Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Questions about resistors!

Status
Not open for further replies.

woodturner550

New Member
1. If I have a 100 ohm resistor and it has a 5% tolerance. That means that the value can be anywhere between 97.5 Ohms and 102.5 Ohms, OR is it is it anywhere between 95 Ohms and 105 Ohms?

2. If Ihave a resistor and when I measure it with a meter and it really reads 100 Ohms. Regardless of what the color band for tolerance is, can I say that that resistor is within 1% of 100 Ohms and treat it as a 100 Ohm 1% resistor? As well as a 5% resistor and a 10% resistor because of what it really read with the meter. We are assuming that temperature is normal. This is not a trick question.

3. If I want to make a resistor array I would want all the resistors to be as close to the needed Ohms reading as possible regardless of the tolerance band.

Thanks. I hope I asked the question clearly enough.

woodturner550
 
A 100 ohm, 5% resistor is guaranteed to be within 5% of 100 ohms in either direction...95 to 105 ohms.

If you measure a resistor that is really 100 ohms, it is within 1% of a hundred ohms today, if your meter is in calibration. If the stripes do not guarantee 1% tolerance, the resistor will almost certainly drift away from 100 ohms as time and temperature changes.

You must remember the tolerance is a promise by the manufacturer. They don't make 1% resistors and label them as 5% and sell them cheaper because they labeled them as 5%. A lot of the art of electronics is making parts that aren't perfect do a job that seems to be perfect after they are all assembled. Trying to make an array of "very close" resistors is a fools errand because they won't stay that close unless you use top quality parts. In other words, you can't make a 1% array by selecting the best 5% resistors. They won't stay the same for more than a few months.
 
The temperature dependence and aging characteristic is different for different resistor families, carbon vs metal film vs wire wound, etc. Only the resistor maker can tell you on their data sheet.
 
Metal film resistors are better than carbon film but they're generally around twice the price, not that this matters because resistors are very cheap anyway.

You'll also notice that resistors come in seemingly odd values i.e 1.5k, 3.3k, 4.7k and 6.8k, rather than more round values, this is because of the tolerance. The numbers I've just quoted a the E6 series which means there are 6 values per decade and are used for components with a 20% tolerance or better, the E12 series is 10, 12, 15, 18, 22, 27, 33, 39, 47, 56, 68 and 82 and is used for components with a 10% tolerance or better. If you take any value and select the nearest E12 value, the number will always be within 10% of the desired value, obviously you need to then factor the additional tolerance of the component in.

In general when designing circuits, try to use the lowest E number values as you can, i.e. don't use 36k when 33k or 39k will do. This makes it much easier for people building your circuit to find the correct parts.

Wikipedia has much more information on preferred values.
Preferred number - Wikipedia, the free encyclopedia

It's possible to make any E12 value with two E3 value resistors and make and E24 value with two E6 value resistors with a reasonable degree of accuracy. Of course don't forget that there's little point in making an E24 value with 20% tolerance parts.

**broken link removed**
 
Thanks for your replies. They have been very instructiveand helpful.

So other than to make sure the resistor is good(reads within the values marked on it), don't use the meter readings.

musing;
So very interesting. Use a less accurate marking rather than the more accurate reading from the meter. .....I could understand that it would be cheaper to not hold material to the tolerances for 1% for mass production resistors. I learn something all the time....and who says old dogs can't learn new tricks.

woodturner550
 
Metal film resistors are better than carbon film but they're generally around twice the price, not that this matters because resistors are very cheap anyway.

You'll also notice that resistors come in seemingly odd values i.e 1.5k, 3.3k, 4.7k and 6.8k, rather than more round values, this is because of the tolerance. The numbers I've just quoted a the E6 series which means there are 6 values per decade and are used for components with a 20% tolerance or better, the E12 series is 10, 12, 15, 18, 22, 27, 33, 39, 47, 56, 68 and 82 and is used for components with a 10% tolerance or better. If you take any value and select the nearest E12 value, the number will always be within 10% of the desired value, obviously you need to then factor the additional tolerance of the component in.

In general when designing circuits, try to use the lowest E number values as you can, i.e. don't use 36k when 33k or 39k will do. This makes it much easier for people building your circuit to find the correct parts.

Wikipedia has much more information on preferred values.
Preferred number - Wikipedia, the free encyclopedia

It's possible to make any E12 value with two E3 value resistors and make and E24 value with two E6 value resistors with a reasonable degree of accuracy. Of course don't forget that there's little point in making an E24 value with 20% tolerance parts.

