Hi,
In a test system we are utilizing the NI 8452 for I2C communication. We are having trouble with one of the test objects getting stuck in the communication. Analyzing further we could see a glitch on the SCL signal, thus the message gets corrupted. The glitch can arise anywhere in the communication, and if unlucky the bus gets stuck low.
We have tried lowering the pull-ups from 4k7 to 1k8. The signals look better ("better" edges), however no change in the occurrences of the glitches. We have also tried lowered the frequency with no change (running at 100kHz as default). The glitches are not always present, and if we are sending the message again, we can sometimes get an OK message through.
It should also be noted that we are measuring a current between the SHIELD of the system and GND on the 8452. 6mA when powering the device, going up to 23mA when initializing the 8452 I2C.
Attaching some pictures of the I2C signals. Is this problem somehow related to clock stretching (the glitch occurs where a clock signal should be, but does not fit, looking at the timing)? Are there some filters we can enable/disable preventing this? Could not find any in the manual.
Thanks in advance!
