I have a customer with a Siemens Modular Metering service on a strip mall. The tenants have been complaining of flickering lights, GFCI's unable to reset, among other issues. After a little research I found what I believe to be the source of the problem. There is a fused 800A main ahead of three meter stacks which has the correct bonding from what is visible. At the first stack there is a difference that varies between 10 and 20 volts between the EGC and the neutral. Between A,B, and C phases and the EGC is about 122 volts, but to neutral its 108, 118, 145 volts. It seems there is some resistance in the neutral buss connection somehow between the main disconnect and the first meter stack.
This morning I shut the main off and re-tourqed the single bolt joint on the horizontal buss. No change.
Has anyone experienced this? Any options aside from replacement?

Did you check all the terminations within the main disco?
Neutrals & GECs/EGCs at the bonding point?

Is the issue within one or all of the meter stacks?

Only visually as far as tightness. They are directly behind the POCO's connections. I did check the voltage between the neutral/gnd buss within the main disco and the chassis and read 0 volts. Between the same buss in the main disco and the neutral bar in the meter stack I get the same fluctuating voltage as in the meter stacks alone.
The issue exists within all of the meter stacks.

There may be an intermittent connection across a set of Bus Jumpers for the Grounded Conductor, as the recorded Voltage between Lines A, B & C to the EGC shows 122V; but Lines A, B & C to N show 108V, 118V and 145V.

If so, it is likely due to an incorrectly installed Belleville Washer, or the Bolt / Nut were never Torqued.

--Scott (EE)

Neutrals tend to be lightly loaded because they carry the unbalanced load and that is a very high voltage difference. Maybe the bolt and nut are missing.

Is the grounded conductor aluminum? I found a problem once with a feeder where the conductor degraded inside the insulation.
We tore into a wall and found a bubbled spot where the conductor was overheating. Good idea to shut it down and check everything. An overheated connection could break loose
and BOOM! Unrelated but check for missing buss supports also. I just opened an older Federal Pacific pull section to find 2 of the critical main buss supports missing.

True for resistive loads, but if there's lots of computer equipment, their power supplies will draw pulses of current that don't match up with other current pulses on the other phases. AKA harmonics. In this case the neutral gets really hammered.