Fluke 8840a/af 4-wire resistance inoperative

[Excavatoree]

I have an old (35 years) Fluke 8840a/af bench meter that I purchased from an E-bay Fluke seller. (funny that I'd know of any, ha ha) 

As far as I can check, everything works  (voltage (AC and DC)  and 2 wire resistance) except the 4 wire resistance mode.   I can't believe it's much, but I've not had any luck finding a service manual. 

Can anyone give me a direction in which to start?  A 1.5 ohm resistor (confirmed using Dave's "poor man's 4 wire technique of using two meters")  reads .5 ohms. 

[alm]

Did the current from the force terminals match the specification? It is not a particularly tight spec, but it should not be 66% low . What voltage do you measure across the sense terminals?

This manual has schematics. I think the /AF version was quite similar (some options default and maybe tighter specs?).

If the values read too low only in 4W mode, then that suggests to me that there is a low resistance path between sense low and sense high. Maybe one of the protection JFETs (Q301/302) is leaky?

[Excavatoree]

Correction:  The 2 wire reading is 1.5xxx (correct) and the 4 wire reading is 1.2yyy.   The error does not scale.   Of course, it's so small for most measurements in the higher ranges, it appears that the 4 wire readings are correct.

I removed Q301, Q302 and clamp Q309, with no effect.

I checked Q303, and even substituted another part, but there was no effect.

I've got to be missing something.  I just can't figure out what could be causing the 4 wire reading to be wrong without affecting the 2 wire mode.

[alm]

How do the voltage and current compare between 2w and 4w mode? I do not have the schematic in front of me, but what about any switching (e.g. mux) that is only used for 4w?

[RobK_NL]

Have you checked the front/rear switch and U301D?

[Excavatoree]

How do the voltage and current compare between 2w and 4w mode? I do not have the schematic in front of me, but what about any switching (e.g. mux) that is only used for 4w?

I can't see anything different so far.

Have you checked the front/rear switch and U301D?

I have bypassed the front/rear switch with no effect.
I'll be honest, I messed up (got turned around and or backwards) when checking voltages at U301 and noticed that when I connect pin 13 to ground via my 10 Megohm Fluke 87,  the reading is correct.  I'll be honest again and admit I have no idea what that could mean.

I've checked the Jfet switching between Q310 and Q301 via U305 and can't find any problems.

[alm]

I can't see anything different so far.

So current source switching is fine, and the problem is somewhere at or downstream of S301.

I have bypassed the front/rear switch with no effect.
I'll be honest, I messed up (got turned around and or backwards) when checking voltages at U301 and noticed that when I connect pin 13 to ground via my 10 Megohm Fluke 87,  the reading is correct.  I'll be honest again and admit I have no idea what that could mean.

Messing up is an excellent way of finding a way out if you are stuck. Too bad it is often associated with extra work and expenses . Pin 13 is one of the input nodes to the ADC. Does it do the same if you ground pin 17 or 18? If not, then it sounds like a bad solder connection or break in trace/via.

According to the table on page 93, U301C and U301D are involved in switching between two and four-wire mode. An open U301D or shorted U301C might explain the symptoms. If grounding the S/H input fixes stuff, then my money is on an open/high-resistance U301D (pin 12/13), either a broken IC or connection. In 4W mode, you should measure close to 0 V between R314 (C306 side) and Q314 pin 3.

[xwarp]

If you find any bad chips, send me a message as I have a couple 8840 boards that I have salvaged parts from.