View Single Post
  #15   Report Post  
Posted to sci.electronics.repair
Arfa Daily Arfa Daily is offline
external usenet poster
 
Posts: 6,772
Default Is this due to RoHS solder?



"Jim Yanik" wrote in message
4...
"Ian Field" wrote in
:


"Boris Mohar" wrote in message
...

Took apart a Volvo 850 fuel pump relay. Cracks everywhere.

http://www.viatrack.ca/Misc/FP%20Relay/



I thought automotive was exempt - after all you wouldn't want the solder
falling off your FI relay half way through an overtake.




Honda fuel pump relays have the same solder problems,have had them for
many
years.
my 1994 Integra had it,I resoldered the PCB and it worked fine for many
years afterwards. I saved at least $50 for a few minutes work.

the relay failure mode was that on a hot day,the car would start
briefly,then die,and only crank for 15-20 minutes(no fuel pressure),then
the car would start and run normally(after the car's interior cooled
down).No further problem until it got another heat soak.
In the mornings,the car would start and run normally.

I think the solder crystallizes with age,then thermal cycling cracks it.

the automotive environment is very tough,hard on electronics.
lots of thermal cycling,shock and vibration.

--
Jim Yanik
jyanik
at
localnet
dot com


Looking at the pictures, I think it is highly unlikely that the relay is
manufactured with lead-free solder. Although the joints have a somewhat
crystaline appearance, the surfaces of the individual 'grains' still appear
too shiny to be lead-free. Also, if you look at the shape of the joints,
they are much more conventional solder than lead-free solder. Whilst I would
agree that there is rather a lot of poor joints to be seen, again, I don't
see any that are typically consistent with a 'volcano' joint that you get
with lead-free. Annular cracks in the actual solder, rather than cracks
around the component lead where the solder has not taken to it, are far more
common with leaded solder. I would say that we are looking at a combination
of circumstances here, including age, temperature cycling and the generally
harsh automotive conditions that it lives in, relay hammer from the relay
operating many many thousands - if not millions - of times in its life so
far, and the possibility of the solder having been 'over-cooked' in the
first place. Smitty with his knowledge of production soldering may be able
to comment better than I on that.

Also, as Nigel says, the RoHS directive was not implemented with a vengeance
in most of Europe, until June 2006. That relay board is fully 10 years older
than that. Like him, I don't think that I have seen any electronic design
that old, that was implemented in lead-free solder.

Just as a matter of interest, I had a 1985 VW Passat for about 10 years. It
was the 2 litre version with the 5 cylinder Audi engine fitted. It had a
fuel pump actually inside the tank, and this was controlled by a relay on
the main board at the front of the car. Towards the end of the time I had
it, so probably around 1996, it started to suffer intermittent engine
stops - always in embarrassing places like four-way junctions, of course. It
would run for weeks without so much as a cough, then it would just die
without warning. Not too bad if you were tooling along on a big highway. It
just lost power and slowed down, giving you time to get off onto the side
shoulder, before you had no power at all, but worst was when you pulled away
from a stop light, and it just died ...

I discovered that you had to wait exactly four minutes - not a second less -
before turning the key and trying a restart. Invariably, it would then start
and run as though nothing had happened. It took me a long time to track the
problem down, but it eventually turned out to be bad joints in the fuel pump
relay. As I recall, it looked in pretty much the same state as the OP's one.

Arfa