Issue with FT232 Adafruit Board

[ScottGibb]

Hi All,

Thanks for putting together this board. It's a really interesting design, and I'm experimenting with it while trying to create a Rust driver for it. You can see my current progress here if you are interested: https://github.com/ScottGibb/AP33772S-rs/tree/add-dev-and-cicid-setup. I would love to link it to this repo once it's confirmed working!

However, I seem to be encountering some issues with getting the I2C to work.

My setup is as follows:

This seems to result in a NACK on the I2C Bus when the logic analyser is disconnected, and it sometimes works with an ACK when the Logic Analyser is connected.

What is interesting in a setup where there is no logic analyser in the setup. In this case, it's FT232 to RotoPD with the power supply connected, it will not ACK and I cant send I2C commands. This issue doesn't happen when connecting to the Raspberry Pi, however.

I'm unsure what to be looking for, as I said I have a logic analyser and Rigol1074Z at my disposal so happy to do further debugging. Was wondering if anyone had any advice or experienced similar?

Equipment List

• FT232H breakout Board https://shop.pimoroni.com/products/adafruit-ft232h-breakout-general-purpose-usb-to-gpio-spi-i2c?variant=1004798821
• 100W USB C Charger https://www.amazon.co.uk/dp/B08CN12NXH?ref_=ppx_hzsearch_conn_dt_b_fed_asin_title_2&th=1
• Logic Analyser https://www.amazon.co.uk/Hobby-Components-24MHz-Analyser-1-1-16/dp/B00DAYAREW/ref=sr_1_2?dib=eyJ2IjoiMSJ9.TOu1xeea4_Fgjb9CsDMEhU6U_gTqmQZA3pKKeScBIsL5Ai8zAXHaywrsA3Fc4GK0onniiQ0CuqUfSpoYteG4U3i-TNZVlBn1qdUQtO-hClXQALZlsdtiu5fXbw8EzTHSmJmAiUUmcgQcZgajInAlcd5BM9L00HTJaSbAqD0MCEyZBmkp6XHHM6WidzEWws1EL8WG60cmBQQ1qDf9k8kfPT6sbGlUkE5k1gTYIQ8O7kN0QsPVJgYJa-eV3G9c07bsow_kInAdtQ154wHk3PImT-LQgHUxcxjUuinDkF5Vk4I.5rNDJwIsdIWk-bVCOLGrsXtA2JmRSCrcHpeQiEBPm5Y&dib_tag=se&keywords=hobby+components+logic+analyser&qid=1753653084&sr=8-2
• https://www.rigol-uk.co.uk/product/rigol-ds1074z-plus-70mhz-digital-oscilloscope-mso-ready/

[CentyLab]

Hi @ScottGibb ,
Linking your repo would be the first we would do, then we thank you a lot!
At the moment, we have no knowledge how to work with Rust. I will have to look for someone to assist you on this.

[ScottGibb]

@CentyLab I managed to make some progress on this front with my oscilloscope and some trial and error. A couple of things were going on. It led me to the following realisations

• ESP32C3 testing only works with a baud rate no higher than 100_000.
• FT232H testing only works with a baud rate no higher than 10_000.

Any baud rate higher than the specified one results in a NACK.

Has anyone used/confirmed that this board can run at 400_000Hz?

When working with the FT232H Breakout board, I also found that the I2C SDA line (blue in the image below) never pulled down to 0V but rather hovered around 500mV

The above graph had the configuration of FT232H -> Stemma -> RotoPD -> Dupont Wires -> Oscilloscope probe

Using the Hobby Components Logic Analyser

I'm not entirely sure what was happening with the Logic Analyser added into the circuit. It seemed to cause reference issues with the SCL line. See the graph below for more details

This also resulted in the following capture

I'm putting this information in here for the time being, in case anyone has any ideas as to what might be going on. Hopefully will report back with more success as time goes on.

@horho77 @petekubiak adding if you're interested

[ScottGibb]

Upon even more investigation, I found that I couldn't get the FT232 at 10_000Hz to list all the Power Data Objects. For example, the following code here:

fn main() {
    let i2c = setup_i2c().expect("Failed to set up I2C");
    let mut ap33772s = Ap33772s::new_default(i2c).unwrap();

    // Read The Status Register
    let status = ap33772s.get_status().expect("Failed to get status");
    println!("Status: {status}");
    // Get the Power Source Delivery Capabilities
    let power_delivery_capabilities = ap33772s
        .get_all_source_power_capabilities()
        .expect("Failed to get Power Source Delivery Capabilities");
    println!("Capabilities: {power_delivery_capabilities}");
}

However, it would ACK as normal using 1_000Hz.

[CentyLab]

Running on C++ ESP32C3. I am getting 10us for one cycle -> 100kHz

Adjust Wire.setClock(400000); //400kHz

We can confirm Arduino C++ lib work at 400kHz

For SDA voltage. Testing at 400kHz, we are getting low at ~80mV.

[ScottGibb]

Interesting!!! I'll have a deeper look when I get back home. On Holiday, so away from Scope and Logic Analyser. Thank you for the graphs!