I am using a level shifter with high impedance input pins. The input to the shifter is coming from a 50 ohm output impedance component through a 50 ohm line. For matching purposes and according to the driver components data sheet, this signal need to be terminated into a 50 ohm load so I am considering just connecting a 50 ohm resistor between the level shifter input pin that I am using and ground. Is this acceptable and is there a better way to go about this? (shifter is TI NS series)
1
-
\$\begingroup\$ If the source is already 50 ohm terminated you probably do not need to terminate and can simply let the signal reflect back into the source where it will be absorbed. \$\endgroup\$– user1850479Commented Jun 27 at 2:56
Add a comment
|
1 Answer
\$\begingroup\$
\$\endgroup\$
5
Will I run into issues if I connect a shunt 50 ohm resistor over a high impedance input pin on an IC?
One result that may or may not be an "issue" is that the voltage at the high impedance input pin of your IC will be half of what it could be if there were not shunt resistor. You haven't provided enough information to determine whether that would or would not be an issue.
Something else to consider is whether or not you actually need to terminate the 50 ohm transmission line. If the transmission line is less than, say 10% of the wavelength of the highest frequency of interest, then terminating the transmission line with a 50 ohm shunt resistor may be unnecessary.
If neither of the above considerations apply, then terminating the 50 ohm transmission line with a 50 ohm shunt resistor is an appropriate choice.
answered Jun 27 at 2:50
-
\$\begingroup\$ In Fundamentals of Applied Electromagnetics, Ulaby & Ravaioli, 8/e they write this on page 64. About a factor of 10 different than you suggest. But many do use the figure you provide, too. :) \$\endgroup\$ Commented Jun 27 at 3:28
-
\$\begingroup\$ Interesting. I have never seen the 1% of \$\lambda\$ used before. \$\endgroup\$ Commented Jun 27 at 3:31
-
1\$\begingroup\$ 1% seems overly cautious, but to be clear, it's an arbitrary threshold relative to the lowest resonant mode (1/4 or 1/2 λ); most accurate would be to say "much less than λ/4", where how much depends on the accuracy one needs in the low-frequency / lumped-equivalent approximation. Mostly, small phase errors can be ignored and the closer threshold suffices, but one can conceive of (or contrive) cases where a much lower threshold might apply. \$\endgroup\$ Commented Jun 27 at 3:49
-
\$\begingroup\$ This page from Altium provides a quick overview. \$\endgroup\$ Commented Jun 27 at 4:04
-
1\$\begingroup\$ Another issue is that in many cases, having either end of a transmission line properly terminated will be sufficient to prevent ringing; some devices that drive transmission lines supply source termination, but others may be relying upon the far end to properly terminate them. The latter pattern seems to have been pretty popular in 1980s video games. \$\endgroup\$– supercatCommented Jun 27 at 16:13
