Air Header Flow drops to zero when compressor unloads

[rcchap]

I am replacing most of the major components on an Instrument Air system for a two unit power plant. Each unit has two compressors, one receiver (137 cu ft), and one internally heated desiccant dryer. The systems are crosstied and a single compressor is normally used to provide air for both units. The crosstie connection is downstream of the compressors and upstream of the receiver tanks. The normal demand for each unit is between 200-300 SCFM.

The existing air compressors are Atlas Copco ZT3 rotary screw oil free compressors. The new ones will be similar (but newer models). The existing compressors have a capacity of 750 ACFM, and I have observed that the LEAD compressor will run loaded for approximately 40 seconds, then will unload for about 15 seconds before loading again.

Each unit's header has an ERDCO Armor-Flo 3600 series mechanical flow indicator. This flow indicator is downstream of the dryer, and there is one for each unit's supply header.

I have noticed that when the compressor unloads, the header flow indication goes to zero. Since the air demand is constant, I can't come up with a logical explanation for this.

Has anyone seen this before or have a plausible explanation?

 

[DLite30]

When the compressor is not loaded, the air at opposite end of a header from the outlet or inlet will have a velocity that is close to zero.

Imagine laying a slinky toy on the ground and pulling one end...looks kind of like a worm, where one ends starts to move, then the middle, then the end. Same idea with air flow in a long header.

 

[rcchap]

That sounds like a possible explanation.

But I can't get around the fact that the receiver tank (which is upstream of this flow indicator) is depressurizing from 110 psig to 104 psig in 15 seconds, and as this tank depressurizes there is about 200 SCFM of air flowing in the direction of the flowmeter to the header.

 

[rzrbk]

Depending on the size of the components, your total header volume may be equal-to or more than your receiver volume. If there are users upstream of the meter (cracked open bleeders, dryer purge, leaks, etc), then the receiver may supply these, while the expanding header supplies the units. This could result in zero or reverse flow thru the meter(s).