Recently installed water heater included an expansion tank per city code. Water meter flow indicator started small intermittent flow indications every 5 to 10 seconds, equivalent to 1 to 2 gallons per hr. Problem disappeared when expansion tank was isolated with shut off valve. Water Authority told us we had a probable leak and to get plumber who subsequently recommended re-piping house as he believed it was an under slab leak. $8000 later problem still exists.
Water Authority come out and calibrate water meter and find no problem but change backflow valve attached to meter as a "precaution" and instantly problem disappears and hasn't returned.

I am arguing that a leaky backflow valve is the source of problem ie when mains pressure drops slightly the higher pressure in expansion tank creates a small leakage through backflow valve, which does not register on flow indicator. However, when mains pressure recovers the expansion tank is re-filled quickly and the quantity of water is registered on meter.

Water Authority claim they tested valve and found no issues, I believe their test was conducted with the expansion tank providing 50 to 60 psi differential pressure across the valve which helps to fully seal it. Under the problem situation there would be a very small differential pressure across the valve which could be resisted by a "sticky" valve and thereby allow the small leakage.

Any thoughts or info would be appreciated.

what is your street pressure and house pressure?

Would there be a thermal expansion pressure increase, if the check valve was bleeding back to the street?

Your plumber was incompetent. He didn't properly test for a slab leak, which would have taken him a few minutes to do.

Yeah, an $8000 expenditure on a guess was quite a boo boo.

mains pressure is typically 60 psi +or- 1 psi. House pressure is the same except for when there is a sudden drop in mains pressure.

Thermal expansion is unlikely to be involved especially if the backflow valve is leaking as that would relieve any increase in pressure. The mains pressure fluctuations create the differential pressure across the backflow valve.

I have a similar situation that I suspect is backflow caused by a newly installed expansion tank between the utility and my DHW tank.

I’ve been chasing a .01 - .02 GPM “leak” less than but similar to Alf’s “leak”. It is recorded by my Flume water flow meter installed at the utility meter but not detected by the utility meter itself. This minimum flow is always present except for the occasional hour or two in the dead of night. The “leak” appeared sometime after the expansion tank was installed.

I can instantly stop the Flume from detecting the “leak” by closing the supply valve before the expansion tank, and the “leak” resumes the moment I reopen the valve. I am confident there is no water leakage anywhere after the expansion tank.

I have also noticed irregular, tiny bi-directional flow from the spinning flow indicator at the utility meter. It sporadically and momentarily reverses direction for a quarter-rotation throughout the day.when there is no water demand. I can prevent this from happening by closing the same supply valve I described above. When closed, Flume reports zero flow and the flow indicator doesn’t move in either direction.

Perhaps the Flume is reporting only the positive flow portion of a bi-directional flow caused by HW expansion tank pressure changes? This could explain why the Flume sees a leak and over-reports water use, which could be as much as 15-20 gallons per day. My service is 80psi and has no water pressure regulator or backflow valve that I can find.