carbon film vs metal film resistors

[gardner]

I have a a couple of cheap packs of resistors in E6 and E12 series.  The construction is not specified, but I have been assuming that the blue ones are metal film and the tan ones are carbon film.  Is the body colour actually a reliable indicator of the construction or do I have to do some tempco measurements to know for sure?

I've been shopping around for kits/sets and I find a lot of E6 and one or two E12 sets going from 1R to 1M or so.  I find I do not get a lot of use out of the low and high value end of the range and wish there were a lot more choices in the middle of the range.   Is there a well known source for E24 or E48 packs covering, say, 1K..200K?

[bd139]

That's usually the case, although there are some forgeries out there from the Chinese side of things where the metal film ones are actually carbon films. It depends on the supplier really. If you bought them from ebay, really you have no idea what they are. Measure them before you use them and you should be ok though.

You shouldn't need an E24/48 pack really. You can sub nearly any two resistors from E12 to get close to the value you would require anyway. I tend to use E12 and a few selected values which are handy on top: 2k, 20k, 200k, 49.9k, 51 ohms and that's it.

You can lazily use the following tool to work out what two E12 resistors will get close to what you need: http://www.qsl.net/in3otd/parallr.html

Really though, most circuits are happy to 10% or so therefore I wouldn't worry too much.

As for suppliers, take a look at Tayda's line of Royal Ohm resistors. They're really good quality for the money. Avoid the ones not marked as Royal Ohm as they're shitty ones: https://www.taydaelectronics.com/resistors/1-4w-metal-film-resistors/test-group-2.html

[gardner]

Thanks.  I generally use a divider calculator to find suitable choices.  But even the calculator you mentioned does not have a setting for E6.

I use this one for dividers, and it doesn't have E6 suggestions either.  http://www.cl-projects.de/projects/tools/resmatch-en.phtm

The types of circuits I ever work on where precision resistance matters is generally voltage dividers and amplifier feedback and it does not matter much the absolute resistance, only that the ratios are close.  I find that the actual tolerance of the resistors in the kits is good or very good -- 2% or 1% is normal -- but the gymnastics you have to get up to get a divider in the 2% range for arbitrary ratios is sometimes silly.  Kits with more values would help me keep the number of parts lower and the amount of tuning down.

[Kleinstein]

Carbon film resistors usually come in 5 or 10 % tolerance, thus a silver or gold band for the tolerance.

Metal film resistors usually come as 1%, sometimes 2% tolerance - thus a red or brown ring. The cheap ones are thick film resistors opposed to the more stable thin film metal resistors.

[floobydust]

Sacrifice and take one apart, scrape off the outside coating.  That will tell about its construction.

[not1xor1]

I have a a couple of cheap packs of resistors in E6 and E12 series.  The construction is not specified, but I have been assuming that the blue ones are metal film and the tan ones are carbon film.  Is the body colour actually a reliable indicator of the construction or do I have to do some tempco measurements to know for sure?

cheap chinese no-brand resistors, even 1% tolerance ones, drift like a surfer on a heat wave  .

If you have a 4-1/2 digit multimeter you can see that the small current used to measure resistance is enough to quickly change the resistor value.
Most metal film brand resistors usually do not change more than a digit even if you breath on them and even if you use a 4-4/5 multimeter.

Of course if you need resistors just to regulate the current through a led, that is not relevant.

Carbon film resistors usually drift much more than metal film, so you might check that to tell which is which.

[David Hess]

The body color is not a reliable indicator of the resistor type.