Connect with us

Soldering surface mount components

Discussion in 'Electronic Basics' started by Daniel Kelly \(AKA Jack\), Jun 8, 2004.

Scroll to continue with content
  1. Hi there Jack

    Check out

    for some soldering advice. There is a basic soldering and SMD article on
    there, and SMD breadboards.

    You should be able to put the chip on by hand. I do.

    1.You need a good HANDS FREE magnifier, one of them geeky head set
    flip-up magnifiers. OPTI-VISOR is one I use, and can get different
    lenses for them and has a little monacle magnifier that you push over 1
    eye to see even closer. A bench mounted, spring folding magnifier and
    lamp combo can work, or a goose neck one with a heavy base. You got to
    have both hands free to see, and work. Do not try them stupid little
    tweezer/magnifier combo deals for $3-5, they are worthless. And you
    usually are not lined up to see into it while soldering.

    2. Get a low wattage iron, 10-20 Watts, or so. Don't need a lot of heat
    for the chips. Ebay has Metcal MPX style soldering iron sets with a iron
    and a stand. They are first or 2nd generation now, but still way better
    than ANYTHING on the market for soldering. 5-10 second warm up times.
    2nd generation power supply unit has 2 iron connectors! 1st gen. has
    only 1 iron connector on it. They all use the same type of tips, and can
    get IC size tips, 0603-1210 size tips, many other special ones. What you
    need for IC's is a HOOF tip. Its an angled tip, with a depression in the
    end, to hold some solder.

    3. Tack solder the chip on 1 corner. Put TINY bit of solder on a corner
    pad. Get chip in right PIN #1 Position FIRST. AFTER getting the chip in
    place, REGISTER the corners and sides so ALL the pins line up. If some
    are bent, use the tweezers you have to move them. OF COURSE YOU ARE
    Solder 1 corner, and make it quick, no more than 5-7 seconds on a pad.
    BEFORE COMMITTING YOURSELF!! Then use a lot of flux, preferably from a
    flux pen, and coat the pads and pins good. Then with a WELL TINNED TIP,
    put a small bit on the tip, put it on one side of the pins, and
    LIGHTLY!! DRAG IT ACROSS PADS/PINS to the end. If not enough solder on
    the side, do it again, AFTER fluxing a little more. Flux should keep the
    shorts from happening, if you go fast enough, too slow and it may leave
    a blob on 2-3 pins. Not to worry. Finish up all sides of chip.

    4. FLUX again - Then go back with fine tip and try to drag the blob over
    2-4 pins or so, to get rid of short(s). With FINE PITCH 0.8/0.5 mm QFPs
    and such, just be a little more careful with pressure on pins. For the
    0.5 mm you will need a good magnifier. You can also use the desoldering
    braid, SODER-WICK, or some type of braid to pull solder shorts off pins.

    5. You WILL need a good magnifier, a DECENT iron and different types of
    tips, 1-2 different type of tweezers, and a flux pen for all SMD type
    soldering. I have a bent fine point pair of tweezers, for the tiny stiff
    and picking up QFP's by 1 pin! And a thicker, straight utility pair,
    with square tips for bending the wires, or getting a better grip, and
    general use, as its a heavy pair. Contact East, Techni-tool, or any
    electronics catalog should have what you need. Some tools may cost, like
    the tweezers, but they are for life, unless you drop the fine tip one on
    the floor!

    Hope this helps out.

    Leif Erickson

  2. No, that is not right at all. A material can be piezoelectric
    even when configured such that no resonance can occur.
    Mechanical resonance is an extensive property, applicable
    to a specific object with particular boundaries. In contrast,
    a material itself, (regardless of its shape), has piezoelectic
    parameters defining an intensive property, applicable to
    infinitessimal volumes.

