开云体育

Not hard to imagine that a nichrome wire wound power resistor will have a
lot of series inductance.

Metal film resistors are carbon ink coated ceramic forms. After firing
they are trimmed to value with a barber-pole spiral trim cut either by
laser or diamond cut wheel. The spiral trim gives them more series
inductance. More inductance the greater the spiral trim length to get a
given batch to resistance value so varied RF performance between same value
part.

Chip resistors are made similarly but triming is usually inter-digital
cuts from each side. Some are trimmed by a small sand blaster to one side
of ink. Chip resistors are better for RF performance then leaded metal
film resistors.

Best for RF performance are old carbon composition resistors with their
core carbon brick. Hard to get and expensive these days.

High value resistors will become capacitive at high frequencies, while low
value ones will become inductive.

Both the series inductance and the parallel capacitance of a resistor will
of course have an increased effect as the frequency rises, but depending on
the resistor's value, and its type/construction/size, one or the other will
become problematic first, as you raise the frequency.

So for each type and physical size of resistor, there is a certain range
of values that tend to remain resistive to the highest frequencies, while
those above degrade into capacitors and those below into inductors. The
higher the frequency of operation, the narrower the usable resistance range
becomes.

RF circuits are usually designed so that only resistors inside this sweet
range are used in RF-carrying paths.

With the kind of quarter watt carbon film resistors I used for many years,
and in the range of frequencies I use in my ham activities, typically
values from about 15 to 200 ohm work best. Using carbon film resistors in
the tens of kiloohm range at RF is futile, and so is the use of resistors
around 1? and below.

SMDs are much better at RF than leaded resistors, and the smaller they
are, the better. This extends the range of resistance values that can be
used at a given frequency. If the resistor is so small that you can't see
it, let alone handle it, then it's probably pretty good at UHF! Say, 0201
size.

Thin film SMD resistors are better in this regard (but also less common
and more expensive) than the usual thick film resistors.

Wired metal film and metal oxide resistors are made in the same way as
wired carbon film ones: Thin resistive layer applied to ceramic rod,
spiral-cut to value. Only the resistive material changes. But I suspect
that metal film ones, given the lower specific resistance of their
material, may get more turns cut into them, and thus become slightly more
inductive. At least my measurements with the NanoVNA tend to show slightly
higher inductance for metal film resistors, compared to carbon film or
metal oxide film.

And lastly, series inductance and parallel capacitance are not the only
problems at RF. Change of resistance value happens too, as correctly stated
by Zack at the beginning of this thread.

I really enjoy using the NanoVNA now to test every part that goes into my
circuits. This has removed a lot of surprises and head-scratching. When I
started building RF stuff while still at school, I didn't even suspect
anything about all these real-world effects. As a result, my RF circuits
often had some strange behavior which I didn't expect nor couldn't explain.
The NanoVNA came out about 40 years too late for me... And I think that
many people share this thought!