Why Do Resistors Have A Color Code? | Hackaday

The first thing to learn in electronics is how to recognize the value of a resistor. Through-hole resistors have color codes and are usually where beginners start. But why mark them like this? Just like the red stop sign and yellow line are in the middle of the road, it actually seems to have always been this way.

Before the 1920s, manufacturers marked parts in any old way, just as manufacturers liked to mark them. Then in 1924, 50 radio manufacturers in Chicago formed a trading group. The idea is to share patents among members. The name changed almost immediately from "Associated Radio Manufacturers" to "Radio Manufacturers Association" or RMA. In the past few years, there will be more name changes, until it eventually becomes EIA or Electronic Industry Alliance. EIA actually no longer exists. It exploded into several specific parts, but this is another story.

This is a story about how the ribbon enters every through-hole resistor from every manufacturer in the world.

By the late 1920s, RMA developed standards, one of which was the RMA standard for color coding. The problem is that marking small parts is difficult, especially in the 1920s.

The solution is ribbon, but it is not as we know it today. The color standard is the same, but the main body of the resistor is the first frequency band. Then there will be another two or three bands showing the remaining values. In some cases, the third band is actually a point. Therefore, most of the resistors will be the first ribbon. The "tip" of the resistance will be the second frequency band, and the dot will be the multiplier. Radios using this scheme began to appear in 1930. This is the color code table in the 1941 Radio Today Yearbook:

Be careful with the advertisements in the magazine promoting resistors, they are RMA color coded. The code quickly expanded to capacitors (capacitors in modern terms).

Depending on the position of the resistor, this point may be hidden like the printed text on the cylinder. So in the end, everyone turned to the band.

The color should follow the visible spectrum (remember ROY G BIV?). However, RMA omits indigo because it is obvious that many people do not distinguish blue, indigo and purple into three different colors;

, Obviously. There are four slots left, so dark color represents the low end (black and brown) and bright color represents the high end (grey and white).

Of course, if you are color-blind, none of this is funny. Reading a resistor with a meter or bridge from the circuit is undoubtedly an answer. However, reading one in the circuit is another matter.

In 1952, the International Electrotechnical Commission (IEC, another standards organization) defined the E series, which specified the value of resistance input so that the spacing on the logarithmic scale of the resistance was equal. If this sounds confusing, consider an example.

The E12 series is suitable for 10% resistors, and its value provides you with 12 values ​​every ten years. Basic value

This is why you can get a 4.7 K or 47 K resistor instead of a 40 K resistor.

However, please consider the tolerance. A 10% 39K resistor may disconnect 3.9K. If the error pushes the resistance higher, it is 42.9 K, so the 40 K resistor is not needed. In other words, in any case, a 39 K resistor is likely to be a 40 K resistor. On the other hand, the low 47K resistance may be 42.3 K, which is less than the high value of 39 K units.

As you might expect, when the tolerance decreases, the number of values ​​increases. For example, in the case of 2%, you will use E48, which has 48 values ​​per decade (if you guessed E96, the 1% standard uses 96 values, and you are correct). Using E48, the values ​​close to 40 K are 38.3 K and 40.2K. It is 39.06 on the high end and 39.2 on the low end.

Next time you pick up a resistor and read the code from it, you can review the history behind it. The remnants of the ribbon will continue to the surface mount area, not as a color, but as a multiplier representing the three-digit number of the first two numbers and the resistance value. Nowadays, many electronic devices (such as wireless modules and lithium batteries) contain a data matrix (similar to a QR code). To be honest, it surprised me that there is no microdata matrix of some kind on all components (through hole and surface mount) that allows you to point your phone at them and view their complete data sheet. Maybe one day.

Wow, the body color + belt + dot is a real improvement to the current beige plus color, it can be any scheme.

It always seems ridiculous that we label them, so, given that 10% of the male population is frankly color blind (red-green), more people have subtle shifts or simple color discrimination. Isn’t electronics the territory of mankind?

Then, I realized that in fact this kind of work has quickly become female dominance (and as far as I know, this kind of work continues to exist in the sweatshops of the Far East). – I may actually view these statistics. Women are far better than men in this subtle detail-oriented technique, and genetically speaking, they are more likely to suffer from color blindness and have to be more careful.

I will put my hands here. I use a cheap DMM to test the resistance and write it into the resistance bar. I know the color and how the marking should work, but frankly, whether it is black, brown or green (especially the cheap small resistors I ended up buying) is hard to be sure.

I also suffer from color blindness. The biggest trouble is between brown and red. I have to check with an electric meter every time. However, I found that if I take a picture with a digital camera and zoom in, I can distinguish the colors. Maybe my camera (unintentionally) changed the color, or maybe it's because of a larger sample?

I think it is difficult for everyone to find brown and red on old parts. Similarly, white can become yellow, yellow is pale enough, you want to know if it is dull white, and blue and purple can fade so much, you want to know if they are gray.

This one. There is no consistency between manufacturers, and if there is a value issue (making the color code meaningless), I usually find it easier to just use DMM. About time it gets updated. One person’s brown may be another’s orange, and so on. Hopefully, the next step will be to interpret haha's clear transistor labels without the need for an electron microscope.

^ THAT ^ and the color change that occurs when the resistor is overheated!

B ^)

If the resistor overheats, its value will also change. It is best to take the correct value from the schematic (if possible).

Ah, that 0 ohm 0% resistor.

When repairing the power supply, I found that many power supplies have accidentally installed 110 ohm 1% resistors.

You don't need expensive 0R 1% resistors, you can use cheap 0R 20% types :-)

I think that it is more common than color blindness that many cheap and efficient light bulbs produce poor color rendering. Normally, orange and red will look brown. In some cases, it is difficult for me to distinguish purple from blue or black.

Sorry, I forgot to include "Many people don't distinguish between blue, indigo and purple".

The shocking answer to this is that this defect is clearly psychological to a large extent (especially because of the transparency of the lens, people have subtle color cast defects).

But just as the painter or the people you spend all day around the swatches seem to have the superpower of color discrimination, so do the people whose language is clearly distinguished in color by name.

Native Russian speakers are faster than English speakers in distinguishing light blue from dark blue (goluboy or goluboy in Russian).

[reference

with

It is precisely because the words in Russian are so different (tested by people who learn and speak Russian, better than people who have never learned Russian). Wow.

There is no indigo in the Russian rainbow. Instead, it is "light blue, blue, purple".

