Difference between 5% and 1% resistors

Discussion in 'General Electronics Chat' started by Crispin, Nov 15, 2011.

  1. Crispin

    Thread Starter Member

    Jul 4, 2011
    88
    2
    Hi Folks,
    Now that version 1 of my project is done, I am redesigning it after tips obtained here. One of the things I am paying more attention to are the voltage dividers and resistors for op-amps and using only 1%resistors for those parts.
    What is the difference between a 1% and 5% resistor? Is it the value or the amount of drift?
    I have some 5% ones which are exactly 10k and a 1% one which is slightly below 10k. Is that just luck and I would find that the 5% one would drift over time / temperature?
    Drift would be the biggest issue because I can adjust the base-line values in code.
    The 1% are metal film and the 5% are carbon film.

    Cheers,
    Crispin
     
  2. jimkeith

    Active Member

    Oct 26, 2011
    539
    99
    The % rating refers to the initial resistance tolerance. They are available in 5%, 1% and 0.1% tolerances.
    Temperate Coefficient is commonly specified for metal film resistors at 50ppm/°C maximum.
    The other is long-term drift--I do not know much about this other than to avoid carbon composition resistors.

    Carbon film are not bad even though they may not have a tight Tc spec--very low values e.g. <50Ω and very high values >500K are not as stable and can drift over the product lifetime.

    Since resistors are so cheap, go with the 1% metal film and be done with it--For thru-hole, I use the RN55 type for small size and availability.
     
    Last edited: Nov 15, 2011
  3. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Cheap is relative. I have made a complete kit of 5% ¼W resistors. It cost around $40, each resistor was around 2¢. In most cases you don't need greater accuracy.

    Project: Resistor Parts Storage
     
  4. Crispin

    Thread Starter Member

    Jul 4, 2011
    88
    2
    I bought a lucky packet of the 5% ones which gives me a shed load of most common values. This was a great starting point.
    Problem is now that I buy the 1% as I need and while for 1 or 2 it is ok but I erks me that the shop is making such a profit on them (£0.24 per resistor).

    Off to find a 1% lucky packet now which will be a nice starting point.
    Off to read Bill's thread.
     
  5. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Firstly, to answer your questions, the % tolerance has nothing to do with temperature drift. The tolerance gives the % deviation that the real value is allowed to be different from the stated or nominal value.

    For example, a 100-ohm 5% resistor can be anywhere between 95 and 105 ohms.
    If you find a 5% resistor that is exactly the nominal value, i.e. 100 ohms, then that is just plain luck.

    1% resistors and other electronic components are over-rated, and that is to say that they are unnecessary in a large number of applications.

    In analog circuits, in many cases the resistor could be off by as much as 100% and you would not notice the difference, that is, a 1K resistor may be replaced with a 2K resistor and you would not see the difference. The devil is in the details.

    In digital circuits for example, a CMOS circuit may call for a pull/up resistor of 10K. You may find that any value from 1K to 100K may work. Again, the devil is in the details.

    When would you choose a 1% component? You require more precise components when you require more precise behavior such as precise gain, frequency response, timing or exact matching of two circuits. This is even more important in manufacturing if you desire each unit in the production line to have the same characteristics. Each unit may be acceptable on its own but the variation in components will result in each unit behaving slightly differently.

    In a lot of cases, it comes down to cost of a manufactured product. If you are using 1% resistors at a higher cost than 5% resistors, then you could be wasting money.
     
    PackratKing likes this.
  6. Crispin

    Thread Starter Member

    Jul 4, 2011
    88
    2
    Thanks!

    So if the tolerance is about the value and it would not drift then I would be fine with my 5% ones. It’s a once off project and all the “calibration” can be done in software (MCU) at design and left as is.
    So long as a voltage divider remains, within reason, what I was expecting it to be, all is good.

    I did find someone on fleabay who is selling 6100 (610 different values) of 1% metal film resistors for £38. A bit too much for me and my projects ;)
     
  7. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    I have lots of old 5% that have drifted way out of spec over time. 1% are much better.
     
  8. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    It is really the materials used, the construction, and the quality of manufacture that determines long-term drift. In principle, this is not inherently related to the selection tolerance. In practice though, you are more likely to find highly stable resistors in the closer tolerance ranges. This is partly because many applications which can accept looser initial tolerances can also accept more drift, so the extra cost involved in providing a lower drift component may not be justified.

    Carbon film types are cheap, not that bad for general purposes, but not so good for a precise circuits like say a measuring instrument. They are not the same thing as the old carbon composition resistors. Although the latter do have some good characteristics, like low inductance and excellent short-term overload capability, they are generally not very stable and can be quite noisy.

    If you want really good long term stability there is no doubt that metal film (or even metal oxide) are better. You pay for the extra performance though, no point in that if it is not needed.