    Perhaps, if you are going to "correct" people and set out
    to "educate" them, you should be careful that you are not
    simply throwing out some vague notions that you have
    inadvertantly collected.
  3. Clarence

    Clarence Guest

    I am not sure I understand what your trying to say. It appears to be
    I was about to reply, then I read your disavowal. I see you are aware that you
    are only vaguely familiar with the phenomena. A piezoelectric material which
    has not been shaped, or which has no electrodes attached to excite the
    piezoelectric properties is not affected in a predictable manner. So if a
    material with those piezoelectric is applied in a product it "MAY" behave in a
    way which creates stress. However AFAIK piezoelectric quartz, barium titanate,
    or other piezoelectric materials are NOT routinely used in the production of
    capacitors. It takes relatively high voltages to get much movement in
    piezoelectric materials. Lose molecules would not have any effect at all since
    it is the matrix of crystal formation which exhibits the property.

    I designed equipment to solder leads on crystals and we measured the flexure in
    tens of micro inches. Placing material in a position where movement was
    inhibited, as in a potting compound, reduced movement to fractions of a micro
    inches. Resonance would vanish with inhibiting pressure applied.

    Unlike a ferrite, or crystal quartz, alumna used for SMT components do not
    exhibit piezoelectric properties.

    If your aware of any other piezoelectric materials please list them. I can
    think of many applications for a low cost piezoelectric operated mechanical

    BTW: The soldering process should Twine the piezoelectric material if it was
    present. Heat destroys the piezoelectric properties in the same manner as
    magnets are demagnetized.
  4. Ian

    Ian Guest

    Look in the FAQ tab - 48W.

  5. Ken Smith

    Ken Smith Guest

    Both X7R and X5R are slightly piezoelectric. Neither is good enough to be
    put to practical use as an actuator. If you put about 500mA at 2KHz
    through a 2.2uF X7R cap mounted on a PCB you get just enough motion that
    you can hear it.

    Somewhere around here, I have a paper by Advanced Technical Ceramics which
    talks about this subject. They had quite a nice collection of reference
    material related to the use of ceramic capacitors in RF power
  6. John Fields

    John Fields Guest

    Frayed knot, Clarence.

    Also wrong:

    1. Your statement, in another post, that piezoelectric materials
    aren't use in the manufacture of ceramic capacitors. Check it out
    and you'll find that barium titanate is quite often used as a

    2. From another post, you imply that "alumna" is used for the
    dielectric in surface-mount caps. I don't believe that's true, and
    I believe that what you're referring to is "alumina",
    (aluminum oxide) the material which is used for the substrate of
    surface mount _resistors_. Take a look at the dielectric constant
    of alumina VS that of barium titanate for a clue.

    You also seem to be laboring under the misapprehension that merely
    shaping a material like barium titanate will imbue it with
    piezoelectric characteristics. It will not; poling is a necessary
    step in the process and, while it can occur spontaneously under
    certain circumstances, it's usually forced.

    All of this information is available by Googling for it and you do
    yourself a great disservice by not availaing yourself of it before you
    "Open mouth, insert foot".
  7. Clarence

    Clarence Guest

    500mA is a rather substantial level of power for that value. It takes 39.3
    Volts sinusoidal directly across the cap. I could find no part which is rated
    to withstand that level of ripple current at these frequencies. All were rated
    for .03 to .09 ohms ESR for frequencies over 100KHz.
    As little as a reliable motion of 0.002 for a 100V bias would be useful.
    Assume a bar shape for maximum deflection.

    So how much motion do you get at 100KHz? The entire unit is 0.180 long, 0.120
    wide, and 0.080 high (max) with tolerances of +/- 0.010 Inches. The cap
    element is coated with a resilient encapsulate.

    The AC component is 30VRMS max and the DC is 80V.
    I find nothing that confirms the property, nor the magnitude of the motion.
    BTW, NASA, and FAA do not provide any warnings on these devices.

    How about the X5V material?
  8. John Fields

    John Fields Guest

    mA is current. mW is power.

    1 1
    Xc = ------- = -------------------- ~ 36 ohms
    2pifC 6.28*2000Hz*2.2E-6

    E = I Xc = 0.5A * 36 ohms = 18VRMS ~ 50.9VPP
    What levels of ripple current were you able to find which these parts
    could withstand at 2.2kHz?

