We have a water chiller with 2 separate refrigeration circuits into a shell and tube evaporator. These are equipped with electronic expansion valves and the superheat has been checked on the controller and via electronic manifold and temperature sensors. Super heat is 8.0K on both circuits.

Subcooling is around 10K, discharge temperatures are around 73°C. The refrigerant is R454B.

The problem we are seeing is that on the off cycle the oil level in both compressors on a circuit is at the top of both sight glasses. If we operate on compressor on its own the oil levels remain stable. When we switch on the second compressor, with a few minutes the oil level drops in the second compressor, this is actually always the right hand compressor in the pair.

After about 15 to 20 minutes to oil level in the right side compressor drops to the very bottom of the sight glass. If we continue to operate the compressors the oil level in the right side compressor drops further. The left side compressor oil level is above the sight glass.

Danfoss scrolls have an 'organ pipe' installed in the oil equalisation ports of each compressor. This is actually to prevent oil being pushed out of an idle compressor into the discharge pipe. There is a video on You Tube that explains this. We have the exact opposite with oil leaving the right side compressor only when both compressor are operating.

The chiiler manufacturer has received video evidence of what is happening. They have responded by stating that it is impossible. The size of compressor that is installed has the 'organ pipe' fitted as standard, so we know this is not missing. With a superheat of 8.0K we are sure it is not saturated vapour or liquid entering the compressor.

The manufacturer has discounted the idea of pressure equalising lines between the sumps as the 'organ pipe' design should take care of this.

Has anyone come across this phenomena before? If so what was the cure?