This seems a bit strange, because I would say that cyan precedes blue, and it is not "light" but green.

Well, cyan ("blue") is somewhere between green and blue ("indigo"). Before entering the computer, I have never heard the word Cyan, here is "turquoise", Americans call it "teal". Therefore, it is no wonder that it is not widely used. Yes, it is green, but I think if you shift the sample further to blue, it will look less green and more light blue. In any case, this is a language problem. Obviously, it is difficult for Eskimos to distinguish certain colors visually because there is no word in their language. It's really not surprising, I bet they have a lot to say about "white people".

The language/visual connection is real. Without a word, you usually don't distinguish colors well. According to Stephen Frye (QI) about QI, the ancient Greeks did not have the word "blue", for them the sky was "bronze". However, whether it is true or false, it is a QI fact of about 50:50. Generally speaking, the same applies to most things that technologist Stephen Fry said.

"Cyan" has been used in photography for more than 50 years and is part of the subtractive color triplet "cyan, yellow, magenta".

In printing, it may also be the most famous. Including consumer printers. However, most people are still not printers or professional photographers. In addition to making their own color film, do photographers need to use these colors in large quantities? Photography uses the RGB color model, which is the output of modern camera sensors. Even internally, they use more innovative colors in the color filters.

A young man named John Savard / Quadibloc owns the most interesting personal website of mental bric-a-brac. He is a very smart person. He has a page about filters used by the camera, not just Bayer, but also his own suggestions. You can use other colors to let more light through, then mathematically find the color of the pixel from each of the three sensors, and perform some addition and subtraction operations. Letting more light in almost always means higher quality and accuracy. Of course, the exposure time can be made shorter and the noise is also less.

Blue-green is blue-green, and blue-green is blue-green.

Only color-blind Americans call cyan blue-green "blue-green", because ordinary people only regard blue-green or "blue" or "green" as a category, so they don't have to worry about being so specific between similar colors.

Teal is not common among artists outside the United States, but any American who can see colors knows cyan and cyan. But they still call it "blue" or "green" most of the time.

Blue-green is only a greener one than cyan. When I set the grayscale on a color TV, I will start by establishing cyan. There is a sweet spot in the middle between blue and green. Once the requirements are met, red will be brought up to create white.

Too much, it looks brown, although some people do like to turn red, if the audience who raises the color control on their scene has the right to vote. I personally prefer "flattering" answers to borrow Hi-Fi terminology.

Why is it cyan? If I guessed it, it seems to me a quirk. But this is something you don't mind watching, because the most difficult part of the TV image, the skin tones of various people and the woods (tables, walls, cabinets) all look like they should be. If I can take advantage of this quirk, my customers will be happy. Yes, I do have a photography background.

However, as I grow older, I am now facing presbyopia. This will affect my ability to look at work. If I want to see the tiny details, I prefer the +3.25 card reader. What a pity, because I was able to solder SMD ICs with my naked eyes. Not so much anymore. At least night vision and color sensitivity.

Fun fact-In Welsh, the same word is used to denote the color of grass and sky.

The ancient Greeks called the sky "bronze"

Is it just me, or do other people have problems when distinguishing the color code that the resistor body is some strange brown or green color? But also noticed that compared with male colleagues, female technicians and engineers have less trouble distinguishing colors.

With SMT things, this has become meaningless.

I think this part is related to color blindness, which is more common among men. And it is more common in the red and green parts of the spectrum (brown has a lot of red).

From my experience, it seems that women see more colors than men because I have been tested for color blindness and it is normal, but it is difficult for me to distinguish shades of pink. The color is usually called salmon, peach, pulp...

There are many good things for color vision on the second X chromosome. I am color blind and I absolutely despise the old color coded long resistors. the worst. Can't we stop doing this and just print small text on them as we do with all other types of components now?

I assembled a small object, which includes a disposable multimeter and two Y-shaped brackets made of wires. I can put a resistor into it and read its value quickly. Of course, it has no effect on reading the value of a resistor that is already part of the circuit. I mean, one in ten men is color-blind, and even more people have difficulty coloring even without a generalized disease-how did this marking scheme pass? Or at least why does it still pass? This is not the seventies anymore, we can print out detailed details on these things.

They also chose the absolute worst color for the beige background and the most used bands. They are really addicted to brown, green, red, orange, yellow and other junk colors, which are the most laborious colors in deuteron eyes.

They can print beautiful details on things, but can you read it? SMD resistors are flat and rectangular, usually larger than through-hole resistors. Stripes are definitely more suitable for small cylinders. There are also printing costs, which must be very cheap for the profit of resistors. Small exquisite printers may be beyond their management capabilities. The printing on the SMD is shaking enough.

What I see with one eye is a little blue, but the other eye sees more red. Therefore, one eye may be a little red, while the other eye lacks blue. Both eyes have (I think) the correct color.

Maybe you are developing cataracts! Otherwise, as you age, your lenses will turn yellow. There is a famous artist who I have forgotten. As he grew up, he suffered similar pain. Compared with when he was young, experts have studied his paintings, which have a completely different palette from when he was old. Together with experts, they analyzed the colors and accurately matched some age-related color degradation. The painting from the middle shows the progress between states, even along the correct time scale.

Do not!

He just forgot to take off his 3D glasses!

To be sure, there is almost no difference between the normal color vision of men and women.

Some women are tetrachromatic and can provide better color vision, but this is certainly not the norm.

However, in addition to the common red/green, there are other forms of reduced color vision.

Wow! I don't know that some people are four-color, so I did a web search.

Explains the concept that men can only see in primary colors :-)

What exactly is "purple"? ?

Therefore, 10% of men are colorblind (most of them are dichromatic, such as 0.1% of women), and theoretically 15% are tetrachromatic (0% of men).

There are considerable differences imo

But as far as I know, in any of my research from color change to a thorough and better distinction between women, there has never been a visual difference between men and women. Some of it may be the psychology I described above.

"Another study showed that up to 50% of women and 8% of men may have four pigments and the corresponding increase in chromatic aberration compared with trichromatic mirrors."

Yes, I found out that I had this problem a few years ago. The solution is to ensure that my entire workshop is equipped with fluorescent lamps. Incandescent lighting will only make the problem worse.

Due to the toxicity of marking dyes, many modern SMD resistors leave the factory without marking at all. Switching to other types of bulbs will not help, you must be organized.

