Digital Isolator - part to part skew and propagation delay Question

[FAE] Andrew Wang

Genius 10685 points

Other Parts Discussed in Thread: ISO7240M

Hello Team,

We take ISO7240M as an example.

Its propagation delay MIN: 10nS, MAX: 23nS. The delta is 13nS. Part-to-part skew MAX is 3nS. Ch-to-ch skew MAX is 1nS.

Two question, please help.

(1) If there is one clock source connect to the input of two separate ISO7240M CH1, how do we determine the worst case output skew?

(2) What is part-to-part skew definition? And does it have any relationship with propagation delay delta (MAX-MIN)? Just wondering why propagation delta seems bigger than part-to-part skew for all digital isolator?

Thanks!

Andrew

over 8 years ago

0 Koteshwar Rao over 8 years ago

TI__Guru 55815 points

Hi Andrew,

Propagation delay of an isolator primarily varies with VCC and ambient temperature. For 5V, 10ns to 23ns of variation in propagation delay is primarily due to the variation in temperature from -40C to 125C. Whereas part to part skew is the difference in propagation delay between different samples under same operating conditions. For example, if sample1 has a propagation delay of 15ns for VCC1 = VCC2 = 5V at TA = 25C then sampleX should have a propagation delay of 15 ± 3ns (max).
Channel to channel skew is also similar parameter with one major difference, instead looking at different samples this parameter refers to different channels of the same sample.

To answer your listed questions,
(1) Skew shouldn't change with input type.
(2) It doesn't have any relationship with propagation delay delta.

Regards,
Koteshwar Rao

0 [FAE] Andrew Wang over 8 years ago in reply to Koteshwar Rao

TI__Genius 10685 points

Thanks Rao! So fast! I got it.

Isolation

Isolation forum

Digital Isolator - part to part skew and propagation delay Question