ADT7410 problems with high temp

Hi T_Barry,

Can you capture waveforms of the I2C signals and also ensure there is no noise on VDD at +110°C?

Ensure that you have placed 0.1uF and 10uF coupling capacitor close to VDD and that they are rated above 125°C.

Regards,

Rob

Dear Rob,

100nF coupling cap is there. Didn't foresee any 10µF Cap

At the moment its not possible to measure I2C signals, but there are other devices on this bus... and they don't have any problems communicating

Hi T_barry,

have you resolved the issue? Is there any noise on the power supply?

Have you tried multiple IC's?

Regards,

Rob

Dear Rob,

What i figured out at the moment.

It seems the Ack from the temp sensor is far to high 0.7V, probably because of to heavy pull up @ master side.

When the temp sensor is heated >100°C it seems this level is even higher 0.9-1V.

My master is seeing 1.15V as a low, very close to the 1V ack of the slave, so from time to time it doesn't generate the clock pulses needed for the slave to answer as the Ack was not valid.

Im playing around with other pull ups to see if i can get the ack to a lower value.

regards

Tim

Hi T_Barry.

10K resistors are recommended to pull up the I2C lines.

Regards,

Rob

Dear Rob,

10k at slave side was used in the design.
Master is using stronger pull-up of 2k, so we need to fine tune master pull-up.

We changed the master pull-up to 10k, running in loop for the whole day without problem.

So probably this was the root cause, but we will leave it running over the weekend to be sure.