**broken link removed**

I know that in the past very high quality audio equipment designers had very narrow specs. for all the parts used. All transistors, resistors etc were checked before being used.

For a reason though. Top end stuff is at the cutting edge of design. Hence critical that components used are are on matching or of better spec. than the minimum requirements of the circuit design.

Krell and AudioLab come to mind with this practice. Very expensive and very,very good at accurate musical rendition. And beautifully built, engineered etc, etc.

However, for the normal circuit designer like myself, it is best practice to design my circuits in such a way that specific resistor and other component tolerances do not significantly hamper the circuit functioning as should. Cheap and easy to mass produce if necessary (just ask the Chinese...they are the new masters of this art).

Ughh...I don't know how to really explain myself here....
 
Last edited:
I know that in the past very high quality audio equipment designers had very narrow specs. for all the parts used. All transistors, resistors etc were checked before being used.

For a reason though. Top end stuff is at the cutting edge of design. Hence critical that components used are are on matching or of better spec. than the minimum requirements of the circuit design.

Krell and AudioLab come to mind with this practice. Very expensive and very,very good at accurate musical rendition. And beautifully built, engineered etc, etc.

Tolerance is normally unimportant for audio circuits, you can use 5% tolerance resistors all the way. Who cares if the gain of an amplifier is 110 or 90? It won't make an audible difference to the volume.

What matters more is low noise and distortion so use metal film resistors, not noisy carbon film, use polypropylene capacitors, not ceramic capacitors which cause distortion.
 
You must remember the tolerance is a promise by the manufacturer. They don't make 1% resistors and label them as 5% and sell them cheaper because they labeled them as 5%.

That's odd.
I have been under the impression for the last 50 years that the tolerances are sorted out after manufacture.
That is the reason that 5% are cheaper than 1%, because there are less 1% available from a given batch.
If the content of the resistor was the only determining factor, there would be no reason to make 20%, just make them all 1%.
Of course, then what would you do with the ones out of tolerance?
 
50 years ago, sorting may have been done. Process control has improved such that nobody sells 20% or even 10% resistors. With 10,000 5% chip resistors going for $13.20 (Digikey) and $19.80 (1%) you can bet they weren't individually tested.
 
Tolerance is normally unimportant for audio circuits, you can use 5% tolerance resistors all the way. Who cares if the gain of an amplifier is 110 or 90? It won't make an audible difference to the volume.

What matters more is low noise and distortion so use metal film resistors, not noisy carbon film, use polypropylene capacitors, not ceramic capacitors which cause distortion.

Speak to the Krell dudes. Yes, caps have more influence....but at the level of performance offered, only the finest quality components are acceptable. Period.

Ever heard a Krell Class A pushing plenty Amps into a speaker and driving it with absolute control???. That my friend is something to behold. I witnessed it way back in 1988.

I stand corrected, but it could of been a little Krell KSA 50 and it was driving a pair of Acoustic Research "AR" and basically took those speakers and wrung the living daylights out of them....on rock, classical,pop. Those speakers going flat out for around 8 hours never showed a sign of strain. It was rock n roll. classical and all rolled into one.

And sounded like being in a concert hall

Live show of note

No, I don't work for Krell but I aspire to their levels of absolute electronic audio design excellence.
 
Last edited:
I Hope this works. I'm trying to put the pictures where they are availiable to forum.

**broken link removed**[/IMG][/IMG]

hope this works.

woodturner5509
 
tvtech,
Sounds like you know a load of audiophools who waste loads of money on components which are way overkill.

woodturner550,
The picture of the resistor doesn't match the reading on the meter.
 
The sixth band is temperature! Now that is interesting. I can understand that maybe the electronics need to be kept air conditioned or maybe used for space in sub zero climate. What else would need a temperature banded resistor?

If I can't use the parts in sub zero climate( I have no need) what can I do with them?

woodturner550
 
It means temperature coefficient which is the amount by which the resistor's value will change with the temperature.

**broken link removed**

You'll normally only find five six band resistors for 1% tolerance and below. This is because there's no point in specifying a 5% resistor to three significant figures and worrying about the temperature coefficient when the value can vary by 5% and it's not used in an application where accuracy is important.

EDIT:
I've just noticed that the six band resistor in this picture is 5%, which is wrong, the person who drew the graphic obviously hadn't thought about tolerance and standard resistor values.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top