Facebook
Google
GitHub
Linkedin
I am using this fan controller: https://www.digikey.com/en/products/detail/microchip-technology/EMC2101-ACZL-TR/3872079
to control a fan: https://us.rs-online.com/product/eb...YfDjRIDEyvn06QDFQIhoCN-IQAvD_BwE&gclsrc=aw.ds
And I have this I2C mux switching I2C channels: https://www.digikey.com/en/products/detail/texas-instruments/TCA9548ARGER/4497280
There is an i2c mux and one controller is on channel sda/scl 0 and the second controller is on channel sda/scl 1. The firmware engineer is saying that the i2c bus eventually becomes corrupted and he cannot read or write. On the second controller, he can read and write but the fan stays at 100%. When I sent this board to him, the fan + controller was working perfect. Firmware engineer gave me an image that goes from 0% - 100% in 4 steps (25%, 50% etc) with 10 second dwell time on each step and I scoped the pwm and the i2c and it all looks great. I tested for hours.
He got the board and for a second time (I've replaced chips twice now) he is getting the above problems.
Can anyone see a problem with the way I 'designed' the circuit? It's so basic.
I am aware of the control inversion, that isn't a problem. I'm just looking at potential causes that would damage the em2101 or glitch the i2c bus.
when this happens, all other channels on the mux are usable and he can read/write to the sensors on the other channels.
The fan does hit about 80W (datasheet says 67=70w) and the ptc is 3A hold, 6A trip. But even if the 24v cut out (which it hasn't yet) that wouldn't cause the problems we see.
to control a fan: https://us.rs-online.com/product/eb...YfDjRIDEyvn06QDFQIhoCN-IQAvD_BwE&gclsrc=aw.ds
And I have this I2C mux switching I2C channels: https://www.digikey.com/en/products/detail/texas-instruments/TCA9548ARGER/4497280
There is an i2c mux and one controller is on channel sda/scl 0 and the second controller is on channel sda/scl 1. The firmware engineer is saying that the i2c bus eventually becomes corrupted and he cannot read or write. On the second controller, he can read and write but the fan stays at 100%. When I sent this board to him, the fan + controller was working perfect. Firmware engineer gave me an image that goes from 0% - 100% in 4 steps (25%, 50% etc) with 10 second dwell time on each step and I scoped the pwm and the i2c and it all looks great. I tested for hours.
He got the board and for a second time (I've replaced chips twice now) he is getting the above problems.
Can anyone see a problem with the way I 'designed' the circuit? It's so basic.
I am aware of the control inversion, that isn't a problem. I'm just looking at potential causes that would damage the em2101 or glitch the i2c bus.
when this happens, all other channels on the mux are usable and he can read/write to the sensors on the other channels.
The fan does hit about 80W (datasheet says 67=70w) and the ptc is 3A hold, 6A trip. But even if the 24v cut out (which it hasn't yet) that wouldn't cause the problems we see.
