illustrate that the two can jointly pull the body back to the target during overshoot and external disturbance; however, the so-called "as large as possible" is always limited by vibration, noise, motor temperature and frame conditions.
P doesn't just make the plane accelerate
The P term works based on the error between setpoint and gyroscope feedback. After hitting the stick, the command usually changes first, the body responds later, and the gyro lags behind the setpoint, so the P output pushes the body to accelerate and track; this is a common source of "P acceleration". But when the body's angular velocity rushes in front of the setpoint, the sign of the error reverses, and P also reverses. At this time, it actually helps decelerate.
The D term focuses on the rate of change of the gyroscope signal and has an inverse effect on rapid changes, so "D acts like a brake" is still a useful intuition. When understanding the two, don't memorize a permanently fixed direction, but look at the signs and amplitudes of setpoint, gyro, P and D at the same time.
Why do P and D move in the same direction in the black box?
The video shows a log: the gyro curve runs in front of the joystick setpoint, indicating that the aircraft is turning faster than the command requires. At this time, P requires the body to decelerate due to tracking error, and D also applies damping because the angular velocity is still changing. The two outputs have the same sign. Rather than canceling each other out, they work together to pull the gyro back to the setpoint.
This phenomenon often occurs in action overshoot or external disturbance. When the pilot keeps the joystick stationary, wind gusts, propwash or mechanical disturbances will still cause the gyro to deviate from the target; P pulls the body back according to the deviation, and D suppresses the change speed. The cooperation of P and D determines whether the body can quickly return to the command without swinging back and forth excessively.
"simultaneously increasing" should be understood as finding available control margin
Increasing P can enhance the correction of tracking errors, and increasing D can increase the damping of fast movements and overshoots; on an aircraft with good mechanical conditions and controlled noise, both have the opportunity to make the response tighter and resist interference stronger. But they are not always better: if P is too high, high-frequency oscillation or rebound will occur, and if D is too high, gyro noise will be amplified, motor heating will increase, and control margin will be consumed.
Repeatable actions and black box records should be used when adjusting parameters. First check the frame rigidity, propellers, motors, soft installation and filtering, and then change P/D in small steps. Each round compares the gyro's rise, overshoot and settling times to setpoint and immediately checks the motor temperature. Different sizes, weights, firmware versions, filtering and Rates will change the upper limit. The video explains the control mechanism, not a set of values that can be copied across machines.
Operation steps
- Put four key signals in the same view
View setpoint, gyro, P and D at the same time in the black box, first confirm whether the aircraft is lagging behind the command or has rushed ahead of the command.
- Find the fragment with the same number of P and D
When the two have the same number, check whether the gyro is ahead of the setpoint; this usually means that P is also assisting in braking, rather than just accelerating.
- Associating overshoot with control output
Observe the speed of gyro's return after crossing the setpoint, the number of oscillations and the stabilization time, and determine whether the corrections of P and D are sufficient.
- Use external perturbations to test synergy
Record propwash or gust segments in safe flight, confirming that P pulls back bias and D suppresses rapid changes, rather than relying on a single parameter.
- Adjust in small steps and check heat and noise
Only make small changes at a time, and retest with the same actions in each round; roll back immediately when high-frequency oscillation, noise increases, or the motor becomes obviously hot.
FAQ
Does the P option always accelerate the aircraft?
No. P depends on the error between setpoint and gyro; when gyro lags behind, it pushes to catch up, and when gyro leads, it helps decelerate in reverse.
Does the fact that P and D have the same number in the black box mean an error?
Not necessarily. When the body overshoots or is disturbed, the two may simultaneously exert force in the direction of restoring the target. This is a normal cooperative control phenomenon.
Why can't both P and D be raised all the time?
P Limited by stability and oscillation, D also amplifies high-frequency noise and increases motor temperature. The appropriate upper limit depends on the chassis, filtering, firmware, and flight behavior.
What should be checked before adjustment?
First rule out loose frame, damaged blades, motor bearings, abnormal vibration and inappropriate filtering; if the mechanical noise is not resolved, the risk of simply increasing D is very high.
Full timeline transcript
Transcripts are arranged according to video time, making it easy to quickly locate the explanation content. Transcript language: Simplified Chinese.
Hello everyone, I am Quarank. Welcome back to my channel. In today's episode, we are going to answer a very interesting question from a listener. That is why we need to increase P and D at the same time when adjusting PID parameters. We often say that P is to accelerate the aircraft and D is to brake the aircraft, right? But in fact, this statement is a bit biased. Why is it said that D is still the brake?
Because after Betaflight 3.5, we D will always pay for the signal of this top D. So there is no problem if D is used for braking, because it always counteracts the gyro D. But it is a bit biased to say that PID is used to accelerate. Why? Our PID is used for acceleration when the PID is activated, because at that time our set point runs faster than our gyro, right? Because the aircraft has a delay in its response.
But in some cases, this P can actually be used to brake. For example, let's look at a section of my own flight schedule on the screen. Please pay attention to the red line here. The red line is P. The red line is D. I found that their two symbols are actually the same. The same symbol means what it means. This means that the direction of force of P and D is the same, that is, they want to turn the plane in one direction. So why does this happen? We can see that the very interesting thing is that here I have this
The Gyro curve is the blue-purple curve, and the green is the Setpoint curve. The Gyro line ran in front of the waist line and stayed in front. It reacted faster than I asked it to go. At this time, our P and D are in the same direction, and their strength direction is to slow down the aircraft.
In this case, if I increase P and D, can the paper-colored line return to the green line faster? Look at P and D. We usually say that they often exert force in opposite directions, but in some cases, P and D will also exert force in the same direction.
When does this happen most often? One situation that occurs is when the propellers are being washed. Let’s think about it. When we are washing the propellers, suppose that the rocker of our aircraft does not move, but because the aircraft is affected by external forces, its gyro curve will swing left and right, right? When it swings in one direction, do you think that the P term will also pull it back at this time, and the D term will also prevent it from swinging in that direction?
This is my understanding of the role of simultaneously improving P and D in the PID algorithm. I hope I can answer this listener’s question. That’s it for today’s video. Everyone is welcome to like, comment, retweet, and connect three times with one click. Post it as soon as possible and we will see you in the next video.