…true? First of all, I have heard of it. Don't think that ink is highly toxic in the field of electronic components. Organizing is a good idea.

Most SMD components are applied by machines, and they don't have to bother trying to read them. Through holes are still used by hobbyists and prototyping staff because they can be used with breadboards.

No one tells you, shouldn't you eat electronic parts? :-) But even if you do, the two most common white pigments, titanium dioxide or zinc oxide, are non-toxic. I think it is even classified as a food additive. Only lead-based pigments should be avoided.

I think this is just a cost factor for printing on small 0402 or even smaller components. Resistors in 0603 and higher are still marked.

In Australia, I have seen many foreign tourists asking how to tell which snakes are poisonous.

I just told them I didn't know because I don't eat snakes.

That may just be part of your luck. Usually the cause of the problem is fluorescent lighting, because the peak spectrum may have a tilt aligned with the color band, making orange or red look brown, etc.

High-wattage incandescent lamps (not dim, light yellow 40w bulbs, etc.) will have "perfect" color rendering, so they are actually the best choice for distinguishing colors.

A high-quality fluorescent lamp or LED is almost as good, so if your fluorescent lamp has a good uniform spectrum, or has a peak in the right position without causing any interference, you are good, but if you want to Choose a lighting fixture workbench and read the many color codes, it is best to try any lights you plan to use before submitting to them.

What I want to say is that only sunlight (6000K) can have perfect colors. Unfortunately, incandescent bulbs are limited by the melting point of tungsten. Therefore, for some rooms, I prefer to use 6000K LEDs. Although the color rendering of some rooms is really poor, you must choose a good room with a high CRI.

LED mainly adjusts the color temperature by changing the amount of blue light in a very narrow frequency band-therefore, it is a low CRI light with lower blue light. It does not help you see colors, because the iris of the eye mainly responds to blue light, so by having a strong peak, the eye restricts the amount of light entering the retina, causing the other side to lose color contrast. Because of the Purkinje effect (Purkinje effect) at the end of the spectrum (red becomes black).

Ironically, due to its nature, cos incandescent lamps have a full spectrum. Although they may lack the power of blue, their light is usually yellowish. The solution to this problem may be to throw more photons and fluorescent lights are brighter, although yes, cool white lights also have more blue.

The color temperature of the halogen bulb is slightly higher.

When the light level drops, the sensitivity of the eyes to red wavelengths decreases and blue increases. When the natural color temperature starts to shift to red, we react to seeing the color "correctly" under sunset conditions.

I don't think it is necessarily color blind (@zé) or lighting (@Medix). Those cheap resistors with almost lime green are really scary. I think the paint they used for the stripes was a bit too transparent, or maybe it was applied when the body paint was still wet, so it was mixed. Either way, the colors will be mixed and become something that only people with extensive mixing palette experience can read.

Lime green! I have seen light blue, but most resistors are official "salmon". I think the 11th color must be found, not to be confused with the other 10 colors too much, and they have used gold and silver, it must be a challenge! It would have been easier to genetically transform more cone cells into engineers.

Okay, not exactly lemon green, a little darker, but this is the closest shade I want to talk about. I received them when I bought very cheap items from Ebay sellers in China. I also saw them being soldered to the PCB of consumer electronics. They are really hard to read, and I often have to use an electric meter. I hate to do this, because then I don't know if this is the actual expected resistance or out of specification. Will the latter be stable? It is best to only buy new ones. Although you don't want to wait a month or more, it is becoming increasingly difficult to do so now.

When working on art, my color discrimination is above average, but I often look at a resistor obliquely under bright light. It is different from the colors used by other manufacturers and lies between the two colors. Their bad colors made me question my vision.

I hesitated and said: "Well, it's nearly 5% more than this color, so it must be this color!"

I try to remember that before squinting too much, the DMM can reach and is faster.

I'm not a color-blind person, but I agree that the color codes of modern resistors are harder to read than those of the 60s, which have a dark brown body and opaque color stripes.

Cataracts and age-induced retinal dystrophy can also cause loss of color vision. Especially in the first case, as people get older, everyone’s lenses will become a bit foggy, which can cause a drop in contrast and saturated color perception-the brain is used to it, so you don’t notice until it really deteriorates.

I believe Bill will not talk about his memory of resistors for more than 50 years.

You can compare old resistors in old equipment with modern equipment to understand the difference Bill said.

I am dealing with many old-fashioned technologies, and I can confirm his words.

Thank you for confirming that this is true. I don’t like to do this when people are skeptical of random guesses such as those mentioned above, which indicate that serious health problems are the cause (or other human “defects”). Luke's answer was (probably unintentionally) a rather hostile reaction.

This is a general reaction, indirectly indicating that the "observer"/human being is always at fault (due to imperfection) and defending the problematic technology/solution/method.

A mentality that is too common among various technicians.

I don't think a person will be hostile unconsciously. Maybe not sensitive. Although on the other hand, they can also be paranoid, spiny and overly sensitive.

If you are using a resistor 50 years ago, it is not unreasonable to assume that it is 50 years ago. There are many grocery stores here! This may not be the case for you, but guesses based on forum posts are acceptable guesses. You can refute it freely, just like in the past.

Everyone’s eyesight is declining. Especially those who spend a lot of time looking at the screen while the rest is staring at tiny colored stripes. Therefore, this is not an unreasonable assumption.

Again, you can refute. It's ok. Don't feel sad, that guy doesn't know you. This is just a suggestion based on part of the information in the forum posts. For many people who cannot see the colors of modern resistors, this is their age! Belongs to the best of us. Wait until your eyes start to spin, and then you will find gray hair everywhere on your head and body!

Strangely, when I was middle-aged, my eyes lost some flexibility. The optician said it was normal. But in this way, I can now read better without glasses, including computers. I am cured! Unless there are far away places, I still need them. Therefore, when I mess up my phone or read something, I tend to stare at them like a clever guide, and then push them back into my nose when I actually see them. In front of the computer, I just took them off. Put them on the table. They are there!

Making assumptions about health is never a good thing.

My eyesight gradually weakened, and I could barely see even with glasses.

My lung function dropped to 52%. I was diagnosed with COPD and I was looking forward to spending a short life in a wheelchair because my lung function gradually declined so that I could not support my vital organs.

Now, my lung function has reached 98%. A recent eye exam showed that my vision is better than the average vision of my peers, and there are no people wearing glasses.

