View Single Post
  #2   Report Post  
Posted to sci.electronics.repair
Cydrome Leader Cydrome Leader is offline
external usenet poster
 
Posts: 1,910
Default It beggars belief ...

Arfa Daily wrote:
Today, I had a biggish Trace Idiot cross my bench. The owner's complaint was
that it produced a continuous buzz (the shop that had taken it in had
quizzed the owner about this, and they were adamant that it was a buzz
rather than a hum).

They claimed that it was unaffected by any of the controls on the amp,
except the "mute" button, which reduced this buzz, but only marginally.

I tried it, but apart from a slight mechanical buzz from the torroidal
transformer, whose bolt was not fully tightened, I could find no problem. I
phoned the shop to recheck the reported fault symptoms, and to see if it was
possible that it was the tranny buzz that the owner was hearing, but again,
the shop said that they had suggested this to the owner, but he was certain
that his buzz was coming from the speaker he had connected to it. Whilst I
was on the phone, I was idly poking at some very thin all-grey wire
interconnects that went between two boards stacked one above the other on
the back panel, and contained various effects send - receive jacks and level
controls, and the main board. If you tugged on one of them just right, the
sound actually went off - but still no buzz was produced. I followed the
wire bunch back to the rear- panel board, and was amazed to see the
termination block easily rock in the pcb. This was one of those slim white
blocks that looks for all the world like a plug / socket, until you look a
little closer, and see that it is just a termination for the wire bunch,
that solders straight into the board.

There are two of these on each of the two boards, so I removed both boards
from the rear panel to see what was going on. When I could see the
underside, I was stunned to see that all four blocks were just sitting in
'virgin' print. Not a jot of solder had ever been anywhere near any pin on
either of the two blocks, on either of the two boards. The actual 'pins' on
these blocks are made from a tail of metal that has been 'split' to form a
springy 'point'. Over the 8 or 10 of them that were in each block, they had
obviously provided enough friction to hang in the pcb holes, and enough
spring tension to make to the thru' plating in those holes. This amp is a
few years old, and is used regularly out on the road, and it is amazing that
this has never even caused the owner a problem with intermittency, let alone
falling out of the board completely.

What I don't understand though, is how this production error can have
occurred in the first place. I could perhaps understand that these
terminations had been called out for hand soldering as a pre-made cable
assembly, after stuffing and soldering of the rest of the board. It's then
fairly easy to see how the actual soldering could have been missed. However,
the board looked like it had undergone flow soldering, and had normal solder
mask over it. But the unsoldered contacts for the terminations were just
naked tinned pads, so how could these have been prevented from taking solder
during the flow soldering process ... ?

It's a mystery ... d:-\


I've seen connections like this before- they appear to be solder-free
where the pins or stakes (the ones I saw were square) were just pressed
into plated through holes. I never figured it what the deal with those
were, but it seemed intentional. Why would one solder hundreds of other
joints, but skip just the weird connectors? If they did have a touch of
solder at the edges of the pins, why were the rest of the voids never
filled in?