What is the typical failure mode of resistor networks? (Dale 472G)

[lobust]

While chasing scsi problems on my Quadra 800, I have determined that RP4, a Dale 472G (bussed, 4.7K) has some abnormal readings on a couple of it's legs, and those same legs do have continuity to the some legs 53C96 scsi controller (no extensive testing yet, not sure which legs specifically).

Those legs read around 1K, with it still soldered to the board. I have not removed it from the board yet.

Is it common for those to fail to a lower resistance as opposed to open or dead short? Or is there some other path causing an erroneous reading, and it will measure correctly when I remove it from the board?

I have ordered some replacements (couldn't get the original Dales, so I got some Bourns 4816P-2-472LF which seem like a reasonable equivalent) and will go ahead and try it when they arrive, but just curious and would like to expand my knowledge.

Red-handed culprit or wild goose chase?

 

[Elemenoh]

I've been working on a bugged SE/30 board and ran into the same issue checking the Bourns filter packs. Out of circuit they measured something like 45k across, but in circuit was a variety. I compared to a referenced known-good board. You can see the specific values here if you're curious. Short story is that the other components in circuit impact the readings across the resistor packs, at least in the case of the SE/30.

 

[lobust]

I've been working on a bugged SE/30 board and ran into the same issue checking the Bourns filter packs. Out of circuit they measured something like 45k across, but in circuit was a variety. I compared to a referenced known-good board. You can see the specific values here if you're curious. Short story is that the other components in circuit impact the readings across the resistor packs, at least in the case of the SE/30.

Yep, just wishful thinking as it turns out - removed it from the board and it measures good on all legs.

Ho hum.