[Boatanchors] Carbon Comp Resistors: The Darkness Gathers?

[Donald Chester]

>  From: David Stinson <arc5 at ix.netcom.com>
 
> I've got piles of NOS Carbon Comps that have been
> following me around for eons.  The majority of
> these have started reading Hi-Z by 5% or more.
> Many by much more.  I just threw a half dozen NOS
> 100K away because they read from 107K- to
> 122K-Ohms.  Other values are showing similar
> drift.

I wouldn't recommend throwing them out just because they are off-value by only 5-10%.  In most applications, even 20% or more won't affect circuit performance.  That's why they originally sold resistors at up to 20% tolerance.  Newly manufactured 1-watt and 2-watt carbon comps are practically unobtanium; every one that is thrown out because of a slight drift in value means one less on the face of the planet.

I tend to discard ones that have drifted 30% or more, or more accurately, toss them in a reserved box.  Sometimes, an off-value as measured with my ohmmeter is the closest I can come to a needed value.  I'm  not sure if off-value resistors usually continue to drift at the same rate as good ones, the same rate they  have already drifted, if the drift accelerates with age, or if it  slows down.  Wonder if anyone has made that observation.  If the drift rate is the same, slower or at least no higher than it always was, I can see no reason to not use it for its measured resistance, but I would document it on the schematic for future reference.

There are many applications where I believe carbon comps are the only satisfactory alternative.  One of the problems I find with film  resistors is that they don't tolerate peak overloads very well.  Carbon comps may heat up, but when they cool down, the measured value usually hasn't significantly changed if there is no visible damage.  Resistors also heat up when soldered into the circuit (I usually try to heat-sink the leads just as I would with a transistor or diode).  OTOH, even a brief overload may destroy a film  resistor by burning or evaporating the minuscule metallic or carbon film.  My solution has been to use much higher wattage ratings than nominal, when that is a possibility; for example, use a 5-watter in a circuit that may call for only 1/2 or 1 watt.  Modern day films are often physically no  larger than an original carbon comp 1/10 its wattage rating.  But the smaller size dissipating the same amount of heat means higher temperature and more likelihood of evaporation during a power surge.

Carbon comps or Carborundum (Glo-bar) resistors are the only thing I will use for rf applications such as parasitic suppressors, dummy loads and reflected power meters.

My junkbox contains a number of 5-watt, 10-watt and even some higher wattage carbon comps that are still reasonably close to nominal value.  I guard those like they were made of gold.

Don k4kyv