Rich wrote:
>
>
>
>>
>>I am open to correction on this (Eu members should be able to confirm) but I
>>think the standard Phillips product resistors are metal film and very
>>stable.
>
>Higher values of R may be carbon film. Values under 10â?¡ are typically
>metal film.
In Europe, small resistors are available in both carbon film (cheaper,
higher tolerance) and metal film (a little more expensive but also
closer tolerance). Medium power (1-3W) are almost exclusively metal
film, and in higher power ratings it's mostly conventional wirewound.
The carbon film and metal film resistors up to 3W are available in all
values from 1 ohm to 1M. The same applies for other materials used in
"film" resistors, for example cermet or tantalum nitride but often not
specified.
>10â?¡ - 100kâ?¡ are usually metal oxide film.
>
The only "power oxide" resistors commonly available over here seem to be
the 5W/7W square type. Some makers supply these as wirewound up to 150
ohms, and "power oxide" for higher values.
> Unless one performs a 1-hour overload test, "think" = guess.
>
Resistors are more commonly rated for stability at their maximum power,
not above it. FWIW I've cooked metal film resistors up to 10x normal -
glowing bright red - for several minutes, and they've returned to within
a few percent of their original resistance afterwards. Carbon film
resistors increase dramatically under this kind of gross overload
because the carbon literally burns off.
> For some applications, intrinsic L is important. An example would be
>VHF/UHF parasitic suppressor service. For suppressor service, >15nH is
>probably not good engineering practice. Measuring intrinsic L is not
>easy.
On the other hand you can very easily get a rough estimate. Scrape the
paint off and you generally find a spiral of a few turns. Count the
turns and measure the diameter and length, and then plug the numbers
into the usual formula for calculating inductance. Obviously
small-diameter resistors with a small number of turns are better,
because the inductance depends on the square of both those values. The
European 1W MF resistors are particularly small.
If the intrinsic L is still too high, you can parallel several resistors
of a higher value (but scrape the paint off those too, because the
spiral may be different). I've used 10 * 1K 1W in parallel for a grid
load at 144MHz with no difficulty, and 4 * 330 ohm 2W for a parasitic
suppressor. Obviously it's also important to use layouts that let you
solder close up to the resistor body with little or no wire leads
showing.
The 20W and 50W power film resistors in TO-220 packages are also good
for RF, but they have to be mounted on a heatsink (unmounted power
rating is very low indeed) and this involves a few pF of capacitance to
ground for each package. Even so, attenuators and dummy loads can be
built for as high as 144MHz with low SWR if you keep the wire leads very
short. The main problem is that the TO-220 resistors are only available
in a limited range of values, so maybe you have to juggle the values and
use a combination of TO-220 resistors and bundles of wire-ended 1W and
2W types.
--
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.com/g3sek
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|