Stealthmax v2 ESP Home Controller

This project aims to create an alternative controller for the Stealthmax V2 based on an ESP32 micro controller using ESP Home. It offers native integration into Home Assistant, for improved reporting and more granular automation control, while also being able to interface to Klipper using the ESP Home API.

This project assumes a level of familiarity with setting up, flashing and configuring ESP32 devices with ESP Home.

Why?

Firstly, I am a huge fan of home assistant, so I wanted to be able to gather historical sensor data in it for visualisation and to judge my carbon effectiveness over time.

In addition, this avoids any direct interaction between the filter controller and klipper. There is no custom code to install in klipper, no separate MCU to link klipper with, no potential for the controller to cause klipper shut downs, no USB connectivity to klipper required etc.

On the flip side, this is very much a skeleton set up and just does three jobs:

1. Controls the servo vent (open or closed)
2. Reports on VOC values and intake and exhaust VOC deltas to gauge carbon effectiveness
3. Adjusts filter fan speeds with the fan directly connected to the printer's main board.

Having the setup decoupled from Klipper and focused on its core jobs was appealing to me, hence this setup.

Important notes on the VOC sensor values

The stock ESPHome SGP4x component does not expose the raw VOC sensor values for use. What this means is that over long prints the VOC values will converge to 100, as the software, over time, calibrates itself to the increased ambient VOC values, which it considers the new "baseline".

However, this is of no benefit for use in a 3d printer! The VOC sensors are used as a proxy to illustrate the carbon effectiveness and to identify when/if carbon replacement is needed. To achieve this, we require the raw VOC sensor data and to calculate the VOC drop between intake and exhaust sensors.

To do this, this configuration calls a custom external component that

1. Extends the stock ESPHome module, exposing the raw sensor values.
2. Implements a "calibration" routine, which creates a static baseline (offset) for the intake and exhaust sensors, which is triggered on demand.
3. Implements a simple sigmoid function to report the statically calibrated sensor values to a range that is familiar and similar to the stock VOC read outs.
4. Calculates the delta between the statically calibrated intake and exhaust VOC sensor values and shows the "raw" delta - ie not adjusted through the sigmoid function.

The user needs to set the "baseline offset" for the intake and exhaust sensors, ideally at the start of the print, after the chamber is heatsoaked and before the print begins. This allows to monitor the effectiveness of the filtering media through the VOC drop between the two sensors as reported in the "delta" sensor value.

The component exposes all of those sensors, the baseline offsets and the raw data in the Home Assistant UI as below.

The sensors section contains the data useful for automations:

1. The VOC value suffixed by Manual Calibration is the manually offset value that is set to 100 when the set baseline button is triggered.
2. The non-suffixed values are the stock reported sensor values.
3. The Intake - Exhaust VOC value is the raw delta between the two sensors.
4. All sensor values are compensated for temperature and humidity.

Bill of Materials (BOM)

1. ESP32 D1 Mini, or similar ESP32 micro controllers
2. A buck-boost converter to power the ESP32 from the 24V printer power supply
3. The standard VOC and Temperature sensor stacks for Stealthmax (sgp40 and bme280)
4. The standard BOM servo (FT90M)
5. The standard Stealthmax V2 BOM Fan (24V - MX7020GBH2)

Hardware Setup

Power Wiring

1. Connect your Stealthmax fan PWM pin (yellow) to your printer's main board PWM fan control header pin
2. Connect your Stealthmax fan Tach pin (blue) to your printer's main board tach header pin
3. Connect your buck-boost converter inputs to the printer's 24V power supply output. Trim the converter pots so you get 5V output.
4. Connect the fan power (+) and (-) wires to the input of your buck-boost converter (24V).
5. Flash your ESP32 controller over USB with the included configuration ESPHome YAML file. Disconnect from your computer.
6. Connect your ESP32 controller to the 5V output of your buck-boost converter. I used a micro USB cable cut to size in my setup above.
7. Boot the system and verify that the PWM fan works correctly in Klipper. Verify that the ESP32 device boots and you can see it in Home Assistant.