I found that my condition was the result of toxin exposure, not related to age. I left the environment where the toxin was.

Never think that health begins with age, otherwise you may be left behind and fall into a debilitating state that does not require pain.

It is indeed difficult for me to distinguish certain colors, but not always.

Stupid aging eyes;}

I did not consider how the color of the resistor body might change, I must pay attention next time.

But even if it doesn't, my red/orange will mix, and my brown/purple will mix.

Just the night before, I had a 5-band resistor, and I swear that the gold band was always yellow, and then cursed my eyes again until I noticed that the 4th band was purple.

The brown-black-black-purple "gold" will be 1 billion ohms, which (useless) makes no sense.

It is strange to realize that I knew the correct color at one time and read silly things backwards. That is 1% 470.

There is indeed a gigaohm resistor. Usually, they are not color-coded, but I am pretty sure I have seen at least one color with normal "stripes".

I have done a lot of work in audio equipment electronics and condenser microphone preamplifiers. Very high resistance resistors are common. Sometimes as much as ten megohms.

The resistance of the grease left on the finger may be less than 10 megohms.

Next step-what is the range of 0 ohm 5% resistance?

I do have some zero ohm resistors. Of course, a single black belt lacks tolerance information.

But your comments did make me want to draw my own silver or gold bands at the end and somehow try to get them back into the wild.

Can you imagine the look on that person's face ten years from now? >:}

Once upon a time, I ordered a roll of 1206 zero ohm resistors, and Digikey sent me a roll of 1206 fuses. I realized that I can't say that they gave me the wrong part, it's just not the part I ordered.

According to my experience, the range is about 5 meters.

The joke is that when you look at the data sheet for zero ohm resistors, many times they do specify tolerances.

The product code of this resistor is RC0603FR-070RL, where RC0603 is the size, F is ±1%, R is the reel type, 07 is the reel size, 0R is zero ohms, and the last L represents a custom label. When ordering these parts, if needed, you can technically order a resistance with a zero ohm resistance of 5%. Anyone will guess what you will get.

Ima buys cheap Volt or Leaf batteries. The battery has been burned out and equipped with a zero-ohm resistance container with a tolerance of 5%. It is tested to find out the resistance of less than 5% ohm, and then the car Drive away and sell the rest of the back.

*Starting from 0 ohms, derp, should be obvious from the context, but for nitpickers...

If you bought it from eBay, then 5% of the time will be a 0.4 ohm resistor.

@RW: 5% of zero is still zero.

Also, please don't forget that some of them have very high TK, up to 4380ppm/K.

But for most "zero-ohm jumpers", they do not specify a tolerance, but a maximum value. Value, for example 50 milliohms.

The "maximum" is also strange. A manufacturer provides a 200V 0R jumper in the case of the "working voltage" specification, which will be huge power.

Oh, I know they exist, but I have never bought or used anything myself, so it doesn't make sense to put them in my parts box.

Of course, the parts bin may resemble a garbage drawer. I found someone there somehow only surprised me a little.

1 Gohm, what glass are they made of?

Never understand why smd capacitors have no code for resistance

This is because the assembling robot does not query the physical map like we do. Compared with girls who do more complex jobs, they are paid higher.

I guess this is because SMD parts are directional, there is an "up" part that can identify troublesome parts from any angle, and these parts have no codes, and the space on the parts is not too small.

There is no "left" or "right" for through-hole resistors. Or think about it "upward". You can determine the end that starts with gold/silver/no streaks at one end. The same thing can work like a cylinder on a rectangle.

Don't fucking tempt them! Their actual numbers are much better than those bad, useless colors! They should modify these numbers in turn and place them on the resistors.

useless. Until you use low TC parts printed with that value, but the machine that loads them puts all the values ​​on the PCB. The schematic is not helpful, because the code is under the part. Read the documentation carefully until you find the board layout, then cross-reference the schematic and know the value after 5 minutes, and you can continue troubleshooting-if you remember why the value is needed.

I also don't think SMD resistors should be installed in reverse. The resistive element is usually on the top, so if it is mounted on a board, the maximum dissipation may be worse.

do not know! I always thought they were a mixture of various solids such as carbon and binder. interesting!

No, they are usually mostly solid mixtures of aluminum and oxygen. Usually called Al2O3 ceramics. With a thin layer of metallic glass resistive glue.

In the Soviet Union, the denominator of resistance is represented by numbers

The link is broken or forbidden:-(

Just press ENTER in the URL line.

Unfortunately, it is not used today.

!

When I was very young, I learned how to read basic resistor color codes and never looked back. Of course, when I need a strange value in the parts kit, I still need to find something, but this is a value that has never been used before. I do admit that I first noticed this strangeness in a pamphlet about standards published by Radio Shack a few years ago.

Interesting article-thanks! But I am still confused about the concept of E series and tolerance. I have heard two explanations:

1. The most common situation is that 10% of the resistance is random, and the error may be as high as 10%. For example, due to the manufacturing process I think, a 100 ohm resistance may be between 90 and 110 ohms. However, I have never been able to find relevant basic statistics. Does this mean that the resistor has an average value of 100, a standard deviation of 10, and a Gaussian distribution? Or, the distribution can be uniform between 90 and 110 ohms? If anyone has a good reference, I would appreciate it.

2. The "non-random" interpretation of the E series is that if the design requires resistance R, the nominal value is always within 10% of the required R value, and with a 10% resistor, the designer will never exceed the required sum The error between the nominal resistance values ​​is 10%. Does this mean that resistance randomness is not assumed? (That is, the actual R value is very close to the nominal value, otherwise there are two sources of error, namely the difference between the actual value and the nominal value and the randomness of the actual R value). Maybe a 10% 100 ohm resistor actually has a small standard deviation, maybe one ohm or less?

Any clarification would be great! If you are not an electronics expert, if this falls into the category of "stupid question", I have to apologize in advance!

There is also a third version:

3. The manufacturer measures and produces the same batch of resistors in batches. First, they remove 1% of the resistors, then 2% of the resistors, then 5%, and then 10%, and then discard the remaining resistors or sell them to night brands outside of China Retailer.

This means that the resistance of 10% will almost never be lower than 5% of the actual value, unless the factory issues an order to transfer part of the 5% or better bin to the 10% bin. In other words, unless you buy a 1% resistor or a 0.1% resistor, you will never get close to the actual value.

