Testing to see if you really killed the Zipit

Before you ask, no, there isn't a fuse that you can replace...

A lot of people seem to have plugged the wrong power supply into the Zipit and killed the Zipit. The Zipit electronics appear to be only able to handle a maximum of 6 volts on the external power plug. Here is a quick way to determine if you totally wiped the electronics out or if it is in the Zipit power supply circuit.

The Zipit power supply consists mainly of a Linear Technologies LT3406 (U12) to derive the +3.3 volts from either the External Supply or the LiION battery. The LT3406 has an absolute maximum rating of 6 volts, if you plug in an external supply greater than 6 volts you will let the magic smoke out of the regulator!

There is a way to see if the main board is completely wiped out. You can make a cut with an X-acto knife and use two fresh alkaline AA batteries to try to power the Zipit. Do not use the 4 volt LiION battery for this test, that will blow the electronics!!! Typically, a new akaline AA cell is 1.6v, two of them will yield 3.2v, this is well within 5% of 3.3 volts.

TopStep #1

Open the bottom cover, pop the speaker loose from the case and tack it onto wax paper or plastic wrap to protect the adhesive. Next, use a sharp blade with your X-acto knife and carefully cut the PCB trace that runs between the inductor (L2) and pin 3 of the LT3406 (U12 marked as LTA5). You can cut the trace by simply bearing down on the knife, it is not necessary to "slice" it, just make a pressure cut. Make two cuts in the trace and scrape it clear.

Make the cut as shown below.

TopStep #2

Once the trace has been cut, you have now disconnected the +3.3 volt regulator from the rest of the Zipit. Using the two akaline AA batteries in series, connect the GROUND (negative) onto the large metal shield of the WIFI transmitter (large 1" square metal shield). Next, using an X-Acto blade, inject the POSITIVE battery voltage into the Zipit on the other side of L2 (inductor) from where you cut the trace.

See picture below for where to connect (note that the AMBER LED is lit?).

TopStep #3

The GREEN LED of the Zipit should come on, this is good, the ARM CPU controls the LEDs. Wait a moment and the GREEN LED should go out and the AMBER LED will now come on. If you've gotten this far, you have just verified that the Zipit power supply is burnt and the main electronics appear to be okay.

However, if the LEDs do not cycle or are not lit at all, you have a fried Zipit (or lousy connections to the batteries).    :( The power button is bypassed at this point, if you cannot get any LEDs to light, don't bother with pushing the button.

TopOk! It is only the supply!

What to do with a bad power supply? Some things to check.

  • Well, you can always try ordering an LT3406 from Linear Technologies and replace U12. Also, jumper the trace you cut when freeing up the 3.3v supply (L2 .. U12 pin 3 cut).

  • You may have also blown the LT4054 (U11) as well. This is the LiION charger and may be dragging the External DC supply voltage down (shorted).

  • Or, you could have blown CR11 / CR9 (PNP transistors of unknown part number). Something like a Philips BCX17 (SOT23, PNP, 500mA) device may work here (DigiKey P/N 568-1640-1-ND)? CR11 and CR9 select between the LiION battery or the External DC source.

  • CR14 is appears to be a 1 Amp diode to protect against reverse polarity. Please note that CR14 is also involved (with CR11 & CR9) in selecting which power source is activated. Check to see that this diode is good with an Ohm meter in the diode test position.

  • CR12 is a BAT54C, though unlikely it is bad...

  • You may want to check to see that FB15 is not open for some reason.

  • Another solution is to put together another 3.3volt power supply to replace the bad supply. Tuck that into the case somewhere.

  • Or, use an external 3.3volt supply of some sort and use the Zipit as a web server, or remote weather station, or a target system to develop software for your "other" Zipit (the replacement you got from ebay ;).