Fluke 8840 Front End Repair

[ItFinallyWorks]

I have a Fluke 8840. Its generally in good condition, but has a bit of an issue.  For positive voltages, its reads off, but in the ballpark (its maybe a few hundred mV off with say 10V in).  I'm not so concerned about this, since I think its related to my other issue.  For negative, it reads approximately correctly up until I get to about -6 to -7V in.  After that point, the input is clamped to read about -6 to -7V (it behaves a little like a diode clamp, increasing slowly as the input voltage is increased greatly).  Since it looks like a clamped input, I tried measuring the input current during this condition, but saw no noticeable increase in terminal current (as close to 0 as a 34401a can read). My strong suspicion is that this is, in fact, a front end issue.  In fact, a coworker of mine used to work at Fluke, on a project to replace the 8840 in fact, and confirmed that it sounds like a front end issue.

I have the schematics for this, and I'm capable of troubleshooting it, but I was wondering if anybody here has had this failure and can shortcut my troubleshooting.  If not, I'll share what I find when I troubleshoot the issue.

[alm]

I haven't seen this problem before. Note that there may be a large series resistor at the input (for example a 10x voltage divider with a total resistance of 10 Mohm and a Thevenin resistance of about 1 Mohm, so the current even if clamped may be tiny. It sounds indeed like something is acting like a diode junction somewhere past the voltage divider (so clampling to about -0.6V results in a 6V reading).

One thing that comes to mind that you might expect to find in the front-end circuit is two series-connected zeners (in reverse polarity) to clamp both positive and negative voltages. If one of them were to fail short, the other would look like a normal zener in one direction and a regular diode in the other.