Resistors - carbon composition vs carbon film vs metal film?

[prefabpeti]

Hello all, I'm working on my 1989 Pioneer laserdisc player still.

I'm trying to source several resistors but I seem to have opened up a can of worms as to what type to get.

I'm led to believe this age of appliance will have carbon composition resistors and I should try to go like-for-like.

From what I can find, carbon composition can handle spikes in power better compared to newer types - but are susceptible to damage from moisture and over time can go outside of tolerance.

Carbon film and metal film appear to be better in most ways.


Can anyone offer advice on whether I should stick with original type or not?

For reference I've attached a photo of the area of one of the boards in question.

[Gyro]

Carbon composition (and it's ceramic composition replacement) is relegated to high pulse applications these days, as you say. They really have no place in normal analogue circuitry (they also have bigger package sizes.

The resistors that you have circled all appear to be carbon film, 1/8W and 1/4W. There is no point in changing them, unless they have failed. All of the ones shown look to be in perfect condition. You are likely to risk more damage to the PCB in changing them for metal film than any likely benefit.

The best policy is to leave things alone unless they are faulty.

[prefabpeti]

Thank you @Gyro!

Something to bear in mind is this is a circuit that has been shorted. I believe the short has been resolved (as this particular appliance has lots of wiring looms connecting the different boards - the wires are quite fragile and reassembly damages them easily. I've repaired a number of crimped wires already). I'm looking at replacing any resistors near the two transistors that getting overheated.

Perhaps I need to go back to the drawing board - if the transistors are still getting hot, perhaps the short is still there.

[BrokenYugo]

A failed or over stressed resistor usually gives a lot of visual clues, you can overheat them to a point of blackening the body and they'll often still work.

[TimFox]

If you have good reason to suspect an individual resistor (such as the visual clues for overheating mentioned above), a relatively mild overload can result in a measurable change in value, without outright failure.  To test the resistance, however, you need to disconnect one side of the resistor from the circuit, which must be done carefully (also mentioned above).