Good article, still really don't know why the band is needed in the first place. The new resistors illustrate their value, and the resistor is only replaced once about once, and the only time if it has burned out and the strap is not visible. If you want to clone the circuit, remove the part so that the trace can be seen. I think this might be useful for recycling, but I can't imagine that resistor manufacturers would want to help. Are they more useful when the cost of dmm is higher than the battery that comes with it?

I remember that I once found a resistor for laboratory use in a class. They use some 1/8w small resistors that have been in use for more than 20 years. Even with a microscope, it is impossible to distinguish yellow to brown and orange to gold. The lecturer didn't even know what they were, so he only gave 100% to everyone. Very useful stuff!

Well, the working performance of these belts is much better than that of cylindrical parts.

But none of your arguments are about ribbons, they are just against any form of labeling.

I guess that once the circuit is put into use, you no longer have to quote the part label, which gives you even greater motivation! (Completely a pun), but I am glad they are there.

If you consider older chips with nicks, it is more of a color issue. But now imagine this problem, even the poor engraving effect will only appear in the 360-degree rotation of your head!

Well, I often use old gears from Germany and Russia, and there is no problem with resistors playing their value in "plain text". But I must admit that I also know some color codes.

In the right way; this part may not need marking. We live in a world where half of the chips obtained on the board do not have a usable data sheet, and even many discreet components lack any kind of identification. I know the cost of these wire gauges is not high, but I also know that resistors are very cheap, so can I save a lot of money? If nothing else, please consider the school wasted hours on this... Get rid of the mark, these courses become useful things!

In my case, I usually buy the cheapest resistor that best suits my requirements. Sometimes they have blue bodies, and these lines make me unable to read or even new. Don't care at all, it's actually like looking at it from a visual appeal point of view. The minimum price per 100 units is reduced by $0.01, which is the part I want to buy, even if it is not marked.

Or, this may be a somewhat strange idea, and the manufacturing may like...provide each resistance value in several different colors. In this way, if your product may have 3 through-hole resistors, you can buy red, green and blue resistors, it is more difficult to use the wrong resistor, and it is easier to check the resistor used! Industrial machine vision cameras are expensive, and there is a big price difference between a camera that can check the correct skin tone and a camera that can read the ribbon. Unlike we see hundreds of through-hole resistors on modern boards... Generally, if you use them, they are only used for some high current projects.

Have you ever encountered a 30 ohm resistor, it obviously decided to have its own little rebellion, and can withstand greater resistance until it reads 330 ohm? This is the reason why the VCR does not work properly because it is in the circuit at the end of the IR LED of the tape detector.

Yes, but once you take them out of your school bag, they are practically useless! Of course, you will use a multimeter, but the striped ones are independent of any type of measurement required.

In terms of their Gaussian distribution, I think you will get "random" as in "random". unknown. This is the point. Therefore, you do not know the exact value of this one or any of the other hundred. It means that what it says is nothing more. Even if there is some known random distribution in the manufacturing process, how do you guarantee that they are boxed? Random is random, no more information!

I used to have a cheap DMM blowing a resistor.

According to the schematic, it needs about 111.1 ohms.

(I have the schematic because I built it with the Vellemann (?) kit.)

I checked the resistance with a reliable digital multimeter, found an equivalent resistance and soldered it!

This is an interesting idea. But I want to know whether it's not a big deal to prove that by making various series?

If the target value moves only slightly, 10% of one person may be 1% of another.

I guess the change from the scheme you described to the better scheme I made is to reduce the proportion of more resistors produced and sold in various series, making it easier to make low tolerance settings

The distribution of value largely depends on the manufacturer. Back in the bad age of carbon composite resistors (belonging to the Allen-Bradley category), a good carbon composite material manufacturer could almost control the quality of the entire batch of products at 5% or more even if they were labeled at the factory. Good level. 10% is sold as such. If trucks are parked by the lake for a day, they may absorb enough moisture to reduce the value of the moisture below the tolerance limit, so they must be baked to restore specifications.

Carbon film resistors are better and can be rubbed to high precision before coating. The same is true for metal films and cermets and any common technology today. Many modern trimmings are done with lasers instead of abrasion. If the manufacturer does not waste it, there is no excuse for a batch of 10% resistors without resistance within 5%.

There are other questions. Resistance has a temperature coefficient, and its value will drift with age. If a resistor must handle power close to its rated power, it will age faster. If you really need 1% resistors, you don’t want to buy resistors using technology that can produce cheaper 10% resistors and meet the specifications.

These choices are not entirely correct.

Resistance is set by material characteristics and size, both of which need to be controlled during the manufacturing process. The chemical reactions are relatively easy to keep consistent-volume and mass measurements have been well understood for centuries, and some laborious measurements have been used to make a large number of components. The formation of physical resistors is difficult to accomplish consistently, requires measurement and adjustment of each part (very expensive), or expects that the part is allowed to miss the target with a certain allowable tolerance. That will infer the Gaussian distribution.

Over time, the costs associated with this process dropped sharply. Early resistors may have tolerances of 20%, but the automation of manufacturing has allowed tighter tolerances to be achieved at ever-decreasing costs. However, when you want to manufacture 100,000 pcbs and 100 resistors, the price difference of each resistor is 1 cent, which is $100,000, so in most cases, parts with looser tolerances are a better choice.

When selecting this series, the given tolerance should be considered. It is almost certain that the measured resistance is closer to its specified resistance than the next value in the range (next direction). that's it. If you choose a 47K 10% resistor, it may be 42.3K to 51.7K. The next value (39K) in E12 can be 35.1K to 42.9K, and the next value (56K) can be 51.4K to 61.6K. Even at the 10% tolerance limit, if I want 47K resistance, I will most likely choose 47K. Now consider whether the tolerance is 20%, but still choose from E12. The range of 47K is from 37.6K to 56.6K. The overlap of 39K (31.2K to 46.8K) and 56K (44.8K to 67.2K) is very important-if I want a 47K resistor, any of the three values ​​can be provided. The measured resistance of the resistor marked 39K may be higher than the resistance marked 47K, and the measured resistance of the resistor marked 56K is less than the resistance marked 47K. In other words, it doesn't make sense to use the E12 range to select a resistance with a 20% tolerance, because it will not give you more certain values ​​than selecting from the E6 range. Similarly, for a tolerance of 10%, it is almost meaningless to choose from the E24 range, because it gives you less certainty than the E12 range. and many more…

Even if the process control is improved and the variance is reduced, the tolerance value is still only a guarantee. Although 47K 10% resistors manufactured today are more likely to measure 47K than resistors made 50 years ago, the outlier may still be 42.3K. In mass production, guarantee matters are very important. For most circuit designs, resistance values ​​outside the guaranteed range are disputed. Unless absolutely necessary, good circuit design does not require components to meet strict requirements, such as high-end analog electronic equipment (such as professional audio) or test and measurement equipment.