    If you search the literature, you may find that piezoelectric
    actuators lie typically in the microinch of length change per inch of
    material per volt of excitation. I think even Edmund might have some.
    If not, Google "piezoelectric actuators" for a clue.
  9. Ken Smith

    Ken Smith Guest

    I didn't suggest it was a good idea to put that much AC through the poor
    thing. I just said that it screems if you do.

    BTW: 1/( 2 * PI * F * C ) = 36.17 on my slide rule

    0.5 * 36.17 = 18.08 Then we have to add a little for the ESR.

    Also the ATC capacitors are rated for several Amps at 30MHz.

    A lot depends on what you call reliable. If you want cheap use a plastic
    piezoelectric element. There are many makers of piezielectric ceramic
    elements. If you are doing something like pushing a grating to tune a
    laser, don't use the plastic stuff. It is a good thermometer.
    I've never been interested enough to measure it.
    Try google. I'll bet you will find something from AVX's web site that
    confirms the effect and says words like "but our do it less".
    Im not surprised about the FAA.

    NASA does surprise me a bit. They wouldn't allow PVC parts to be used in
    the U2/ER2 because of the fume risk to the pilot. This even applies to
    the parts that are at outdoors pressure.
    I think all high K materials do it to some degree.
  10. (snip)

    Agreed. I have a little pump and dump LED pulser that charges a 1 uf
    100 VDC X7R 1210 smt capacitor. I gcharge it up to 50 volts over 900
    us and dump it through the LEDs with a peak current of almost 1 amp in
    100 us with the current peak controlled by a series inductance. I can
    always tell when the circuit is working by the thin whine emitted.
    Mylar caps in the same circuit made no noise.
  11. John Fields

    John Fields Guest

  12. Thanks?
  13. Clarence

    Clarence Guest

    Before it was stated that they would be cracked by the piezoelectric effect!

    I buy caps by the roll, and they are installed by many different people, some
    by flow soldering, some manually soldered by a convention Iron. Never heard a
    peep, nor had a failure due to a crack from soldering. Thermal cycling has
    causes a few failures, but only a few. Less than 0.1%!
  14. Clarence

    Clarence Guest

    "Ken Smith" wrote
    Did you allow for ESR? I did!
    Which is NOT 2Khz!
    I'm looking for a pump impeller an implanted medical application.
    I Goggled, and found no published studies which gave numbers. Only that the
    motion was detected.
    The FAA is very stringent about any condition which might cause a fair or spark
    on a board. For instance "Gold caps" often used to maintain a RTD are not
    acceptable since they fail (rarely) by out gassing with a flame.

    I used many of these in a satellite program.
    Very small Degree!

    These got clipped from my last mail:

    Per AVX:
    Effects of Mechanical Stress – High "K" dielectric ceramic capacitors
    exhibit 'some' 'low level' piezoelectric reactions under 'mechanical
    stress.' As a general statement, the piezoelectric output is higher,
    the higher the dielectric constant of the ceramic. 'It is desirable to
    investigate this effect before using high "K" dielectrics as coupling
    capacitors in extremely low level applications.'

    Also found:
    Piezoelectric Effect. One minor drawback for X7R dielectrics has been
    their sceptibility to 'piezoelectric-induced stresses.' Although this
    effect is 'marginal' and 'may be neglected for case sizes smaller than
    the 2220,' for larger capacitors it can lead to catastrophic failures
    caused by cracking. Modifying the dielectric composition to avoid any
    piezoelectric effects within the range of operating frequencies may
    skirt this problem.
    Whereas the standard X7R material shows piezoelectric noise, this is
    almost absent from the new ceramic. These measurements were made
    under a 300-V bias at 90°C. Note this effect is not present in film
    or aluminum-electrolytic capacitors. With this new development,
    ceramics now offer a viable alternative to film and electrolytic
    capacitors in the large case sizes.
  15. Ken Smith

    Ken Smith Guest


    ESR = sqrt( (39.3/0.5)^2 - 36.17^2) = 69.78 Ohms

    Holy smoking capacitors Batman were do you buy these caps so I can avoid
    Yes I noticed. The spec. is based on the heating. It doesn't vary that
    much with frequency.
    You have bio compatibility issues then too don'y you? The plastic stuff
    is based on Kynar and nickel plated IIRC. Does the Kynar cause trouble?
    Will you keep it out of contact with the host? The Kynar causes larger
    motion than the ceramic. The force it can produce is much less and the
    losses are greater.

