Connect with us

Magnetic Drive Etchant Pump

Discussion in 'Electronic Design' started by D from BC, Aug 7, 2007.

  1. D from BC

    D from BC Guest

  2. Jim Thompson

    Jim Thompson Guest

  3. AKA peristaltic pump.

    Best regards,
    Spehro Pefhany
     
  4. D from BC

    D from BC Guest

    That sounds familiar...

    I think that's the pump often seen on medical dramas. It's used to
    pump blood.

    D from BC
     
  5. Jim Thompson

    Jim Thompson Guest

    Yep, and NaK and other such fun stuff ;-)

    ...Jim Thompson
     
  6. whit3rd

    whit3rd Guest

    Is this for agitating the etchant while etching? Pumping is only one
    way
    to agitate, others are stirring (titanium stirring rod with plastic
    vanes
    on the end) and bubbling (hook an air pump to a perforated nest-of-
    tubes on the
    bottom of the tank).

    If your pump has enough pressure, you can spray-etch (park the boards
    over a sump,
    and spray directly onto the surface, let etchant drip off). That
    causes some mist,
    though, and EVERYTHING in your workroom will rust. Spray-etching is
    not
    an application for a 'mini' pump. With suitable (titanium) heaters to
    keep the
    etchant at the right temperature, it IS quite efficient, though.
     
  7. D from BC

    D from BC Guest

    It's for agitation and flow...
    I'd like to try to get a river of etchant flowing down the board.
    I think I read somewhere that there's a risk of over-etch.
    However, I'm not making brain implants.

    Spraying would be more pro but I don't think I need my boards that
    good.. A little rework is tolerable.

    I'm interested in using minimal amounts of etchant that way I don't
    have to play with a tank.
    It's safer and quicker to handle small amounts of etchant..

    I believe a tank is required for a bubbler set up.

    The flow(fountain) idea I have just fills the tray enough to prevent
    the pump intake pipe from sucking air.

    ~^-----<-----+
    ~ / |
    / ^
    \ / / |
    \/\/\/\/\/\/\/ |
    \__tray____/-->[pump]
    D from BC
     
  8. D from BC

    D from BC Guest

  9. The problem will be nonuniform etch. Wherever the flow
    thins the boundary layer, etching will speed up. But you
    will have a hard time getting a fairly uniform boundary
    layer everywhere. Little corners will be more stagnant, and
    traces that run along the flow will have lots thinner
    boundary layers than traces than run across the flow. Spray
    etchers try to overcome this by producing a violent, random
    impact pattern that has as uniform an average density as
    possible. But they are never perfect, especially
    considering the effect of run off drool and hanging drops.
    You might try convection etching. The board is supported
    right on the surface of a shallow puddle of etchant, etching
    taking place on the under side, only. As the copper loads
    the etchant, the density of the etchant increases and that
    loaded etchant falls to the bottom and fresh, lower density
    etchant rises to make contact with the boundary layer. No
    additional agitation is needed or is helpful. Usually I can
    find a plate or saucer that will support the four corners of
    the board over a quarter inch or so deep puddle. On a few
    occasions, I have taped flexible copper laminates to a slab
    of polycarbonate (sold in building supply stores as safety
    glass for screen doors), all around the edges, to etch this
    way. It provides enough buoyancy that the whole assembly
    floats like a raft.

    The trick is to get the board in contact with the surface of
    the puddle without trapping any bubbles between copper and
    etchant. I dunk the board in etchant to wet it and brush
    etchant on any spots that seem to resist wetting, till the
    whole surface is coated. Then I hinge the board very slowly
    onto the surface of the puddle so the air is expelled by
    the surface tension of the etchant. After a couple minutes,
    I slowly lift one edge and have a quick look, to make sure
    the whole surface is etching. Then repeat the slow hinging
    dunk.

    The convection process is so efficient at delivering fresh
    etchant to the ceiling of the cell that it etches faster in
    the upward direction than it does sideways, because the
    sides of the etch slots and holes are coated with a sliding
    boundary layer of used etchant. This effect produces almost
    vertical walls and can etch slots as narrow as the thickness
    of the copper.

    If your board material is translucent, you can watch the
    etch finish through the back side. Otherwise, every 5
    minutes or so, you have to hinge the board back and take a
    look. The process is gracefully slow, but surprisingly
    uniform. This process also produces minimal splashing and
    mist formation, so is more suitable to kitchen table
    operation than things that make fine droplets that corrode
    everything within a couple meters of the etching.
     
  10. D from BC

    D from BC Guest

    Doh!
    Long ago and new to etching, I used to attach the PCB to long pieces
    of masking tape and then suspend the PCB upsidedown near the surface
    of the etchant.
    I only did that because every other way didn't work as well without
    agitation.

    At present...I use a foam brush and pamper the PCB with fresh ferric
    chloride.

