I have a Heathkit signal generator that had a bunch of carbon comp resistors that had gone way out of spec. Now I have a couple amps that have carbon comp resistors. Should I be worried about their performance? These units are 40 years old, so not quite as old as the Heathkit. I have a parts list and schematic, so it wouldn’t be hard to identify and replace them.
Carbon composition resistors have a well deserved reputation for drift (usually upwards) and the higher values are especially prone to this. I don't trust any of them, especially values above 100K.
Even the new ones cannot be trusted; I have regularly encountered modern audio equipment that uses carbon comps that were laughably beyond tolerance within just a few years of service.
When it comes to old coupling caps and carbon composition resistors, measure everything, trust nothing.
Even the new ones cannot be trusted; I have regularly encountered modern audio equipment that uses carbon comps that were laughably beyond tolerance within just a few years of service.
When it comes to old coupling caps and carbon composition resistors, measure everything, trust nothing.
Why do you specifically mention coupling capacitors? Is it because they are more important that other caps, or because they are more susceptible to drift?
Suppose you have an output stage consisting of one or two valves with a transformer to the supply, and suppose their control grids are connected to the previous stage using a coupling capacitor and a grid leak resistor with some high value.
A slight amount of leakage in the coupling capacitor or capacitors then suffices to change the bias current of the output valves to a much too high value, possibly blowing up the output valves or the output transformer.
Antique paper capacitors are notorious for this. Even people who try to keep their equipment as original as possible often replace the output stage coupling capacitors if they are paper capacitors (sometimes they make the new capacitors look like antique paper capacitors so their equipment still looks original). Antique paper capacitors have often absorbed some moisture, which much increases their capacitance and especially their leakage.
Although output stage coupling capacitors are the most critical, if the leakage is really bad, it can also cause problems at other locations in the circuit, where there is a large voltage across the capacitor. For example, the capacitors from the anodes of the output stage to ground that you find in some radios, or capacitors in the power supply that have hundreds of volts AC across them.
A slight amount of leakage in the coupling capacitor or capacitors then suffices to change the bias current of the output valves to a much too high value, possibly blowing up the output valves or the output transformer.
Antique paper capacitors are notorious for this. Even people who try to keep their equipment as original as possible often replace the output stage coupling capacitors if they are paper capacitors (sometimes they make the new capacitors look like antique paper capacitors so their equipment still looks original). Antique paper capacitors have often absorbed some moisture, which much increases their capacitance and especially their leakage.
Although output stage coupling capacitors are the most critical, if the leakage is really bad, it can also cause problems at other locations in the circuit, where there is a large voltage across the capacitor. For example, the capacitors from the anodes of the output stage to ground that you find in some radios, or capacitors in the power supply that have hundreds of volts AC across them.
Last edited:
When replacing composition resistors, especially in older equipment, don't just automatically replace with a carbon film. Composition resistors have a much higher surge rating and this can be a very important factor to consider when replacing them - and older equipment with it's higher voltages and surge currents often require them. And if you do use film resistors as replacements, make sure they have a high enough voltage rating - most film resistors only are rated for a few hundred volts and will quickly fail in the high voltage environment of vacuum tubes.
Hal
Hal
Carbon comp I deem as "replace on site", metal film are excellent.
Among the problems with carbon comp is that the value varies over time, temperature and with moisture levels, the excess noise is high and highly variable between devices, and they are more non-linear than any other type of resistor.
The value variation over time can be massive, factors of 2 or more, so they are basically a time-bomb in some circuits, leading to malfunction eventually. Hence replace on sight...
There are a few uses they can be good at, for instance dumping extremely high power short-duration spikes of energy, but that's of no use to audio really...
[ or if you want a museum piece, keep original parts, but don't expect reliable operation without luck being involved ]
Among the problems with carbon comp is that the value varies over time, temperature and with moisture levels, the excess noise is high and highly variable between devices, and they are more non-linear than any other type of resistor.
The value variation over time can be massive, factors of 2 or more, so they are basically a time-bomb in some circuits, leading to malfunction eventually. Hence replace on sight...
There are a few uses they can be good at, for instance dumping extremely high power short-duration spikes of energy, but that's of no use to audio really...
[ or if you want a museum piece, keep original parts, but don't expect reliable operation without luck being involved ]
@Mark Tillotson Do you replace with carbon comp, or metal film? What do you recommend with regards to voltage handling as people have discussed above?
Depending on make/model/type, they are as well known for leakage as carbon comps are for drift and noise. If you want a reliable piece of equipment that performs (at a minimum) up to it's published specifications, these are the things that you need to check.
The "If it ain't broke, don't fix it" philosophy is great, right up until a notoriously unreliable, low cost, easily replaceable coupling cap ends up costing you a set of power tubes, or worse, an irreplaceable semiconductor or output transformer...Frankly, I'm not that inexperienced, cheap or lazy
One thing that’s confusing me in this thread: Usually I’m used to capacitor discussions gonna alone the lines of “electrolytic vs film”. However here people keep talking about coupling caps, without discussing composition.
I think I get it: coupling caps are especially important. But doesn’t the logic that “film caps generally don’t need to be replaced” still factor in?
I think I get it: coupling caps are especially important. But doesn’t the logic that “film caps generally don’t need to be replaced” still factor in?
"But doesn’t the logic that “film caps generally don’t need to be replaced” still factor in?"
That doesn't necessarily hold true. As I said above, it depends on make/model/type. There are many plastic film and dual dielectric (paper & plastic film combined) capacitors that develop leakage due to the encapsulating material letting moisture in. And many aging electrolytics develop leakage too.
Just as aging electrolytic/plastic/paper filter & bypass capacitors can become problematic, so too can coupling caps. It's not the application, it's the part.
That doesn't necessarily hold true. As I said above, it depends on make/model/type. There are many plastic film and dual dielectric (paper & plastic film combined) capacitors that develop leakage due to the encapsulating material letting moisture in. And many aging electrolytics develop leakage too.
Just as aging electrolytic/plastic/paper filter & bypass capacitors can become problematic, so too can coupling caps. It's not the application, it's the part.