> For the complete documentation index, see [llms.txt](https://yagsl.yassrobotics.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://yagsl.yassrobotics.com/how-to-guides/tune-pidf-gains.md).

# Tune PIDF gains

{% hint style="warning" %}
Swerve steering/angle/azimuth motors have PID wrapping enabled at the topmost and bottommost point (e.g. "front" and "back"). When tuning angle PID you may want to test with left or right translation so the module doesn't cross the wrap point.
{% endhint %}

## The basic idea

YAGSL reads PIDF values from [`pidfproperties.json`](/reference/json-schema/pidfproperties-json.md) in your module folder — generate or edit it with [**config.yagsl.com**](https://config.yagsl.com).

This has been hashed and boiled down to the simplest form like this before:

* **P**: if you're not where you want to be, get there.
* **I**: if you haven't been where you want to be for a long time, get there faster.
* **D**: if you're getting close to where you want to be, slow down.
  * [Video](https://www.youtube.com/watch?v=qKy98Cbcltw) found by 78, found by [this CD post](https://www.chiefdelphi.com/t/finally-i-understand-pid/450811)

<figure><img src="/files/BWjBb2Ts88krC8jXy2aK" alt=""><figcaption></figcaption></figure>

## Tuning procedure

Another good way to describe it, [from CTRE](https://pro.docs.ctr-electronics.com/en/latest/docs/api-reference/device-specific/talonfx/closed-loop-requests.html):

Manual tuning typically follows this process:

1. Set $$P$$, $$I$$, and $$D$$ to zero.
2. Increase $$P$$ until the output **starts** to **oscillate** **around** the setpoint.
3. Increase $$D$$ as much as possible **without** introducing **jittering** to the response.

For the drive motor, also set the feedforward terms (`s`, `v`, `a` — see below) before fine-tuning `P`; a good feedforward means `P` only has to correct small errors.

## WPILib walkthrough

WPILib has lots of great documentation on PID, so it's worth reading directly:

{% embed url="<https://docs.wpilib.org/en/stable/docs/software/advanced-controls/introduction/introduction-to-pid.html>" %}
Basics of PID
{% endembed %}

{% embed url="<https://docs.wpilib.org/en/stable/docs/software/advanced-controls/introduction/tuning-flywheel.html>" %}
Tuning velocity, like the drive motors!
{% endembed %}

{% embed url="<https://docs.wpilib.org/en/stable/docs/software/advanced-controls/introduction/tuning-turret.html>" %}
Tuning position, like the steering/angle/azimuth motors!
{% endembed %}

## The current `pidfproperties.json` shape

{% hint style="info" %}
The schema changed in YAGSL 2026.8.05 — `f` and `iz` (integral zone) and the old `output` clamp are gone. Feedforward is now expressed as `s`/`v`/`a` (`kS`/`kV`/`kA`, matching WPILib's `SimpleMotorFeedforward`). See the [schema changes reference](/reference/schema-changes.md) if you're migrating an older config by hand.
{% endhint %}

```json
{
  "drive": {
    "p": 0.02,
    "i": 0,
    "d": 0,
    "s": 0,
    "v": 0,
    "a": 0
  },
  "angle": {
    "p": 0.01,
    "i": 0,
    "d": 0,
    "s": 0,
    "v": 0,
    "a": 0
  }
}
```

If you leave drive `v` (kV) at `0`, YAGSL auto-derives it from the drive motor's free speed and your gearing, so a reasonable feedforward exists even before you've tuned anything by hand.

You can also add a `pidfproperties_sim.json` alongside it — YAGSL will load that one instead when running in simulation (`RobotBase.isSimulation()`), so you can use looser gains in sim without touching your competition robot's config.

## Starting points

The right PIDF values vary per-robot, but these are reasonable starting points to iterate from, organized by motor controller type.

{% tabs %}
{% tab title="SparkMax / SparkFlex" %}

```json
{
  "drive": {
    "p": 0.0020645,
    "i": 0,
    "d": 0,
    "s": 0,
    "v": 0,
    "a": 0
  },
  "angle": {
    "p": 0.01,
    "i": 0,
    "d": 0,
    "s": 0,
    "v": 0,
    "a": 0
  }
}
```

{% endtab %}

{% tab title="TalonFX" %}
When the drive motor `type` for every module is `talonfx_*` (or alike — `krakenx60`, `falcon500`) this is a reasonable starting point:

```json
{
  "drive": {
    "p": 1,
    "i": 0,
    "d": 0,
    "s": 0,
    "v": 0,
    "a": 0
  },
  "angle": {
    "p": 50,
    "i": 0,
    "d": 0.32,
    "s": 0,
    "v": 0,
    "a": 0
  }
}
```

{% endtab %}
{% endtabs %}

Once you have a reasonable starting point, use [**config.yagsl.com**](https://config.yagsl.com) to regenerate `pidfproperties.json` with your updated numbers, or hand-edit the file directly — both work, since the parser only cares about the final JSON.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://yagsl.yassrobotics.com/how-to-guides/tune-pidf-gains.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.