SparkFun will be closed Nov 26th and 27th for the Thanksgiving holiday. Orders placed after 2:00pm MT on the 25th will ship out Monday the 30th.


Member Since: February 15, 2006

Country: United States

  • It would be unlikely that the wire bonding process is done by soldering the wires. Usually this is done using thermosonic bonding, which uses a combination of heat, pressure, and ultrasonic vibrations to bond the wires. At least this is how it is done for IC packaging, so it is unlikely that they use something completely different for COB wire bonding.

  • It looks to me like there is a polarization mark visible from the top side. There is a small circle in the gold lead frame on the lower right corner of the part in the first picture. There doesn’t appear to be a matching mark on the upper left corner, so this could be used as a polarity indicator, as long as there was not too much solder applied.

  • Check out MG Chemicals for chemicals for this. They even have a dry resist film now which can be applied using a laminator, which would have made my job much easier long ago if it had existed then. Alas, there was no internet back then, so finding things was much harder.

  • Digi-Key has photosensitive PCBs. Make sure you get the matching developer. These could require positive or negative artwork, so you’ll need to print the artwork to match (but usually positive now I think).
    These boards have an emulsion on them that reacts to light, much like film. After developing, you will have a positive image of the traces of the PCB, which resists the chemical attack of the etchant. They still need to be etched just like any other board. The advantage, is that you can produce much finer lines, and usually there are very few pinholes in the traces.
    When I was much younger, around 24 years ago, I made many boards using this process. I had the photosensitive resist in liquid form, and I applied it to the board. This prevented wasting board, if something went wrong in the process. I could strip the board, and reapply the resist.

  • The key is to drill at high speed with a sharp bit. Steel bits only last for 20-30 holes before getting too dull. You can tell by how the copper lifts around the edge of the hole, after it’s been drilled, which can also remove the pad if it is small. Carbide bits when drilled at around 20krpm+ don’t have this problem, and can drill hundreds of holes. If drilling by hand, they will usually they break before they get dull, as all it takes is one slip, and the bit is broken. You can buy resharpened carbide bits from a number of places, for around $0.50 each, so it is worthwhile to use them.

  • Drill holes using a Dremel drill press at high speed using carbide bits. Steel bits don’t last for many holes before getting dull in fiberglass. You must use a drill press with carbide bits, as they are very brittle, and break easily.
    If you print your artwork, with holes shown in the pads, then a small dot will be etched away in the middle of the pad which makes centering the drill easy.
    Etch boards in a sealable plastic container to help with smells.

  • Thanks for fixing the links. I couldn’t figure out how to stop them from italicizing, and obviously the underscores had something to do with that.

  • While I’ll agree that the wait times in Canada are sometimes too long, it really does cost us less money per capita than it does in the USA. You should read:
    which makes some useful comparisons between the two systems. The USA spends more per capita than any other country in the world (http://en.wikipedia.org/wiki/Health_care_in_Canada). The WHO ranked Canada at 30th, and USA at 37th for quality of care worldwide.
    The recent case you quote was for a small child that has no chance for recovery. The Canadian medical system did not want to spend further funds on a child that would not live. This sounds to me like a just decision, but I can understand the plight of the parents. The outcome regardless of whether the US stepped in or not, is that the child will die. So I don’t see how the US helped in this case.

  • It would also appear that the minimum spec. for CVDD is 2.4V, and it looks like the regulator used for this supply is 1.8V.

  • It might be worth noting that there is no logic level translation on this board, so connecting output pins on the Arduino directly to this shield is not a good idea, unless the Arduino itself is powered from 3.3V.
    The inputs are only rated for voltages up to a max of 3.6V based on the I/O supply used. At the very least a series resistor should be used to limit the current into the protection diodes.

No public wish lists :(