    I like the float idea...
    Perhaps sticking styrofoam on the PCB with double sided tape might
    work to.
    D from BC
     
  11. amdx

    amdx Guest

    Just want to add my very positive experience with "convection etching".
    I saw the technique somewhere on the net and gave it a try. It works very
    well and I found it reduced the etching time considerably.
    As I recall, I put a little handle in the middle of my pcb using masking
    tape, then just floated the board on the etchant. I was very impressed with
    this method.
    Mike
     
  12. Glen Walpert

    Glen Walpert Guest

    <snip entire discussion of board etching methods>

    A method I used with good results when I made my own boards long long
    ago was the heated rocking tray board etcher.

    A pyrex baking tray was temperature controlled with a heater and
    thermocouple glued to the bottom, using a cheap temp controller from
    Omega. Something over 100 F is required for decent etch time with
    ferric chloride, ISTR using 110 or 120. The tray was glued to a piece
    of plywood with pins in the ends supported in grooved blocks on a base
    so it could rock, and a small motor with eccentric roller used to rock
    the tray at about 1 rock per second, perhaps 3/8" peak to peak at the
    edge of the tray. You want a wave of etchant to slosh neatly from
    side to side with no splashing so some experimentation on amplitude
    and speed may be required. Board supports were cut from rubber
    erasers and glued to the bottom of the tray to hold the board at about
    1/4" from and parallel to the bottom of the tray. 1/2" of ferric
    chloride etchant was preheated (a glass cover reduces evaporation
    losses), the board placed on the supports, and the tray set to rocking
    for something on the order of 5 min for double sided 1 oz copper. With
    the board height set just right both sides etch at the same rate;
    otherwise flip it over mid-etch.

    This method is probably not suited for fine pitch boards, but for
    occasional use in a small shop I like it better than the equally
    effective bubbler circulation in a vertical tank, because it does not
    tend to fill your shop with etchant mist. And you can use the heated
    tray to start seedlings in the early spring :).

    Glen
     
  13. I have no experience with Ammonium P but I suspect that when
    it is loaded with copper it is denser than when it is fresh.
    Does it produce any gas when it reacts with copper. That
    would spoil the convection method.
     
  14. Nico Coesel

    Nico Coesel Guest

    Years ago I build a pump completely made out of PVC except for the
    motor (which is far away from the etchant anyway).
     
  15. D from BC

    D from BC Guest

    I'm guessing those magnetic drive pumps are the centrifugal type.
    It's tempting to DIY.

    ***=====>
    * \ / *
    * - O - *
    * / \ *
    I'll guess at the design..
    Rotor has embedded neodymium magnets.
    The motor head has matching magnetic positions (not shown).
    All parts can be laser cut acrylic.
    Dunno about an easy rotor blade shape. Paddles? Curved?
    I'm imagining that the magnetic pull on the rotor could cause
    friction.
    But I suspect there's a way to get the rotor to "ride" on a fluid
    layer for just about 0 friction..

    Heyyyy....I'm supposed to be doing electronics design! :(
    D from BC
     
  16. D from BC

    D from BC Guest

    Bummer...I can't remember.. Last time I used Ammonium P I had both
    copper and aluminum in the etchant..(It wasn't a normal PCB project.)
    Dunno if one or both metals made little bubbles or if the bubbles were
    just trapped air.
    D from BC
     
  17. Guest

    [...]

    Almost float?. I've never had one sink. Surface tension keeps 'em
    aloft. I put 'em in dry having pre popped any stray floaters with a
    cocktail stick. Loop of tape on the back for launching/retrieving and
    preheat the etchant in the microwave. (40degC about 4 minutes etch).
    Yes. Floating is the best method. (moreso for the occasional PCB)
     
  18. Rich Grise

    Rich Grise Guest

    I'm guessing those magnetic drive pumps are the centrifugal type.
    It's tempting to DIY.

    ***=====>
    * \ / *
    * - O - *
    * / \ *
    I'll guess at the design..
    Rotor has embedded neodymium magnets.
    The motor head has matching magnetic positions (not shown).
    All parts can be laser cut acrylic.
    Dunno about an easy rotor blade shape. Paddles? Curved?
    I'm imagining that the magnetic pull on the rotor could cause
    friction.
    But I suspect there's a way to get the rotor to "ride" on a fluid
    layer for just about 0 friction..

    Heyyyy....I'm supposed to be doing electronics design! :([/QUOTE]

    How about an aquarium pump with all plastic parts?

    Good Luck!
    Rich
     
  19. D from BC

    D from BC Guest

    It's possible my boards won't float...
    I'm doing the poor mans ground plane and gluing 2 boards together
    prior to etching. Like this....

    top
    --------SMD components
    --------Cu
    --------FR4
    --------Epoxy glue
    --------Cu
    --------FR4

    It might be bizarre but my app needs the FR4 on the bottom for
    isolation reasons..
    .....And it's also nice I don't have to put etch resist on the
    additional copper layer.
    D from BC
     
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

-