There are some techniques that can rely on the above techniques. Suppose I have a small part to sell in three versions, one is a normal specification, a value specification and a super high-end specification, and the design relies on the critical 50K resistor to some extent. I can simultaneously use 10% of the components in the E12 series to achieve all 3 specifications. First, I designed the circuit to distribute 50K resistors to multiple components. Normal specifications are connected in series to get 47K and 330 ohms. I relied on modern strict variance to create the specified 50.3K resistance. In the quality assurance process, the resistance and the product performance produced by it have passed the specification verification. The budget version has looser product specifications, which can be met by 99.9% of the components, and solves exceptionally serious abnormalities through a warranty plan. The super-spec model replaces a 300 Ohm resistor with a 470 Ohm trimmer potentiometer, which can be factory and after-sales calibration.

I'm pretty sure that your plan to use multiple components will not help, and will actually make the situation worse. If you are not lucky enough to buy a small part with all resistances at the low end of its range, the final accuracy will be even lower. On average, I don’t think there will be any difference. But in general, this will cause additional problems that cannot be solved.

In addition, you mean 3.3K, but we all know it.

Yes, I mean 3k3. Fart at the end of a long post.

Compared with one resistor, the purchase and installation cost of two resistors will not bring any additional problems. The distribution of two additive Gaussian distributions is itself Gaussian. Add the mean and variance.

However, there is no 50k resistor in E12. Not in E24 either. Even E48. Therefore, when choosing a "50K" resistor, would you choose a 47K average distribution or a 50.3k average distribution doubled the variance?

This is easy to visualize in R:

x = seq(40,56, length = 500)

plot(x, dnorm(x, mean = 47, sd = sqrt(2)), type = "l", lwd = 2, col = "blue", main ='normal distribution', xlim = c(40, 60), ylim = c(0,0.5), xlab ='R', ylab ='φμ,σ²(X)')

curve(dnorm(x, mean = 50.3, sd = 2), add = TRUE, type = "l", lwd = 2, col = "red")

The E24 series (usually 5%) allows you to reach 51k. Even there-two 5% resistors with a total resistance of 50.3k are a better choice:

plot(x, dnorm(x, mean = 51, sd = 1), type = "l", lwd = 2, col = "blue", main ='Normal Distribution', xlim = c (40, 60), ylim = c(0,0.5), xlab ='R', ylab ='φμ,σ²(X)')

curve(dnorm(x, mean = 50.3, sd = 1.41), add = TRUE, type = "l", lwd = 2, col = "red")

You must go to E96 to get 49.9k, but this is still not strict enough to meet the over-spec requirements you want to calibrate.

I did a rough Digigikey search on 1/8W through through-hole resistors (because after all this topic is about color codes). It is almost impossible to buy a 10% resistor today. So...Stackpole 1/8W 5% resistor is $0.00589, similar Vishay Dale 1% resistor is $0.0675.

If you build 100,000 small parts, the cost of using a single 1% resistor and two 5% resistors is $5572. If that can meet the market demand for small parts, it will be a very respected design choice.

In fact, my example is artificially designed, and given that there are many better options today, in their correct thinking, no one would choose a 10% tolerance for "some critical" components, let alone critical components . Horowitz and Hill will be shocked. What matters is not just tolerances, but also other attributes such as temperature coefficient and humidity resistance. In my defense, I did not say yes, but only said yes. The value selection in each resistance series can be traced back to the days when you might use a 10% tolerance. Even so, the trick is still valid, but today you will choose two 1% resistors so that the average value is close to the nominal value available only in E192. The comments on eevblog indicate that the difference in resistors is usually much better than expected. From their tolerance specifications.

Similarly, "the given tolerance should be considered when selecting this series. It is almost certain that the measured resistance is closer to its specified resistance than the next value in the range (next direction)."

I think this is for the manufacturer's sake. This means they can bin all resistors, regardless of their resistance value, almost every resistor they make will be within 10% of the resistance value of one or the other. As others have said, resistors within 1% are sold at a price of 1%, but with a tolerance of 10%, almost every resistor will have a place.

Logically speaking, those with a tolerance of 20% will be almost completely between the other values. Assume that everything that falls within 10% is removed and sold with better tolerances.

Chip resistors are sampled in batches and are not individually boxed. Due to the variation of NiCr thickness and the accuracy of the optical and motion control systems aimed at the laser used to trim them, there is still a normal distribution within each batch. In turn, these will depend on the life of the machine. Older machines, newer machines have better initial specifications, and the machines will wear out over time, reducing their performance. My view is that it is usually more advantageous to replace the production line with newer equipment and introduce a new product line with better specifications than trying to pick cherries in batches from an older production line. As a result, the older tolerances eventually become obsolete.

An example-Vishay TNPU chip resistors have a tolerance of +/- 0.02% and a temperature coefficient of 5ppm/C. Older product lines with larger 1% tolerances cannot achieve this goal. It is worth noting that high-precision chip resistors are usually thin-film NiCr, not thick-film. I guess that the film thickness is the control variable and the main reason for the increased tolerance.

Precision wirewound resistors are usually measured individually. For decades, their manufacturing process has not changed much. The process variables include iron core (diameter and cylindricity), welding wire (diameter, cylindricity and tension), and mechanical accuracy of spot welding and winding processes. Most of these are easy to control and can be mass-produced with high precision. However, the spool can only be wound with one resistor at a time. The resulting high production cost reduces the cost of a single measurement.

In all these areas, it is important to remember that good circuit design does not require strict tolerances unless necessary, and even today, this is much less. Few people now use discrete components to build A/D converters, so buying ICs is usually more effective. Wheatstone bridge? I know. Audio preamplifier? I know. The analog system has been replaced by the digital system. Bipolar transistor with MOSFET. Discrete resistors have mostly been downgraded to support roles, such as pull-ups and protections, and strict tolerances are not required in these applications. I would not be shocked if more 5% tolerance chip resistors with 102 and 103 values ​​are used than the sum of any other tolerance values.

