PCA9306-Q1: Output current and Ron

Hey Ryuji,

"(1) Output Current
There is the Vi=0V and Io=64mA at the test conditions of Ron in the EC table.
And there is the Ipass specification in the Recommended operating conditions.

Is this Io same with Ipass?"

-Yes, you are correct the Ipass is the same in both references. The device is recommended to be kept below 64mA of current to operate in the guaranteed datasheet specs. To reiterate: the current flowing through the device from point A to B should be under 64mA to ensure proper operation of the device.

Example1: current flow of 3mA is perfectly fine and within spec.

Example2: current flow of 65mA is outside the recommended and now may not behave as datasheet specifies but will not break.

Example3: current flow of 129mA is outside the max operating and may become damaged.

Note: if you are using this device for I2C then you should not be trying to pull that much current into the device.

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

"Also, there is the description of "The pullup resistor value must limit the current through the pass transistor, when it is in the ON state, to about 15mA." in the 10.2.2.2 Sizing Pullup Resistor.
Is this typo of 64mA?"

I2C spec states the max VoL for a transceiver should be 400mV at 3mA (minimum). I would base my calculations for a pull up resistor on the VoL of the transceiver with the maximum current it takes to reach the maximum VoL. A 64mA is incredibly high and would imply the pull down FET of the transceiver is incredibly strong. 15mA in my opinion is also fairly strong. I would recommend this app note as a way to select a pull up resistor instead of this section of the datasheet.

15mA and 64mA could work but require incredibly strong pull ups. If we say the impedance for a pull down FET to be 20ohms then 64mA would result in a 1.28V VoL and violates I2C spec for max VoL. 15mA in this example would result in about 300mV which meets I2C spec.

Note: This is the voltage on the pull down FET, the resulting VoL seen by the receiving transceiver would see a larger value due to the voltage drop across the PCA9306 as well. In the 15mA example if the impedance across the A/B sides is also 20 ohms then the VoL seen is actually 600mV and could cause the receiver  to see a "high" or 1........ (Some FPGAs require ViL to be 0.5V or less)

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

"(2) Ron
Could you please let me know the Ron at the condition of SDA1(Vref1) is 1.8V and SDA2(VDPU) is 3.3V both pull up resistor is 1.5kohm?
VEN is 3.3V via 200kohm. The schematics is same with Figure 7. "

-The answer to this varies as the Ron is not exactly a constant value and changes when the device is passing a zero from one side to the other. The answer is dependent on the VoL of the pull down transceiver and which side is pulling down as well.

I can order parts and do some lab testing to verify later this week if you need an exact answer but I suspect it to be incredibly low (a max of 30 ohms or so but my assumption would be less than a few ohms for both sides)

-Bobby

Hey Ryuji san,

I was able to get some measurement on Ron last night.... When B side pulls low, the maximum Ron was found to be about 2.75ohms @105C. When A side pulls low the maximum Ron found is about 6 ohms @105C. At room temp (25C), I found them to be 2.25 ohms and 4.5 ohms respectively. These tests used VoL at 0.4V which assumes worst case and will likely be even smaller for the customer.

If you want to specify a maximum Ron, I would probably bump them up 1 or 2 ohms to include deviance and instrument accuracy.

I hope this helps,
-Bobby