Short or expected operation?

I am experimenting with an AD9958 chip that has 56 pins (all underneath the chip!) that I'm attaching to a SchmartBoard prototyping board. At the moment all I've done is to solder the chip to the board - there are no other connections.

Before I started to load the board with the other components, I decided to see if any pof the pins were shorted together by using my DMM and checking adjacent lines.

I have found several pins that show up as short circuits but I can't see any solder bridges etc, even using my 100(+) USB microscope - unless they are underneath the chip itself!!!!!

One thing I have noticed is that the pins that appear to be shorted are all "digital" output pins that are next to a digital VDD (+1.8v) line. Looking at the datasheet it does show the equivlent circuits for the digital I/O pins in question but it does for the +3.3V digital I/O pins and they are 2 FETs in series betwqeen the supply rails with the inputs to their common gates and the outputs from between the 2 FETs.

Assuming something similar as the equivalent circuit for the higher voltage digital ports, I was wondering if it is possible that I'm actually measuring the path through one of the FETs rather than an actual short circuit?

Thanks

Susan

Put your DMM in the "diode" test mode, and repeat the measurement, both ways (swap the leads). If you read 0.05 or less, that is a short. If you read ~0.6 or ~1.2, that is the parasitic diodes inside the CMOS structure.

Thanks for the quick response - I'll try it out when I get home tonight.

Susan

Mike,

Going by that test it is a short - sigh.

Thanks for the tip - I know it will be used again in the future.

Susan

Have you used one of these? https://www.schmartboard.com/schmartboard_pd_202-0011-01.pdf

You need an adaptor like that, but the problem with the schmartboard adaptors is that they are not allowing for the massive ground pad in the middle of the AD9958. You should trim the longer pads in the middle of each side so that they do not extend more than 0.7 mm under the IC.

You can solder ICs like the AD9958 from the edges with a good iron, by heating the PCB tracks that stick out from under the IC. I need a binocular microscope for that. To unsolder one, you really need a good hot air gun.

The big pad underneath the AD9958 is there for heat sinking. You might have problems on a board that doesn't have a big area of copper there. I don't know how much the power consumption of the AD9958 depends on the clock frequency. If it does, you can reduce the power by clocking it slower.

Actually is is more like https://www.schmartboard.com/schmartboard_pd_202-0019-01.pdf except that it doesn't have the pad in the middle (not sure why not but it doesn't).

I took the chip off the board last night and found a couple of bridges between some pins which is probably the source of the problem.

I have been using my "frying pan reflow" method but I've just bought a new soldering iron that can take a fine enough tip (my old Weller that I've used for the last 20+ years finally broke last week) so I'll try hand soldering the chip this time to see how I go.

Thanks

Susan