LAUNCHXL-F28379D: C2000 I2C pull-up resistor

Yang Lei

Part Number: LAUNCHXL-F28379D
Other Parts Discussed in Thread: PCA9554A

Hello,

I'm using LAUNCHXL_F28379D to control an IO expansion chip PCA9554A. I use GPIO104 and GPIO105 as SDA and SCL with 4.7kOhm pull-up resistors. However, I observed the following waveform on SDA and SCL.

What I didn't expect is the rising edge of these signals. I thought the 4.7kOhm pull-up resistor should generate a quite sharp rising edge but looks like it is actually a slow rising edge. I checked my circuit and there is no capacitor on these two signals. Then, I removed PCA9554A and the waveform didn't change. So, my question is why the rising edge is slow? Thanks!

Best regards,

Yang

over 4 years ago

+1 Kevin Allen18 over 4 years ago

TI__Mastermind 45606 points

Hi Yang,

Your waveforms don't look too bad, some comments below:

It looks like your I2C clock frequency is around 312.5 KHz (Fast mode). A lower valued pull-up resistor thank 4.7K homs may be recommended for those speeds, see this App Note for details https://www.ti.com/lit/slva689. Also, you can check if the F2837x internal device pull-up is enabled and test with/without it.

You may be experiencing a glitch on your I2C lines, similar to what's discussed in the E2E below:

https://e2e.ti.com/support/microcontrollers/c2000-microcontrollers-group/c2000/f/c2000-microcontrollers-forum/676372/tms320f28069-glitch-on-rising-edge-of-i2c-or-gpio

The issue is further discussed in the F2837x device errata, advisory 'I2C: SDA and SCL Open-Drain Output Buffer Issue'. The workaround is to add a small amount of series resistance near the SCL/SDA pins. This may change the look of the signal somewhat.

Best,

Kevin

C2000™︎ microcontrollers

C2000 microcontrollers forum

LAUNCHXL-F28379D: C2000 I2C pull-up resistor