Carbon resistors characteristics.

[wa1vwl]

Occasionally, I see a post relating how someone found a carbon
composition resistor that was way out of tolerance. This is not
surprising. Years ago, I learned a good lesson about resistors. We
needed some 330 Ohm resistors but only had 300 Ohm in stock. In a vain
attempt to prove how clever I was, I simply heat the resistors over a
large soldering iron until they measured the proper value. I had a VTVM
connected to the leads as I did this. All was fine until the equipment
was sent to a temperature cycling chamber for 6 cycles of 0 to 70
degrees C. When the test was completed, all resistors had reverted back
to about 2 or 3 Ohms above what they were when I started. Fortunately,
my boss took it pretty well but I learned a good lesson. I cracked open
the books. What I found out is briefly listed below. Much of it is
gleaned form MIL specs.

The resistor manufacturers did not have good control of the carbon
mixtures during the manufacturing process. Temperature coefficient
varied from batch-to-batch and was non-linear. This was even worse for
resistors of less than 1,000 Ohms.

The coating on the resistors was intended as an "adequate" moisture
barrier. In other words, they're not hermetically sealed.

Resistors require an "adequate" heat sink. What you say? Yes, the leads
of the resistors ARE the heat sinks. A good rule of thumb is to allow at
least 1/2 inch lead length on each end of the resistor. Most of the heat
is carried away by the leads. Ambient temperature lowers a resistor's
power rating as well as having it mounted in close proximity to another
heat generator. In other words, resistor wattage must be derated when
operating in a high ambient temperature environment.

Exposure to humidity under non-operating conditions such as storage may
increase resistance as much as 15%. As some have discovered, this can go
even higher after 30 some-odd-years of exposure in a humid environment
or climate.

Applying voltage may change resistance another 2%. This is called
voltage coefficient.

Recommended maximum voltage applied across a resistor to prevent
breakdown varies with the power rating of the resistor. For those
enquiring minds who want to know, here it is:
1/8 Watt = 200 Volts, 1/4 Watt = 400 Volts, 1/2 Watt = 700 Volts, 1 watt
& 2 Watt = 1,000 Volts. None of these are hard and fast but are pretty
good guidelines.

I've read somewhere that you can restore old stock out of tolerance
carbon composition resistors by baking them in a dry oven for about 24
hours at 100 degrees C but I've never tried that. Resistors are so
inexpensive that it's much easier to replace them with metal film
resistors and be done with it once and for all. The only draw back is
they don't look authentic if you're attempting restoration. On the other
hand, it's under the chassis and it won't show. For my part, I'll take
functionality over good looks.

I'm not a resistor expert and I'm not a Physicist. I make my living as a
Technician. So if I've made some errors, kindly correct me.

73/Gerry/W1ID

--- --- --- --- --- --- --- --- --- --- --- --- --- --
To subscribe: listserv at listserv.tempe.gov
and in body: subscribe HEATH yourfirstname yourlastname
To unsubscribe:  listserv at listserv.tempe.gov
and in body: signoff HEATH
Archives for HEATH: http://www.tempe.gov/archives
--- --- --- --- --- --- --- --- --- --- --- --- --- --