I have multiple adjustable DC/DC buck converters in a design and I'm looking to save lines in the BOM for assembly. Is there any significant drawback to using more than two resistors in a feedback network?

For example, if I need R1 to be 300k and R2 to be 150k can I replace a single 300k R1 resistor with 2 150k's in series? Assuming I use high precision resistors and can tolerate an increased ±% output tolerance what other negatives might I encounter?

Same question applies for resistors in parallel.

DC/DC Converter


Using multiple standard value resistors in the way you suggest (to stay away from hard to find parts or for BoM proliferation reduction) is a very common practice.

Unless you are trying to get the best possible predicted reliability (as found in MIL-HDBK-217) there really is no practical downside providing you have space on the PCB (although if you cannot fit a couple of extra 0402 parts, things must be really tight).

Note that MIL-HDBK-217 reliability predictions are really a numbers game, so minimising components maximises reliability, all other things being equal.

Note that if you replace a 300k, 1% part with two 150k, 1% parts, you still yield 300k, 1%

  • 1
    \$\begingroup\$ In reality, slightly increasing the number of unstressed components has very little effect on reliability, of course. Worst-case tolerance is no worse, and typical might be a bit better (or not). \$\endgroup\$ – Spehro Pefhany Apr 20 '18 at 16:07
  • 3
    \$\begingroup\$ Of course, Spehro; I was looking at the extreme case only for reliability and tolerances are generally never worse in a measurable way. \$\endgroup\$ – Peter Smith Apr 20 '18 at 16:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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