Author Topic: Parallel path resistor contamination for large value resistors (~1MEG and up)  (Read 1849 times)

0 Members and 1 Guest are viewing this topic.

Offline Jim from ChicagoTopic starter

  • Regular Contributor
  • *
  • Posts: 53
  • Country: us
  • Power Electronics
Some colleagues of mine were saying to avoid using large value resistors whenever possible (~1MEG and up) because contaminants (e.g. dust in a humid environment) over years can create a parallel path resistance that is low enough to significantly lower the total resistance. My question is, is it possible to have a rule of thumb for what the value of this parallel "R_contaminant" is worst-case? For example, if I have an 0603 resistor is it reasonable that contaminants could form a parallel 10MEG path across it? Or is that way too low a value that would ever be seen?

I realize it's a difficult question to answer because it depends on the environment and application. Nonetheless if you've dealt with this issue before I'd be interested in any thoughts.

 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 13948
  • Country: de
So far the parasitic leakage on the board I encountered was more in the 10-500 Gohm range - at leat with a reasonable clean board. It may get a lot worse with mold growing and than high humitity.
With modern PCBs the solder mask helps quite a bit by keeping the contamination away, though appications that want super low leakage (e.g. Tohm target) may not want the solder mask.
Another really bad case was with some of the old style glue (AFAIK some neoprene type) that over time changes from white / yellow to brown and becomes conductive. There I once got some 100 K.

Resistor of more than 1 Meg usually have a relatively large TC and often not so great stabilty on there own. So it is not just dirt and board leakage. The better / thin film types often max out at 1 M.
 
The following users thanked this post: Jim from Chicago

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 730
  • Country: us
Use a conformal coating to protect the board from leakage.
 

Online tszaboo

  • Super Contributor
  • ***
  • Posts: 7227
  • Country: nl
  • Current job: ATEX product design
So far the parasitic leakage on the board I encountered was more in the 10-500 Gohm range - at leat with a reasonable clean board. It may get a lot worse with mold growing and than high humitity.
With modern PCBs the solder mask helps quite a bit by keeping the contamination away, though appications that want super low leakage (e.g. Tohm target) may not want the solder mask.
Another really bad case was with some of the old style glue (AFAIK some neoprene type) that over time changes from white / yellow to brown and becomes conductive. There I once got some 100 K.

Resistor of more than 1 Meg usually have a relatively large TC and often not so great stabilty on there own. So it is not just dirt and board leakage. The better / thin film types often max out at 1 M.
I've found "no clean flux" together with a small amount of water, due to condensation, completely wreck my measurements, in the tens of KOhm range. The sensible solution is to keep the electronics away from condensation of course, and say no to "no clean flux".
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 13948
  • Country: de
No clean flux is mainly a problem when not fully heated. It than can be a little hygroscopic and even corrosive. This can be an issue especially for small parts flowing off to the sides when using a iron.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf