"Tom" wrote in message ...

On 11/14/2015 2:11 PM, highlandham wrote:
On 12/11/15 18:20, Robert L Wilson Jr. wrote:
On 11/7/2015 8:25 PM, Jeff Liebermann wrote:
I just repaired an old MFJ-269 antenna analyzer. This time, the 4
diodes around the RF connector were NOT blown. Instead, it was crappy
SMT soldering for the components surrounding the diodes. The solder
joints looked ok before I resoldered them, but were obviously a bad
connection because several resistors just fell off the board when I
touched one end with the soldering iron, indicating the other end was
not properly soldered. With the resistor removed, the PCB pad looked
like it had never been soldered. If you don't have good soldering
tools, a steady hand, and a good microscope, DON'T try this as you'll
probably make a mess, as I did before I realized what was happening.

Because the various parts on the RF board were not originally making
good contact to the PCB pads, the calibration is off. The display
reads about 12% too high for impedance and dead on for reactance. I'm
debating if it's worth calibrating.
http://www.w8ji.com/mfj-259b_calibration.htm

Now working on an MFJ-259A, which looks like the usual blown diode
problem:
http://802.11junk.com/jeffl/MFJ-269-repair/

I've generally found MFJ stuff has decent design and rather good
components, but terrible wave soldering on the PC boards (and sometimes
hand soldering on components off the board like chassis mounted
connectors). On my 259B, it worked well for almost a year, and then
became erratic. When I saw how many soldered joints weren't, I decided
not to try to figure out which ones were bad: I just went around and
resoldered everything on the board. Since then it has worked perfectly.

Wave soldering is a marginal thing, I think it is amazing how many times
it does work. Think about how it goes: Basically the PC board is
suspended just above a vat of molten solder, and a wave is caused to run
across the vat. The wave, we hope, just reaches up to the connections on
the board and solders them. Too high and it makes solder bridges. Too
low and it either misses connections or at least does not stay on them
long enough. And it is not long on a connection at best, so does it get
it hot enough?

It is not just fairly inexpensive things like my MFJ analyzer that have
had problems. I did the "resolder the whole board" thing on my Phase
Linear preamp many years ago also, and that was by no means low end!

Bob Wilson
=======
What strikes me in this newsgroup thread , time and again, is that MFJ´s
quality control people (if they exist) pass badly soldered PCBs and
other imperfections which affect their products´ reliability . But for
some reason they get away with it. To me this seems to be an attitude
brought about by management.........Martin F. Jue ?

Frank , GM0CSZ / KN6WH





# It is too often that some of these things get past QC. The good news is
# that some of their products, like their linear amplifiers have allowed
# the ordinary ham to run power. .

MFJ seems to have gotten a bit better as the years go by.
Once I bought something (a cw filter perhaps) that had the PC board mounted
by wedging it at an angle and using a bead of epoxy to hold it in place.

I haven't seen that kind of sloppy work from them in a while.