    There are not many pilots on these satellites are there?

    AVX is trying to sell you these capacitors. The fact that the makers seem
    so unwilling to put a number on it makes me suspect that they are trying
    to downplay it. I know that the piezoelectric effect can be real trouble
    in low signal level analog circuits. they make microphones out of the
    coupling capacitors.
  16. Clarence

    Clarence Guest

    Decimal error somewhere. I got 3.6 ohms which is worst case for the family.
  17. John Fields

    John Fields Guest

  18. Terry Given

    Terry Given Guest

    I'm curious about using Y5V - you have a 100uF 10V cap. TDK spec Y5V as
    +22%/-85% over -30C - +85C range. Using the AVX data (TDK dont seem to
    mention this) the voltage coefficient is -90% at 60% rated Vcc, -80% at
    40% Vcc. Assuming Vcc = 4V = 40% of rated voltage, you therefore only
    have 20uF of overall capacitance. You probably use a 10V cap at more
    like 6-7V, so its more like 10uF. Add in the temperature effects, and
    its looking more like 1.5uF.

    X7R = +/- 15% over -55C - +125C. AVX give -10% at 100% rated voltage. A
    22uF 10V X7R cap comes in a 1210 package from TDK, and will have MORE
    CAPACITANCE than the 100uF part you are currently using. Unless of
    course you are using it at very low voltage (< 20% rated), with
    temperature between 0-45C. Given that this is a space application, I
    doubt the temperature is that well controlled.

    Mind you you might be using the cap as a DC bias/temperature/vibration
    detector, in which case Y5V would be a very good choice.
    No its not. The IEEE standard dictionary of electrical and electronics
    terms, 6th ed. defines piezoelectric effect as:

    "Some materials become electrically polarized when they are mechanically
    strained. The direction and magnitude of polarization depend upon the
    nature and amount of the strain, and upon the direction of the strain.
    In such materials the converse effect is observed, namely, that a strain
    results from the application of an electric field."

    Note the complete and utter lack of the word "resonance" or any of its

    It also appears you are suggesting there are no forces whatsoever inside
    a capacitor, other than those caused by the piezoelectric effect. That
    would be a fairly neat trick if it were true, and would doubtless be a
    major concern to the manufacturers of electrostatic loudspeakers, which
    by now ought to have all stopped working in light of this revelation.

  19. Clarence

    Clarence Guest

    Actually,NO! But the higher values are generally too expensive.

    This has nothing to do with part damage due to a piezoelectric property. Many
    types of Capacitors including AIR dielectric are subject to "PURELY MECHANICAL"
    I used to use WOBULATORS to modulate a VFO for a particular type of FM. Driving
    a split stator ball bearing variable capacitor with a motor to generate a wide
    frequency variation at a controlled rate. So what has this kind of mechanical
    change of capacity have to do with piezoelectric damage to a part?
    Since I do not use the larger package size, I see no relevance.
  20. Clarence

    Clarence Guest

    Rest assured that the components we used were applied properly and with the
    approval of the design staff, and NASA, who along with myself, recommended this
    approach to solve another problem.

    This appears to be a change of subject from the effect being due to forces from
    piezoelectric characteristics. I see nothing relevant to discussing this!
    I'll concede the IEEE is correct, I should have quoted precisely.
    Again, your attempting to change the subject, unless you are asserting that
    electrostatic loudspeakers operate on and are damaged by piezoelectric forces.

    I have not addressed such speakers, and INDEED have no interest in them.

    I think this is no longer a matter of sharing different experiences, rather is
    appears to be you trying unsuccessfully to prove you knowledge is greater and
    more correct than someone else's. I don't care to play. I designed weapons
    system components for use in Vietnam which are still working and in use. I
    have nothing I care to prove to you. Your entitled to your opine. So am I!
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day