Resistor Color Code

[Vanadium 50]

In the Dark Ages, resistor values were indicated by colored bans, and we all had to memorize the code. (This was back when they were bigger than a speck of dust)

My question is - why colored bands? Caps have numbers printed on them (as do some resistors). Why fool with bands if you can just print the value?

 

[phinds]

In the Dark Ages, resistor values were indicated by colored bans, and we all had to memorize the code. (This was back when they were bigger than a speck of dust)

Ah, yes. Bad boys ravage our young girls but violet gives willingly, except we didn't say ravage.

My question is - why colored bands? Caps have numbers printed on them (as do some resistors). Why fool with bands if you can just print the value?

Well, the small ones (quarter Watt) WERE pretty small even back then. Maybe printer technology wasn't up to it at the time, given that it was a small cylindrical shape. Bands were easy to apply, I assume. Still, good question.

 

[phinds]

The source that is not to be used says

The use of color bands on resistors dates back to the mid-20th century and was primarily for practical reasons:

1. Space Constraints: Early resistors were often small, and there wasn't enough surface area to print values clearly. Color bands are compact and convey a lot of information without requiring much space.
2. Durability: Printed values can wear off over time due to heat or environmental factors. Color bands, being part of the coating or the material itself, tend to be more durable and resistant to fading.
3. Standardization: The color code system became a standardized method across the industry, making it easier for engineers and technicians to quickly identify resistor values without needing to read fine print.
4. Simplicity in Manufacturing: Applying colored bands during production is often simpler and more cost-effective than printing precise values, especially for mass production.
5. Visual Recognition: Experienced technicians can quickly gauge resistor values using color codes, making troubleshooting and assembly faster.

While printed values are becoming more common in modern electronics, especially in larger components, the color band system remains a classic and effective method for indicating resistor values.

 

[Vanadium 50]

All of those are good reasons to have a cap color code as well, but that seems not to be how things evolved. So what makes resistors special?

I'm sure you all have all seen resistors that are mostly cylindrical but flat on one side. So printing (likely silk screen) was a solved problem.

 

[berkeman]

I'm sure you all have all seen resistors that are mostly cylindrical but flat on one side. So printing (likely silk screen) was a solved problem.

Picture? I don't remember this description...

 

[Vanadium 50]

I don't have any around. They are cylindrical with one flat side, which is printed. They seen to come mostly in pastels.

 

[dlgoff]

I don't remember this description

I don't remember that either.

 

[berkeman]

Hmm, my Google Images search came up empty, so maybe they were very unique parts? Where do you remember seeing them? Inside an accelerator someplace?

Well, the small ones (quarter Watt) WERE pretty small even back then. Maybe printer technology wasn't up to it at the time, given that it was a small cylindrical shape.

I think this was probably the reason, with the small diameter cylindrical shape being hard to print on at the time (especially with the coating material of the resistors). Capacitors were more disk-shaped, so easier to print on their mostly flat surfaces. I don't recall any caps with the same small radius shape as small-wattage resistors at that time.

32kHz watch crystals were probably pretty common at the same time, and definitely were printed on. Was it because of the metal can material versus the ceramic-like material coating the resistors?

 

[Averagesupernova]

Capacitors have had color codes in the past.
-
https://www.tpub.com/neets/book2/3g.htm
-
I've seen leaded resistors with the value printed on. These were precision parts. Less than 1% tolerance if I recall correctly.

 

[DaveE]

Back in the day when our PTH 1/4W resistors were hand stuffed on PCBs, the engineers had to ask our manufacturing people (yes we did our own PCB assembly back then!) to please make the printed values point up where we could read them. This was because they made a lot of mistakes that we had to correct. Part of their fix was to go to back to the color code for a while.

Now that this is automated, many tiny components have no labels at all. You don't need them if you have good manufacturing processes.

 

[Rive]

It did not start with color code. The oldest (I met with) were mostly printed.

By size, that time there were some smaller printed ones than banded ones. So I think size constraints does not apply as reason.

My guess: they lost on competition. The banded ones are with centered leads, the printed ones were often oriented.
In manufacturing, that means banded could be rolled around fast and easy => cheaper.
In supply, centered leads could be packed more compact.
In assembly, centered leads (bended on the fly during assembly) requires less attention and cheaper machinery.

The centered&printed ones still required attention on orientation to be useful.

As for why the caps stayed with printing... I think that might have been about the variation. More details needed to be visible? Voltage and type did the trick? So it was about the caps being special, not the resistors? Honestly, no idea.

 

[Vanadium 50]

so maybe they were very unique parts? Where do you remember seeing them? Inside an accelerator someplace?

Mostly detector readout. Nothing I had anything to do with designing, but I looked at a lot of failed boards. (Before replacing one, it is a good idea to look for signs of overvoltage - if the power supply is killing boards, you don't want to blindly replace them.)

They were precision (4 digits) and had a good number packed near each other on the board. ADC maybe? A discriminator?

Capacitors were more disk-shaped,

Not electrolytics. They can be cyllindrical, like typical resistors.

I had a bunch of blocking caps labeled "Japan". That was it. A C-merer showed their values were all over the place. I replaced all of them.

 

[tech99]

On older resistors the value was read as a colour on one end, then the body colour, then a dot, which was the multiplier. At that time, capacitors often varied in size and style according to their value. In the very early days, late 1920s, components were not available in all values, but were sometimes sold for specific tasks, such as a grid leak, reaction condenser and short wave choke.

 

[Vanadium 50]

Thus far, the answer seems to be "cost", although the question of "why not caps too" is still out there. But do we really think a major cost of resistors is printing on them?

 

[Tom.G]

Try debugging a circuit with printed-value resistors versus color-coded ones.

Unless the circuit in on a PC board with machine-inserted components, in the best case you will be able to read the printed values on perhaps 60% of those resistors.

Not conducive to keeping a full head of hair after yanking on it so much!

 

[Vanadium 50]

Try debugging a circuit with printed-value resistors versus color-coded ones.

a) why not caps then? They seem to be installed printed side down >50% of the time.
b) My experience - and I am no EE - is that most of time time, the circuit is not all that sensitive to the exact resistance. That's why 20% resistors are so common. Obviously you can't substitute a 10Ω for a 10MΩ, but they are not so fussy.

c) I have never had 1 board incorrectly stuffed. I have had hundreds (which makes people sad), sometimes spectacularly, but my experience is they are all the same: all right, or all wrong.

 

[DaveC426913]

I only have a rusty recollection of resistor band reading.
I would guess that resistors, being cylindrical can easily get installed with printing pointed the wrong way to read. Not a problem with bands.

Contrarily, tha caps I remember were vertical, so no such problem with printing facing the wrong way.

 

[DaveE]

the circuit is not all that sensitive to the exact resistance.

Clearly a digital guy speaking.

That's why 20% resistors are so common.

The big laser company I worked for standardized on 1% resistors for everything because the manufacturing overhead costs of buying, stocking, and handling multiple types of common resistors far exceeded the cost savings of lower precision resistors. A significant cost was how many different component types (eg. reels on a pick and place machine) needed to be loaded. If any 10K resistor needed to be 1%, it makes sense to have all of them be 1%. Granted, we didn't make millions of products like Logitech.

I have never had 1 board incorrectly stuffed. I have had hundreds (which makes people sad), sometimes spectacularly, but my experience is they are all the same: all right, or all wrong.

Yes!!! This is exactly the point. If you have good manufacturing processes, you rarely need to read the values. If you don't you are in big trouble. The answer often lies in paperwork* forensics and repair, not so much in troubleshooting more than a few PCBAs.

* Without any actual paper, of course. That's sooo 1980's...

 

[Svein]

Ah - colur codes. I remember way bak when TV could be repaired and often had to. A common fault in a certain model was that the cathode resistor had a tendency to burn out. The standard replacement resistor was 1kohm (brown - black- red). One low-level repair guy was red-green color blind, something we discovered when his "repair" did not work at all. The solutioon was - he had inserted a 1Mohm resistor (brown - black - green) because he could not see the difference!
Another case - a research colleague of mine almost tore his hair out when he discovered that the culprit in his complex circuit was a resistor he had read "upside-down". It should have been 220ohm (red-red-brown), but he had soldered in 1.2kohm (brown-red-red).

 

[Averagesupernova]

It should have been 220ohm (red-red-brown), but he had soldered in 1.2kohm (brown-red-red).

I would assume that was a 20% tolerance resistor. Tolerance band comes last. Gold is 5%, silver is 10%, no tolerance band is 20%. I have never dealt with anything that used 20% tolerance.

 

[Tom.G]

I have never dealt with anything that used 20% tolerance.

Some very old vacuum tube radios (1920's - 1940's) had 20% resistors. If I recall correctly, many used only 20%-ers.

Cheers,
Tom

 

[Sagittarius A-Star]

In the Dark Ages, resistor values were indicated by colored bans, and we all had to memorize the code. (This was back when they were bigger than a speck of dust)

My question is - why colored bands? Caps have numbers printed on them (as do some resistors). Why fool with bands if you can just print the value?

Resistors with common small power rating were too small for printing on them.

Source: book from 1976 "Einführung in die Elektronik", published by Jean Pütz (out-of-stock).

 

[Sagittarius A-Star]

a) why not caps then?

No color codes on capacitors helps to distinguish capacitors from resistors. The color code contains a number without an SI-unit (Ohm vs. Farad).