[time-nuts] Silicon Labs series of oscillators...
mpb45 at clanbaker.org
Sat Jan 22 14:04:21 UTC 2011
offers a large assortment of various types of oscillator
chips: XO, VCXO, programmable XO, clock generators,
clock distribution chips, Jitter Attenuators, Clock cleaners,
I have a need for a 110 MHz VCXO in a 1.8GHz to 7.5GHz
tracking generator I am building for my Tek 494 spectrum
analyzer. I bought a pair of Silicon Labs 110 MHz VCXO
chips for less than $25 for the pair from Cramer
Distributors. The Si595 VCXO chips are in an
"industry standard" 5mm X 7mm surface-mount package.
Yikes! I knew I was going to have trouble (for lack
of thru-hole leads) breadboarding this chip. However,
I managed (using a magnifier-loupe and a v-e-r-r-r-y
tiny soldering iron tip) to get some "legs" soldered
onto the surface-mount pads. Great... I inserted the
critter into the socket-strips of my breadboard, hooked
up the required 3.3vdc Vdd and ground and checked to
see what it's output looks like.
No joy. Drat. It has a set of complementary output
pins. One sits at around 50% of Vdd and the other is low.
When I pull the Output Enable pin high, the 50% output
pin goes low. The other (complementary) pin just stays
low. If I pull the Output Enable pin low, neither
output pin changes.
Drat. I must have destroyed the little critter during
the leg soldering process. These chips are supposed
to be pretty static from normal handling and-- here in
humid Flori-DUH, handling problems from static build-up
is almost a non-existent problem. Even so, I do all my
breadboarding on a 3-foot X 2-foot static-drain pad.
Sooooo.... I used the utmost care in soldering legs to
the second chip. The surface-mount pads are gold-plated
and it is super easy to just momentarily tap them with the
soldering iron tip and leave a very teensy blob of
solder on each one. Using pre-tinned gold-plated
legs stripped from some surplus 1/8 Watt resistors, I
fastened the legs on the chip with only the briefest
time of soldering-iron tip contact; less than one second,
I am guessing.
Same result with the second chip; the outputs appear to
I guess this sad saga boils down to my question for the
Time-Nutters List: How do you deal with breadboarding
when it comes to parts that are ONLY available in
surface-mount configuration (and are just at the size
limit for hand soldering?
Thanks for any input on this!
More information about the time-nuts