I've long been under the impression that a quarter-watt resistor has a standard package, and a half-watt resistor has a standard package, etc. But I've recently gotten through-hole resistors which are not in the expected package size. I've done an analysis of a few thousand of the top stocked through-hole resistors at Digikey, and I see no pattern.

3.5 +- .3 mm length seems to be one grouping, which I would normally have thought of as a 1/8W resistor. But Digikey has resistors listed in this size range of anywhere from 1/8W to 1/2W.

6.3 +- .3 mm length is another grouping, which I'd normally call a quarter-watt resistor package. But the wattage is anywhere from 1/8W to 1W.

Similar groupings can be observed around 9mm, 12mm, 15mm, 18mm, 22.2mm, 26mm, and 45.2mm. This makes creating our internal part numbering scheme problematic; we can't just assume a quarter-watt resistor is in a "quarter-watt package", if such a thing there be.

So is there a set of size standards for through-hole resistors? And is there any relationship between those size standards and the wattage of the resistor?

  • 2
    \$\begingroup\$ Bandaid: Map your internal part numbers to a specific part, and specify suitable substitutes. \$\endgroup\$ – Matt Young Sep 15 '14 at 15:22
  • 1
    \$\begingroup\$ Many years ago, 1/4-W resistors could be assumed to fit a 10mm/0.4" hole spacing, and 1/2-W would not, but this is no longer true as they've introduced resistors that can run hotter. You can do as @MattYoung suggests, or (somewhat similarly) do what I've done- define a 'generic' resistor (which is one of several series, or picked by the contract assembler) and call out specials (which are typically higher performance/smaller so the generics will not fit. \$\endgroup\$ – Spehro Pefhany Sep 15 '14 at 15:29
  • \$\begingroup\$ @MattYoung Unfortunately our MRP software has no concept of alternate parts. Which is idiotic. \$\endgroup\$ – Stephen Collings Sep 15 '14 at 15:51
  • \$\begingroup\$ Neither does ours, we just put them in the internal memo field, and purchasing sorts it out as needed. \$\endgroup\$ – Matt Young Sep 15 '14 at 15:52
  • \$\begingroup\$ @StephenCollings, You have to be kidding. If you can't map an internal PN to a list of approved MPN's, what's the point? \$\endgroup\$ – The Photon Sep 15 '14 at 16:16

Standard Through Hole Resistors are very commonly, though unfortunately not absolutely always, referred to in the size domain by their DIN-size. Almost all manufacturers still make them in the standard sizes, but not all of them refer to that DIN size any more in datasheets or product numbering, as they used to.

This is a four character number. The most common of these are 0102, 0204, 0207, 0309, 0411, 0414, 0617 and 0817. (although the numbers do go higher). Sometimes these are prefixed with DIN-****. As stated size no longer really equals power, as there are different ways of building up the resistive film that will damage at any temperature between 75 and 200 degrees Celsius, where each build-up can be equipped with advantages and disadvantages in other domains.

If you want to implement DIN compatible size numbering I'm sure someone somewhere on Google has scanned an old book about the regimen of sizing using DIN numbers. They have become so normal/standard to me, that I hardly remember the meaning of the numbers. Probably something to do with those silly inches. ;-).

  • \$\begingroup\$ I think JEDEC or IEC probably standardized some THT resistor sizes as well... let me check. For SMT ones I found somewhat related question: electronics.stackexchange.com/questions/101079/… \$\endgroup\$ – Fizz Nov 23 '15 at 2:23
  • \$\begingroup\$ Following the other IEC links from there IEC 60115-2:2014 "Fixed resistors for use in electronic equipment - Part 2: Sectional specification: Leaded fixed low power film resistors" might be [some of] it because "it includes an extension of the list of styles and dimensions"... which is probably defined elsewhere. The IEC standard is probably the same thing as the DIN one these days. \$\endgroup\$ – Fizz Nov 23 '15 at 2:30
  • \$\begingroup\$ For the US ones this forum thread is the closest thing I found so far. \$\endgroup\$ – Fizz Nov 23 '15 at 2:45

"This makes creating our internal part numbering scheme problematic" - IMHO part numbering is a very complex issue. Smart, sane, Engineers in manufacturing companies have "pie fights" over it.

Approaches which can work:

  • Use non-significant part numbers (i.e. almost random numbers) and make it easy to see key attributes 'at a glance'. The unique number ensures you can manage stock, part changes, etc. Providing key attributes helps humans understand what the part is. This carried the problem that people make mistakes, or think the attributes are the part identity, and really, they are not.
  • Use significant part numbers, i.e. something made up from type of part, and values of key attributes, and make it easy to add new attributes fields without breaking any existing systems. This is often hard to do because computer systems were not designed to be flexible enough.

I have worked with companies which did neither, and it was often painful.


My experiance is that there are conventional body sizes for each power rating of carbon film resistor.

However, metal film resistors can tolerate higher temperatures and some but not all manufacturers use this to advertise a higher power rating for a given body size.


Not the answer you're looking for? Browse other questions tagged or ask your own question.