I'm not sure about the resistors, but choosing 1% surface mount capacitors for 0.05% tolerance applications, it took a puzzling time. (They will be used at low temperatures and it is impossible to obtain more precise components.) Their distribution is definitely not Gaussian. In a batch, you will get about half closely clustered around some very specific values ​​and a bunch of outliers. The next batch will have a very different center value. We didn't bother to record those records that did not meet the specifications, but my qualitative impression is that a certain peak value is different from the calibration value by a fixed value. This is a huge peak with wide tails in both directions.

Similarly, they first select parts with higher tolerances from the lot.

It is basically a Gaussian distribution, because the process is not completely accurate, it will be biased to one side, and then the bell curve is divided into two according to the nominal value, and then all the parts with the highest accuracy are taken out to different warehouses .

If the cluster you measure is higher than the nominal value, then it should have a tail to the right, if it is lower than the nominal value, then it should have a tail to the left.

What I didn't get was that almost every introductory e-book stated the color code of the resistor at the beginning. But almost no one mentioned how to interpret the capacitor marking! Of course, in most cases, electrolysis is obvious, but not many disc capacitors.

As a child, I grew up before parts were cheap and ordered online, and the lack of any markings on most inductors was a strong limiting factor for my budding electronics hobby. I dream of reusing parts from garbage equipment, but most of them I can’t determine their value!

And it doesn't work. At the time, the L/C table was not an option for children's budget. Now everything is cheap!

Ceramic capacitors use the same system as SMD resistors... but the values ​​given are expressed in picofarads.

104⇒10 0000pF⇒100nF

473⇒47 000pF => 47nF

331⇒33 0 pF⇒330pF

Of course, with the help of the Internet, it is not difficult to find these days. But when resistor color codes are almost everywhere, why is this information not included in more of the initial electronic text?

Because it is difficult to write this part clearly, you can read it from the beginning before purchasing the book.

the same.

For years, I have been searching for information on how to read these old resistors and capacitors. At last!

Can't you find the color code of the resistor? In fact, it is given in every introductory electronic text. There is still no rhyme we are going to mention here... I am really surprised, "How to find a resistance value" or something in the Internet search must be resolved soon! Or, you just need to ask an electronic freak. I think your expectations are higher than expectations!

Stuart, I think this is problematic because the microfarad mentioned in the 1940s guide above. But then I noticed that it was distributed in two lines, not a typo... it said "Pico Farah". Didn't they have ancient Greeks in the 1940s? I think they just don't have an ISO standard!

Do not. This is not what we are talking about here at all! How did you read so much from the comments without knowing the content of the article?

What is not in these books is how to read old, vintage resistors, where the color code is not stripes. Even if you know the colors, when they are the main color, the base color and the dots, in what order do you read them? This is not the usual left-to-right streak.

When was the last time you saw the opening electronic text containing this information? I might doubt you, because you only read very old texts on electronic products, but I might want to know that micro-microelectronics is not old news for you.

I'm thinking, today someone's thinking is a few thousand times slower, right?

"Didn't they have ancient Greeks in the 1940s?"

Before learning electronics, I had never met a person who had studied ancient Greek or consulted ancient Greek.

I even read "Anabasis".

You know Mega, Wei, Wei Wei. I speak that kind of Greek! "Pico Farad", like they did not invent other scale words, even though they have capacitors that need them. Not entirely serious.

Yes, this is the case in terms of reading OLD resistance. Who knows what my brain is doing?

"Before I studied electronics, I had never met a person who had studied ancient Greek or consulted ancient Greek."

Have you seen such a married person?

Some analog meters have a capacitance range. Connect the capacitor in series with the meter and connect the 115 VAC 60 cycle power supply.

It will be your trouble to build your own clear bridge for capacitors, resistors and inductors. Using some known good components as references, you can even have calibrated measurement equipment.

I rarely use capacitors that can withstand 115VAC, and the actual power supply voltage (230V) is much less. Therefore, low-voltage methods are essential.

OMG, thank you Al!

I have a few humidors filled with humidors that look like the "Résistancesanciennes annees 50.jpg" picture. I am not sure if they are resistors or inductors, nor how to read their color codes. All I know is the band!

I think inductors older than that will have visible windings. Wire wound resistors may be like this! I really don't use those museum works in the actual tour. I want to measure their actual resistance with meters if needed. I think if they have drifted over the years, they will not drift anymore, but in fact, you can spend a lot of money to buy a humidor equipped with modern resistors. Maybe you can sell a few old records to someone who restores antique radios at once. Again, I don't want to rely on old components for anything connected to the mains. Even if my antique radio is still usable, I would not plug it in without supervision.

Although they are not useful for some museums, their value or rarity is not enough to be useful for museums. Take them as an example, you can show them as well as modern resistors and SMDs to show how they shrink over the years. It is interesting that we still use the same color code itself. You can keep them and hope they become valuable before your grandson dies. Or after a certain future end of the world, no one can get any more parts. You built a 2-way radio and you are known as a hero in the community. In addition, despite this... I would not use them because they are too rare, but there are as many as they are because they are definitely useless and unreliable.

"Maybe you can sell some old records to someone who restores antique radios at once."

What do you think of the day I bought them?

Everything I've read shows that the resistors in antique radios usually do. It is best to replace the capacitor. So.. If my radio does have a burnt-out resistor, it may be the result of a damaged lid that cannot be closed, then one of these resistors is likely to be a good resistor that was properly replaced on schedule.

If it has drifted a lot, I would not use it. I am afraid to believe it. Now that I know how to read their marks, I can finally test them! If they are available or useless.. I will determine eventually.

Nevertheless, although I admire and like to watch antique electronic products, I still spend more time on modern products. I only really own an antique radio, and I want to restore a day’s family heirloom. Even if repairing antique radios becomes my new main hobby, I might use these resistors for life, with a lifespan of 3 days.

They may be very suitable for circuit sculpture. They are more decorative than any modern decoration. Or maybe it's because a small QRP rig without a casing was made. Here comes the mites of Michigan.

"I will not let it be inserted unsupervised." "

How to install an appropriate size fuse in the device?

