Carbon Composition -vs- Carbon Film -vs- Metal Film

This is more of a discussion than a How To.

Typical vintage gear use carbon composition resistor for low wattage applications from 1/4 to 2 watts.  These resistors deteriorate over time.  The failure seen is primarily an increase in resistance.  Some resistors increase surprisingly way out and above their tolerance designator.  I have found that carbon composition resistors, that are in tolerance before a restoration, some times go  bad shortly after the unit is back in the arms of the owner or positioned caringly on a shelf.  See the resistor test table in the middle of the AIRLINE page.

At about seven to ten cents in quantities per 100 for 1/2 watt 5% carbon film resistors, I just replace them all.  It is better for me to spend the time to replace resistors than to have a unit fail and have to be trouble-shot.  And with practice, resistor replacement is almost as quick as snipping one lead, testing and resoldering the original resistor.  

For me resistor shot-gun replacement is a "No-Brainer".  Now if you are restoring a radio for your self and have it in your control then by all means, if you care to, measure and save the original resistors. But I have had to open the same  radio three times over several months, to find a bad resistors.  I learned my lesson.  

There has been some discussion on what type of replacement resistors to use.  Should carbon film resistors be used? Should metal film resistors be use or should good old carbon composition resistors be used? 

Continuing below are my thoughts insights and prognostications about resistor.

Carbon film resistors are robust components that will far outlast the original carbon composition resistors. They are covered in a plastic like coating keeping moisture away from the core.  Also carbon has been deposited on a ceramic tube then laser cut to a precise and accurate resistance.  1/2 watt carbon film resistors are commonly 5% tolerance.  Most resistors in vintage tube gear is 1/4 watt and 20%.  10% and 5% are selected for specific resistor locations.  So an inventory full of carbon film resistor values that handle 100% more power and are up to 4 times more accurate (5% tolerance), that will last longer than the original resistors and seven pennies per unit (quantities of 100), what the heck!  You can't go wrong.  Can you?  Oh yes you can.

Frequency of operation.  What type of current do these resistor pass?  DC, AC and at what frequency? 10 to 20khz (audio) and/or 530khz to 18Mhz (Radio Frequencies of a Zenith Transoceanic).  Or higher up to 30mhz for a communications receiver like a National  NC-300  amateur radio (Ham) receiver. Or a Hammarlund HX-50 transmitter capable of up to 29.9 MHz (10m).  

Boy! Did I find out the hard way on the Hammarlund transmitter.   Fortunately the though occurred to me and I swapped out the carbon film resistors for brand new carbon composition.  But only in the circuits that handle high frequency RF above 18mhz or in this cast the transmitters first mixer frequency.  I put the stake at 18mhz since I have never had a problem with any radio that receives below 18mhz like the venerable Transoceanic. 

Now you may be surprised how little information is available from the manufacture regarding the inductive properties of resistor.  Induction is what causes problems in resistors at high frequencies.  Capacitance is a factor also but not covered here.  A forum contributor send me the below table showing the frequency characteristics of Metal Film resistors. 

Metal Film resistors are noted for their non inductive properties and some manufactures recommend them for carbon composition replacement.   The best that I can find is Metal Film resistors perform better than Carbon Film resistors in high frequency circuits.  So keep this in mind. 

  Click on this table to enlarge.

In this table I have plotted the first stage and oscillator resistors of an NC-300.    [Schematic found here]   I anticipate these stages will see RF up to 30 mhz.  

Several Metal Film resistors change value considerably.  This renders the local oscillator and the front end of the NC-300 non functional. These resistors were replaced using new Carbon Composition Resistors. 

Not all the resistors are affected by the frequencies encountered in the  NC-300.  Any resistor under 10 K ohms would be OK.  However, I did not have this information available at the time of the restoration.  I think that I would have traded the research time on each resistor for the labor time just to use a carbon composition resistors in front of the first IF conversion and not worry about it.

Remember the main choice of resistors is Carbon Film resistors that make no claim of being a good replacement for Carbon Composition resistors.

It occurs that a Transoceanic has a few high resistance resistors operating around 18 mhz.  According the the table above those resistors should be way out of tolerance.

For instance the Zenith Transoceanic L600 chassis 6L40 (click link for schematic) has all before the first IF (455khz): 

As it turns out most of these resistors control AGC which is a DC voltage.  I'm not sure about R7 on the converter tube from the filament to the grid.

OK now a Philco 40-180 (click link for schematic).  

According to the table above, that 47k ohm resistor (part #15) drops to 2000 ohms at 18 MHz.  Perhaps I got lucky with part 15 on the Philco 40-180.  That resistor could potentially shunt the RF signal to ground if it has a too low resistance to high frequency RF that the receiver covers (SW up to 18mhz).  

Could this carbon film or metal film resistor be rolling off the sensitivity as the radio tunes higher into the SW band? This is definitely something that I am going to look into. 

I still have questions to be answered.  Until then I will use carbon comp replacements on high frequency circuits.

Check out this web site.  A good study on aging resistors. 

What Should I buy?   

I searched for hours to find documented frequency characteristics of carbon film resistors.  I can find only generalizations and nonspecific comments.  

 including your thoughts and comments. 

Don't overlook the LINKS page.  I have collected many resources over the years.

71,694 unique web site visitors (7,734,904 hits) from October 2004 through December 2009.

Copyright © 2010. All rights reserved.

  1st snow 2009   New Years 2010 Visitor.  2010 blizzard.