Agilent 34401A error 611 (solved)

[peertux]

Hi,

I've bought an 34401A and the self test gives error 611 also the readings are noisy and quite a bit off.
Some readings with shorted leads:
Ohms 2W: starts around 53 Ohm then ramps down to bounce between 41.3 and 41.7 Ohms
mVDC range:  bounces between 40.7 and 41.2 mV
1.2VDC range: between 0.00926 and 0.00940 V
12VDC range: -0.0006 V so seem to be ok
Low amp range: between 0.155 and 0.159 A
high amp range: between 0.0412 and 0.0419 A

Error 611 description according to manual:
DC gain x100 failed This test configures for the 100 mV dc range. The
dc amplifier gain is set to x100. The measure customer ( MC ) input is
selected to the internal TSENSE source which produces 0.6 volts. A 20 ms
ADC measurement is performed and checked for a +overload response.

I did some measurements while in mVDC setting and I think that the voltage drop across CR203 is to big.

All measurements are in Volts.

Does anyone have a suggestion how to fix this? Personally I think U201A is the culprit.

[Kleinstein]

The shown circuit part are the current sources for the Ohms measurement. Would not expect the error in this area. To check this one can measure the current in ohms ranges with a second DMM.

The trouble is more likely with the DC amplifier and maybe the switches at the amplifiers input (e.g. the one to make the zero measurement). So more like U101 (which would be a bad thing) or maybe U106 / U153 or similar.  A first point would be to check the supply - just in case.

Even the 0.6 mV in the 12 V range are not a good sign.

With a 1 mA current source the 40 mV in the mV range leads to the 40 Ohms reading. So these two a re consistent. However 9 mV in the 1.2 V range is different - so it is not just a simple offset to the amplifier.

Are the readings with AZ mode or not ?

[peertux]

Thanks for your reply Kleinstein.

The supplies are within spec. I did the measurements with AZ on (I didn't change the standard setting).
Some additional measurements with shorted leads in DC mode:
                                    39 Ohm
38mV          3mA            9 Ohm
0.0088V      0.8A          90 Ohm
0.00060V    0.16A      900 Ohm
0.31V          0.04A     0.12 kOhm
0.031V                         12 kOhm
                                   1.2 kOhm (heard a relay)

When I look at the Ohms range there appears to be a problem with the multiplier.
I did some measurements at the input amp everything seems to fluctuate especially after U106. I changed zener CR103 but the power to U103 keeps fluctuating.

[Kleinstein]

Usually the current sources (part around U153) for the input amplifier should be really stable. Already the 16 mV at the input to U103 is odd. However some parts here are relatively high impedance and might get upset from noise / 60 Hz picked up by the probes.  There is also a small chance that there is some ripple of oscillation of some kind at some of the signals or supplies.

So it might be a good idea to use a scope to debug the input amplifier.

[peertux]

I made a pic of U153B output sorry for the angle it's not easy to hold a probe and a camera at the same time

[Kleinstein]

The signal at U153 should be essentially DC if everything is OK. It kind of suggests that some part of the circuit is kind of oscillating or there is a lot of RF signal coupled in.

The oscillation could already come from the 10 V reference or the supply (-18 V, +18 V and the + 13V behind CR103).

[peertux]

The 10v reference doesn't oscillate according my dmm but there is oscillation after CR103 (not on the 18v line) I already changed the zener but the oscillation is still there. btw the voltage after CR103 and the values displayed on the screen of the 34401A changed a bit after changing CR103 so there is a correlation there probably.

[Kleinstein]

I won't expect the zener to be a problem. It is possible that there is decoupling capacitor no longer connected somewhere (e.g. cold solder or crack in solder  or cracked internally).

One could try adding something like 100 nF at the supply of U103 and maybe U153.

Is the AC part changing a lot when the gain changes ?
If so it would suggest that U106 might be involved.

U105 and Q104 could also cause trouble - without sufficient voltage there, the input FETs could work in a wrong way.

[peertux]

I tried 100nF caps over the supplies of U103 and U153 to no avail.
I desoldered R131 to see if the output at U103A would clean up but it was still there, it leaves when I probe pin 5&3 of U153 probing pin 2 and 6 has no effect.
Also without R131 the voltage after CR103 is a steady 14.583 so no more oscillation there.

I'll have a look and see how the AC part behaves.

Edit:
Ac part shorted leads:
000.0800 mV last 2 digits oscillating
others are 0

[Kleinstein]

Without R131 U103-B should run into saturation and thus it can no longer oscillate or couple the AC part to the supply. How could there still be the oscillation (AC signal) ?

It might be worth checking the voltage at Q104: insufficient voltage there might make the input JFETs oscillate.
Another point would be if the input is for some reason open - this could also make the amplifier unstable.

[peertux]

I decided to leave the dmm alone for a couple of weeks because I was afraid I was going to test how big of a dent it could make in the wall. 
I measured 18.8V at the source of Q104 so that seems fine.
I ordered a couple of AD706JRZ on Aliexpress to replace U103 because it looks to me that the nastiness starts at U103B.
I'll post an update after they've arrived.
Thanks for the help Kleinstein 

[peertux]

The AD706 arrived yesterday and after replacing U103 the error is gone and the DMM works like a charm. 

[HighVoltage]

Congratulations on a successful repair.