Yes, but some parts of the radio may become so hot that the power supplied through the fuse may be dangerous. Some parts should at least get hot. They sometimes use only resistors to lower the supply voltage. At that time they used cloth insulation and wax, and the safety standards are not now. I would really doubt anything old. It's not that you can't enjoy using it, but maybe don't put it on the shelf next to a bottle of methylated alcohol, but stay in the room to listen while plugged in.

I might also want to assemble a short-range AM transmitter and play some music from the 1940s/50s through it. Now listening to the radio trash through virtual antiques will annoy me!

A (possibly) interesting side note: an early radio manufacturer (Philco?) installed new lighting equipment, possibly mercury vapor lamps. The increased light level is welcome, but it makes certain colors indistinguishable.

Their solution is to have the engineer change some resistor values ​​to colors with fewer problems!

Nowadays, the same problem: The CRI of LED lighting is poor, and it is especially difficult to distinguish red: red, orange, gold, and brown are beginning to be similar to each other.

I also found this problem, so I installed a bunch of 95 CRI natural white LED strips (designated as 5000K, measured at 4900K) above the workbench, and now the color reproduction is very good.

There is only a 40-watt halogen lamp on my workbench. CRI = 100

40W halogen lamps do not have that much light output. I prefer 40-50W fluorescent lamps or LED lamps.

Although a few years ago, I still needed a short-term solution to overnight SMD assembly. The 300W or 500W halogen lamp achieves this purpose very well, except for its high heat output-it has been in early summer, dripping sweat will not act as a flux :-)

For spotlights, enough.

I think it depends on the phosphor. A single yellow phosphor, together with the blue LED on which they are based, will emit a "white" light, but I don't want to use it to light up my house. In particular, it is not necessary to identify the color under it. Again, although I will use a multimeter as much as possible.

If they add extra phosphors, they can get better light. The name from the appropriate manufacturer comes with an English or European name, at least a name with Latin letters (!) designated for lighting, which may be better for CRI. Put in some red, green and other phosphors to get the widest possible frequency band. The LED is monochromatic. Like every Ebay supplier that sells goods, cheap, nameless Ebay LEDs are purely price-based, and you can search for prices. Price-quality = profit, so quality is the enemy!

Many people have noticed that this may be a problem, but don't know why. The public's ignorance and boredom leads to one of the things that are shy. It's like politics!

The "better" CRI> 80 you can find in any supermarket without special order. They basically just add red phosphor to the yellow phosphor. It is difficult to find bulbs with CRI>90 anywhere, because 80 bulbs are a cheap manufacturing point for these products, and most consumers will not notice or care.

But this is still bad. Even if the CRI of the old compact fluorescent tube is> 87, and to have good color rendering, you want to be greater than 92.

When I discovered that our assemblers used 270K (red, violet, yellow) and 4K7 (yellow, violet, red) resistors interchangeably, I had to do something similar. The 270K resistor has an arbitrary RC time constant, so we switch to the adjacent value (220K or 330K). Close enough!

An older colleague of mine believes that E3 values ​​of 1, 2.2 and 4.7 are sufficient for most components :-)

It does feel that most of the old resistors on the earth are beginning to appear brown-black, red-red or yellow-purple. If I had to guess, the next most common value will start in blue-gray.

Using only those E3 values ​​in the gain circuit R/r, you can approximate the gains that are important in log10 mathematics-1, 2, and 5. Using the E3 series of only 30 years, that is, 9 total values, you can get about 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 50, 100, 200, 500.

The combination of two E3s in series (or an E3 with a 0 ohm jumper) can generate: 1, 2.2, 3.2, 4.4, 4.7, 5.4, 5.7, 6.9 and 9.4, which is a good starting point in the linear range. Adding the value of 6.8 can also generate 7.8 and 9.0.

Please correct your article, this is "Résistancesanciennes annees 50 by François Collard, CC-BY-SA 4.0".

When my grandfather passed away, I learned that he was in "resistance" and passed the British pilots from one place to another, through the forest.

Resistance is not in vain. Oh, sweet!

I like the simple three-letter system used for SMD parts, but the "new" EIA-96 is indeed PITA. I will never remember these things, always have to look for it. Measuring SMDs online is not always feasible, and sticking them in a confined space and removing them without damage will not be fun.

In any case, all this will soon disappear. Resistor manufacturing has started to stop printing values ​​on SMT resistors.

Mainly a few

#1) They can save $0.0000000002 and save ink cost by skipping the printing process.

#2) There is no value on the capacitor, and no one complains, so why do we do this to the resistor.

#3) The machine mainly installs them and does not need to read the value.

In the next few years, you will pay high prices for marked resistors.

#4. In any case, 01005 size SMD cannot be read.

First of all, I think they cannot be handled.

Among the many resistor manufacturers I deal with every day, none of them plan to delete the mark (due to the work of electronic contract manufacturers, I think there are at least 9)

1) The cost has nothing to do with the added value of the mark used for inspection and other post-processing. The resistor material is ideal for adding a marking step, so this is the reason for the standard process for resistors.

2) Due to material reasons, the cap is not marked. Labeling capacitors with standard methods will not lead to very obvious results. Manufacturers can use laser-based processes to obtain caps, but this adds considerable cost and is therefore not standard.

3) The perfection of the machine is only human-loaded machines and traceability tools used by manufacturers. After SMT, flying probe or ICT style tests are not always used to verify the assembly, and when a 10k resistor is loaded instead of a 20k resistor, the functional test cannot always capture, but it will affect the end user. Using AOI, AXI and visual inspection are more cost-effective solutions for quality management.

4) Yes, for 01005 and most 0201s, legibility is limited.

Can a typical AOI recognize the printed content on the resistor? Or is it not possible because ICT is more reliable?

The mnemonic-it is best to be right, otherwise your grand plan will go wrong-(black, brown, red, etc.-purple stands for purple) has helped many learners. Of course, there are also hard-shell variants, but this "clean" version is suitable for hybrid companies!

Bad beer rotten our young courage, but vodka is going well

Hey, Elliott!

What is the German who put the rose on the grave?

Oh, I do not know. I have to ask around.

Dutch: Zij Bracht Rozen Op Gerrits Graf Bij Vies Grauw Weer. I think she brought the rose to Gerrits' grave in the dirty gray weather

Al, I am glad you can cheer for your brother!

I just came here to say that I found a file that was stored decades ago...

Zij Zwart (black, 0)

Bracht Bruin (Brown, 1)

Rozen Rood (Red, 2)

Op Oranje (orange, 3)

Gerrits Geel (yel