    Twenty-four pence for a metal film resistor seems pretty exorbitant though - what power rating are they, and who from - some backstreet reseller of RS Components?
     
  9. debe

    Well-Known Member

    Sep 21, 2010
    947
    184
    A relativly cheap way of starting up a 1% 1/4W Metal film resistor set is purchase a pack of 300 such as RR-0680 from Jaycar for $19.95. The plasic bags i get for $3.75 for 100 from another suplier.
     
    Last edited: Nov 15, 2011
  10. Crispin

    Thread Starter Member

    Jul 4, 2011
    88
    2
    For the adhoc ones I need, Maplin is pretty much across the road from me. They're a great baseline for what you should not pay :mad:

    I might order the ebay 6100pcs lucky packet and be done with it. Will sort me for life :D

    Maybe a Farnell order would be in order...


    thanks for the info folks. For now I can stick with what I have (4 loose 1% ones) but next time I order I'll compare prices.
     
  11. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,395
    1,607
    I'm gob stopped that you found a single 1% resistor with 5% marking.

    Manufacturers typically make a batch of resistor and then sort them for tolerance. Thus all the one percenters should have been in a different bin.

    This must have been an incredibly tight lot where they had to take from the 1% bin to yeild enough 5%.
     
  12. Crispin

    Thread Starter Member

    Jul 4, 2011
    88
    2
    Off to buy a lottery ticket :D
     
  13. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Here is my guess as to what is happening. Long ago they could not make 1% resistors. Therefore 1% resistors were selected out of the batch by measurement. You ended up with 5% resistors with a bifurcated distribution with nothing at the nominal value. (The best were taken out and sold at a higher price).

    Today the process control is good enough that all the resistors meet 1% tolerance. They sell the resistors as 5% or 1% at whatever price the market will bare. Conclusion: buy 5%, they're as good as 1%.
     
  14. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    I'll second that, I buy through-hole resistors in boxes of 1000, on tape (same value). Metal film ones are often purchased as 5% and marked as 5%, but when testing with a good ohmmeter they are better than 1%.

    I came to the same conclusion MrChips did, that the manufacturer just runs a large batch with good equipment and they are all 1% resistors, then they label some as 5% as they still have customers that want to buy 5% resistors! :)
     
  15. Gdrumm

    Distinguished Member

    Aug 29, 2008
    684
    36
    What about longgevity?

    I know that electronic products from China are usually cheap, and only last for a year or two, but what if you want to build something to last for a long time. Does that change the equation?
     
  16. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    There was a time you had to worry about resistors around 30 or more years ago. Technology changes though, and long term stability is not a problem.

    The main thing is pay attention to the 50% rule of thumb, don't use a part past 50% of its absolute ratings for best performance. If you have a resistor that will dissipate ¼W, then use a ½W resistor.
     
  17. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    When I did industrial electronics in the mid 80's they trained elec guys for the steelworks (where things get hot and downtime of machine costs big $$$) and the de-rating guideline was 10%. So if a resistor actually dissipated 1W you use a 10W resistor. Needless to say the place was full of electronics that worked since the 50's and 60's and never broke down. :)

    I still use a similar method when designing auto electronics and industrial control electronics. My stuff doesn't break.

    But have you noticed that typical white block style 5W resistors are considerably smaller now than they were 20 years ago?

    And the big question, have you tried running one at 5W?

    I did a while back (I emprically test power components when designing) and it got so hot its legs got hotter than the melting point of solder and it fell off the test wires. Apparently wattage these days ain't what it used to be... Grumble grumble. ;)
     
  18. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    I built a clock kit about 20 years ago that had the same problem, with 1W resistors! I replaced them and gave them enough lead length that the heat had time to dissipate, still, if I were to design the same clock I would have spread the load over several resistors instead of just one.
     
  19. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    I hear you. We buy 2W resistors now, that are the same diameter as 1970's 1W resistors, but shorter! With only 80% of the mass they must run a heck of a lot hotter at 2W than good old 1W resistors run at 1W...

    But then manufacturers don't expect (or want) their products to last now. That hurts sales if their products last more than a year.

    I repaired cooked caps and resistors in a friends DVD recorder that was only a few months old! In my opinion that's pretty despicable designing a $300 item that fails from cooked 2 cent parts after only a few months... The PCB that had plenty of space actually had the electros and resistors etc next to heatsinks, all very deliberately designed to cook parts and give a short life.

    Anyway, sorry for off-topic ranting! ;)
     
  20. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    I suspect that the present situation of selling mountains of cheap but very nasty products, many of them from China, will not last forever. I do not expect to see the end of it, but younger and healthier folks probably will.

    At some time in the future, it seems likely that someone will come back to using reliability as a selling point. This appears to have been part of the reason for Japanese success at a time when UK industry was not taking it seriously enough, from around the later 1960s. I wonder what region may lead a return to quality in future?
     
Loading...