Re: FW: reactance of 0805 and smaller SMT resistors

Charles Hill ([email protected])
Wed, 18 Jun 1997 22:40:43 -0600

The important thing to keep in mind here is the environment and what is =
the problem to be solved. In physical PCBs with signal integrity =
problems, there are thousands of technical issues to deal with. But =
which ones are significant? An approximate absolute value of inductance =
can help in assessing the magnitude of the effect in a particular =
application. The next signal integrity concern is how to make the =
design better; only relative effects need be assessed. For signal =
integrity what is important is how much inductance a device has in your =
implementation. On the other hand, if you are interested in how to =
measure components and devices, the level of accuracy is very important =
and all the details need to be considered. And that is difficult.

I agree with Mark Randol that how the device is mounted affects the =
actual value since it is dependent on the configuration of the magnetic =
field. The 0.2nH is the value given by the manufacturer for a standard =
short used in a fixture to measure 0805 packages (HP16193A). The =
fixture has an APC-7 connector to an impedance analyzer which is =
calibrated with known short open and load standards. The fixture must =
also be compensated and the instrument does account for transmission =
line loss. So the coax in this case is 7mm diameter.

Mark Randol wrote:

>BTW, there isn't anything magical about coax for this measurement.
You could use coplanar transmission line, 300 ohm twinlead, or
lamp cord depending on the frequency of measurement, availability
of calibration standards, the desired accuracy and precision, etc, etc. =

Coax is used since the field is confined, therefore can be simulated, =
and also is unaffected by other objects. Excellent calibration =
standards are available for coax. If you know where I can get lab =
quality calibration standards (known and traceable to at least 500MHz) =
for twinlead, please let me know.

Charles Hill