[dmjlambert]

You have to be careful not to poke the tester through the printed circuit, especially if it is a sharp pointed tester like mine that is made for poking through wire insulation. What you're dealing with is the metal case of the instrument cluster, which is grounded, then a thin layer of plastic, then the thin foil trace on the plastic, then the metal tab of the bulb holder or of the cluster plug, then plastic bulb holder or plastic cluster plug. So it is a 5 layer sandwich that is all pretty much wafer thin. Except where the bulb holder contacts, screw hole ground points, and cluster plug tabs make contact to the traces, the traces are coated with another layer of insulating plastic. The trace on one side of each bulb is supposed to be grounded anyway, but the other side is 12V feeding the bulb so you don't want to short the trace through the thin plastic to the ground of the metal instrument cluster case. Check that you have a 3A or smaller fuse (and not 5, 10, 20 A, etc.) in the top right PNL LTS position of your fuse block. That will provide some protection for the printed circuit, but only for the gauge illumination lamps. The direction signal lamp traces are unfused, so get unlimited current.

I don't want to discourage you, you need to go ahead, because Rick is on the most likely path and has provided some excellent info. If the printed circuit manufacturer didn't do a great job of making sure the traces around the screw hole ground points can make good electrical contact with the screws, those appropriate traces won't be connected to ground reliably and could cause the problem. My 2 cents.