Sensor Wiring (3.3V & GND)

Stack	SDA	SCL
Intake	GPIO23	GPIO18
Exhaust	GPIO19	GPIO26

⚠️ Note: Some BME280 sensors (eg. AZ Delivery ones) require I²C address 0x76 (default is 0x77, which is used in Isik's combo sensors)

⚠️ Note: On boot up the sensors take approximately 90 seconds to stabilise before they start reporting values.

⚠️ Note: To be able to adapt the stealthmax fan speed automatically based on chamber temperature, you'll need a separate chamber thermistor as the BME values cannot be used in klipper directly.

Servo Wiring

• PWM → GPIO16
• Power from ESP32 5V/GND or from the buck-boost converter output

Servo Calibration

1. Power ESP32 without servo connected
2. Set vent to OFF in ESPHome/Home Assistant
3. Connect servo + install flap in the closed position.
4. Set vent to on → verify no mechanical overtravel
5. Adjust servo_vent_closed/servo_vent_open in YAML if needed
6. Alternatively, use the Servo Control slider to manually set an arbitrary servo % value. Note the servo closed and servo open values, divide by 100 and set them in the above YAML placeholders. The slider is set conservatively, so if you need more range update the corresponding value substitutions in the header.

Software Setup

The included configuration exposes the temperature, humidity, VOC and servo sensors and switch to be used in automations in Home Assistant and via Klipper.

Klipper Macro Setup

1. Install Gcode Shell Command via KIAUH
2. Setup the included macros in Klipper, replacing the IP below with the IP of your ESP32 device. The included macros allow you to open/close the ventilation flap and to set the VOC baseline ("zero out" the sensors) during your print start routines.
3. In addition the included macros allow you to set up dynamic fan speed and vent % control subject to the printer chamber temperature. See section below.

Klipper Macro Integration

Servo Vent

Firstly integrate the servo activated vent and VOC baselining in your print start macro.

My prefered method is to check the requested chamber temperature by the slicer. If, for example it is over 20C, this means you are printing a material that needs enclosed printing, hence close the flap.

After all print preparation is done but before the hotend is set to the print temperature, you can optionally sync the intake VOC baseline to the exhaust so the values match before printing commences.

{% if target_chamber|int > 20 %}
  CLOSE_VENT
  .... preheat chamber, bed etc ...
{% else %} 
  OPEN_VENT
  .... preheat bed etc
{% endif %}

... QGL ...
... Bed Mesh ...

 sync_intake_voc_baseline_to_exhaust # Optionally equalize VOC readings before a print starts

... Heat hotend to print temp...

Fan Speed Control

Firstly set up the PWM fan as below in your printer.cfg file:

[fan_generic exhaust_fan]
pin: PA6
hardware_pwm: true
tachometer_pin: PC2
tachometer_ppr: 2
max_power: 0.8 # 100pc is way too fast and unecessary
shutdown_speed: 1.0
kick_start_time: 1.0
off_below: 0.10

Then set up the fan control routines as below in your print start macro

[gcode_macro PRINT_START]
gcode:
....

{% if target_chamber|int > 20 %}
  CLOSE_VENT
  SET_FAN_SPEED FAN=exhaust_fan SPEED=0.3 # approx 3k RPM
{% else %}
  OPEN_VENT 
  SET_FAN_SPEED FAN=exhaust_fan SPEED=0.3 # approx 3k RPM
  UPDATE_DELAYED_GCODE ID=adaptive_exhaust_fan DURATION=30
{% endif %}

... QGL ...
... Bed Mesh ...

 SET_VOC_BASELINE

... Heat hotend to print temp...

The above snippet sets the exhaust fan to 30%, which is approximately 3k RPM when printing ABS/ASA/high temp material. If we are printing a low temp material, it calls the adaptive exhaust fan delayed gcode that gradually ramps up / down the fan depending on chamber temperature to allow the chamber to remain below 40C.

In the print end macro, it calls the cancel_adaptive_exhaust to stop adapting the exhaust speed and turn off the fan.

In addition the adaptive exhaust macro also adjusts vent open % (25-50-75-100) subject to chamber temperature conditions.

You can view my complete configuration here: https://github.com/igiannakas/Voron-backups/tree/main/printer_data/config