From: cd@notformail.com   
      
   On Sun, 13 Aug 2023 07:34:36 +0100, liz@poppyrecords.invalid.invalid   
   (Liz Tuddenham) wrote:   
      
   >Dave Platt  wrote:   
   >   
   >> In article ,   
   >> Cursitor Doom   wrote:   
   >>   
   >> >>Micro crack in the solder?   
   >> >   
   >> >It'd have to be a pico crack given the hi-mag examination I gave it   
   >> >and found nothing suspicious.   
   >>   
   >> Possibly a bad wave-soldering job?  I've heard of cases where   
   >> bad/incorrect solder or flux caused a mostly-failed joint... the   
   >> solder flowed up over the component lead, formed a decent-looking   
   >> meniscus, but didn't actually bond to the lead properly.  The   
   >> lead could eventually crack away from the solder _inside_ the   
   >> joint, leading to an open circuit which is invisible to the eye.   
   >>   
   >> You might want to try solder-sucking away the existing joints,   
   >> re-fluxing, and re-soldering, and see if that fixes it... although   
   >> I'm not sure if this would teach you anything more than you'd   
   >> learn by just unsoldering the resistor and measuring it out of   
   >> circuit.   
   >>   
   >> An internal crack in the resistor is probably more likely, though.   
   >   
   >If it were a large heavy component, I would look for circular cracks in   
   >the rings of solder around the terminals; the solder blob in two parts   
   >with one attached to the pin and the other as a circle surrounding it.   
   >This used to happen mainly where the dip-soldering conveyor was vibrated   
   >to shake off solder splashes and mass of the large component stressed   
   >the cooling solder blob.  The centre of the blob was cooled by the pin,   
   >the periphery was cooled by the track and the last bit to solidify was   
   >stressed into a ring of 'dry joint', which later failed almost   
   >invisibly.   
   >   
   >I can't imagine that happening to anything as small as a surface-mount   
   >resistor, but include this historic information for the benefit of   
   >anyone repairing through-hole boards with large heavy components.   
      
   No, this is through-hole and about 35 years old. As a mattter of fact,   
   it also suffers from the issue you mentioned above, in the PSU   
   section, where the heat from a couple of 5W resistors exacerbates the   
   problem. In fact the original PSU only lasted about 10 years before   
   requiring a rebuild as a result of this particular design flaw.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)