• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 16 subscribers
  • Views 208 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

SN74AUC2G53: tpd plausibility

Alfred Fuchs1
Intellectual 320 points
Part Number: SN74AUC2G53

Propagation delay at Vcc=2.5V and 15pF load is given as 0.1ns.

Footnote (1) says that this is the calculated RC-time constant with the typical on-state resistance and the load.

For AUC1G66 the corresponding value is 0.1ns and for AUC2G66 it is 0.4ns.

Their footnote says exactly the same.

The typical on-resistance according datasheets is 9 Ohm for the 1G66, 4 Ohm for the 2G66 and 6 Ohm for the 2g53.

So something does not pass my plausibility check ...

Given the higher complexity of the mux can it really be factors faster than the 2G66?

Or is the 2G66 the unexplained outlier? As I understand the footnote it ought to be fastest.

RC=15E-12*4 would be less than 0.1ns actually.

Something is inconsistent, I'd ask for an explanation if there is one.

A second observation:

The disable time for INH decreases from 2.6ns to 2.2ns if the load capacitance is doubled from 15pF to 30pF.

The same phenomenon happens for 1G66 and 2g66, but only at Vcc=2.5V.

Honestly having difficulty to understand why that would be ...

BTW: The device ought to be named 1g53, with "2" we'd expect a dual switch, compare 2G66...!

  • 0
    Guru 243980 points

    An 1:2 mux consists of two switches, so there is no practical difference.

    The tpd calculation assumes an infinitely strong signal source. It might be a good idea to ignore it and to do your own Spice or IBIS simulations.

  • 0 in reply to Clemens Ladisch
    Intellectual 320 points

    I actually did not expect a difference, but between the 2G66 and the 2G53 there is one ...

    I am not hunting for datasheet bugs, but I become irritated when I stumble over anomalies. I have been trained to always dig deeper in such cases and for good reason.

    Application context: I need a fast quadrature demodulator for a BPSK-modulation in a 3-phase data transmission (specifics irrelevant). The inverter ring oscillator question is related to the same design.

    After long thought I happened to find a very nice solution with the AUC2G53, where the (almost) delay-matching between A and INH input (both used!) comes in handy. Except then there is this odd difference that raises questions.

    For the record: Perfect for my use case would be a dual device with two 53-analog muxes at that speed for precise timing matching of signal paths, that's the reason why I balked at the misleading "2G".

    Take that as a product planning feedback: Duals are _always_ wonderful!

  • 0 in reply to Alfred Fuchs1
    Intellectual 320 points

    For the record: The table with the Switching Characteristics points to Figure 5.

    There is a table in figure 5 giving the paramaters of various test circuits, but it does not say clearly which parameters are used for which measurements. But now I see that there are two entries for 2.5V, one for 15pf and 2kOhm and one for 30pF an 500 Ohms.

    Ah, R decreased by a factor of 4 when increasing C by a factor of 2!

    Thanks for the quiz, but finally I passed, it seems.

    Obviously this does not transparently characterize a chip, I must take it to the books as a marketing gag. 

    Still mysterious why this phenomenon does not appear with the A input (only INH) or also at lower voltages, but I assume, we'll not get better data.

  • 0 in reply to Alfred Fuchs1
    Intellectual 320 points

    It does not occur with the A, because in a mux one switch remains enabled. Got it.

    It does not occur at lower voltages, because there the load resistor is not changed by a factor of 4.