I must respectfully disagree with your statement about handling SMT boards.

I, too, learned to solder by dis-assembling old tube (valve) radios, and
building "Heath Kits" during the 70's.  As technology marched on, I had to
learn how to deal with smaller and smaller parts.  I distinctly remember
cartoons in the electronics hobbyist magazines of that era making fun of the
tiny parts and losing them on the workbench.  I'm sure they were talking
about nothing smaller than 2N3904 transistors in TO-92 packages. 

I have typical 50 year old eyes and shaky hands, but I have successfully
built and reworked many boards with 0402 SMT resistors and caps. 

While there are a few tools you need to work on SMT pc boards, they are
neither hard to find nor expensive.  Most electronics catalogs/websites have
a decent soldering iron with a tip small enough to do the job. If you
already have something better than a Radio Shack wood burner, you may be
able to get a finer tip for a few bucks.  Buy a roll or two of solder wick
while you are at it and save your solder sucker for the big jobs. The way I
install fine pitch, 100 pin ICs is to glob on the solder and then remove the
excess with solder wick.

One trick I use is to buy a pair of "drug store" reading glasses that are
more powerful than the ones I use for just reading.  For really tough jobs,
I combine these with a magnifying lamp I got from Harbor Freight Tools.  I
would never buy their stuff through a catalog, without seeing it in person,
but their magnifying lamp with flourescent lamp is a pretty good unit.  

Good tweezers can be found in the cosmetics section of most drug stores too.
If the tip isn't fine enough, take a file and/or sandpaper to them.

There are tons of tutorials on-line, complete with videos to show you how to
do deal with small SMT parts.  Here's a link to a pretty good one:
http://www.epemag.wimborne.co.uk/solderfaq.htm
I also highly recommend "MAKE" Magazine:  www.makezine.com.  I subscribe to
their print and online versions.

Oh, and lay off the quadruple latte's if you need to do some fine soldering
work!  I have to wait for the caffiene to wear off before I do any really
fussy work....

Good Luck!

	Keith

-----Original Message-----
From: psn-l-request@.............. [mailto:psn-l-request@............... On
Behalf Of Geoffrey
Sent: Sunday, April 01, 2007 2:38 PM
To: psn-l@..............
Subject: Re: My broken DataQ AD converter

Someone who has reworked a PWB (printed wiring board) will most always
leave signs like scratch marks or noticibly different solder work.
Resistors will be cut and left in place rather than unsoldering them
because a PWB can only take so much heat before the foil starts
to peal away. No sense in doing unnecessary soldering. You must look
with a magnifying glass at the solder joints to make sure they are
well wetted and no fine hairs of conductive materials are where they
are not wanted. No resistor should ever be burnt or the board under them.
You really must have a schematic because the job of tracing out one yourself
can be next to impossible. In past years the manufacturers would almost
always
provide schematics but today since most US industry has gone overseas
these forign places will refuse to let you get the schematic thill the
equipment
is old and obsolete.
It is not practicle for the common man to work on surface mounted technology
you
need to work on the new boards is expensive and medical like tools require
fine skills.
If you like electronics as a hobby it is best to go to a university and
become an engineer.
Technologist or if all else fails a Technician.
Without the proper tools throw the old one away and buy a new one.
You will waste your life/monies in trivial pursuits.


----- Original Message ----- 
From: "tchannel" 
To: 
Sent: Saturday, March 31, 2007 12:29
Subject: Re: My broken DataQ AD converter


Well Chris, You did it again!  I hope your head doesn't get too big.....But
facts are facts. From way over there, in the UK?, you 
were able to help me fix this silly AD board, here in Boise Idaho,
USA.......

I do believe you were correct at every turn....

1  I do think this is an older 194 8 bit board, from your physical
description.
2  The resistor numbers on the surface mount components were 1003, and 2003.
Of course the two missing ones had no numbers.  But 
there was a pattern as the resistors were in a row, something like
1003,2003,1003,2003,1003,2003,xxxx,2003, so I figured the xxxx 
was the 1003, 100K
3  The other missing resistor was near others which had the 2003, so I
figured the xxxx would be a 200K, the way it is built I could 
not see where these resistors were in the circuit.  I soldered in the two
missing resistors, and it is now working!!!!!!!

4  I did go to AmaSeis .ini files but did not see any "Channel # listed"  so
maybe it is always going to be Channel 1.
and never Channel 2,3, or 4.
5  I do think you are correct about someone modifying the board to run as a
+5v and a -0, as these two missing resistors just did 
not fail and fall off, they look like they were purposely popped off, small
scratch marks in their place.

Again, Many Thanks, Ted

  ----- Original Message ----- 
  From: ChrisAtUpw@.......
  To: psn-l@..............
  Sent: Saturday, March 31, 2007 12:00 PM
  Subject: Re: My broken DataQ AD converter


  In a message dated 2007/03/31, tchannel@.............. writes:


    Hi All, Looking closely at the board I see two resistors which are
"Open" "cracked"
    Its a wonder it works at all, but some parts of it do work. Channel 1,
only puts out + voltage, just a small amount of -
    I think that account for the AmaSeis trace showing only the top part of
an earthquake.


  Hi Ted,

         The input resistors are two 200 K in series. The offset line is 100
K to the centre point. If any of them are damaged, it 
should not be difficult to replace them. Alternatively, could you  unsolder
the resistors on CH4 and replace the ones on CH1? The 
resistor values are printed on them eg 2003.
         I get about -0.59 V on the offset line and about 90 mV on the
input. I suspect that this is because the ADC is not in 
operating mode. The offset line should be -5V for +/-10V input. Check the
voltage on the offset line while the channel is switched 
on?


    However Channel, 2,3,4 all seem to work showing both positive and
negative movement "In DataQ Software Only", in this software 
Channel 1, only shows positive trace movement, and, no negative.


         You could get this with failed resistors. Is the range 5V or 10V? I
suspect that the ADC may have been 'modified' for a 
single polarity 5V signal? Can the friend who gave it to you provide any
information?


    When I connected my Sensor to Channel 2,3 or 4 Using DataQ,  I was
thinking it would then work in AmaSeis, but it does not.  It 
looks as if AmaSeis will not read any Channel except Channel 1. Even though
in DataQ Channel 2,3,4 all seem to work fine.  If I 
connect the sensor to Channel 3,  AmaSeis shows a trace, but the trace does
not move in either direction if I stimulate the sensor. 
This is true for Channel, 2,3,4.  Only Channel 1, moves the trace in
AmaSeis.
    I think you can change the com port, in AmaSeis, but is there anywhere
you can or need to change for Channel selection?
    In DataQ I closed all the Channels except Ch. 3 and saved before
closing.  So If I were just using DataQ software, I would be 
operational.


         Amaseis normally uses CH1 only. However there are two setup files
Amaseis.ini and AS1.ini. It may be possible to set a 
different channel by editing and then saving the file, but I am not sure. I
know that you get your different sample rates this way. 
If you use a Dataq and Amaseis without editing the ini file, you just get 6
SPS.

         This information is taken from an old 8 bit DI-194 board, which I
suspect that you have.

         Regards,

         Chris Chapman 

__________________________________________________________

Public Seismic Network Mailing List (PSN-L)


__________________________________________________________

Public Seismic Network Mailing List (PSN-L)