Pressure/PRV Issue

[utilengr]

I work for a utility and we have been having a long term issue associated with a customer's pressure. The customer has had some leaks, one of which resulted in the loss of a significant amount of water. There are a number of issues, including the customer's internal piping which is a mixture of materials including Qest. We've notified the customer of this and have been working with him - initially trying to provide the information to him to allow him to install and maintain his own PRV properly. Other issues have come up and we've gotten more involved as we are being accused of being the source of the issues. The customer is served by a small diameter line (likely a 1" galvanized line) that we intend to replace as part of our ongoing line replacement program. In the meantime, I have been unable to make sense out of what is happening at this customer location. There 3 other customers on this line, this customer is at the end of the line and is the only one having issues (that we know of).

The description of the set up at this location is as follows:
System (105-110 ave. press.)-->Customer Meter-->PRV1 (Wilkins)-->Tee for Press. Recorder 1-->Check valve (inline ball check)-->Lateral to house-->PRV2 (Watts)-->House plumbing-->Press. Recorder 2

The customer watches his internal gauges constantly and calls whenever the pressure is above 70 psi. We have several charts showing something that I am not able to explain. I have attached one. The pressure routinely climbs above the setpoint of the PRVs. Initially it appears that is could be thermal expansion and we added a check valve to the line in the anticipation that this would provide the data needed to demonstrate this or that something internal to the home is adding energy to the system. However, the data following the installation of the check valve (between the 2 PRVs) shows that they both increased in pressure and continued to track along with each other. The check valve is new and I guess there is a chance that it may not be working properly but it didn't seem probable. We have adjusted the PRVs and they continue to allow pressure to go above the setpoint. As a result, we continued to receive calls. The PRVs are direct acting 3/4" valves - both are relatively new.

To try to keep the pressure from being seen on customer side we also installed a pressure relief valve after the first PRV that would keep the pressure at 65 psi. This worked for a few weeks but we are now seeing that the PRV upstream from the blow off will no longer maintain a pressure setpoint - we make adjustments (while flowing a small volume of water) but as soon as we stop the flow, the pressure climbs over the setpoint and up to the relief valve setpoint (10 psi higher than the PRV setpoint), which will then run/drip almost continuously.

Any thoughts/suggestions on this would be appreciated.

 

[itsmoked]

It's not clear to me but are there other customers after the 'customer meter'?

If so one of them could have a recirculating Reverse Osmosis system - run amok.

Could you be delivering really cold water that is expanding after the PRV?

What is causing the periodic pressure pulses (about an hour apart) in the entire area in discussion?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[utilengr]

There are no other customers after the "customer meter" in the system description above.

I don't believe they have an RO system but we can check this (there wouldn't be a need for one but that doesn't mean there isn't one). I know there is a geothermal system in the house - we've asked more than once about a possible interconnection and have been told repeatedly that the geothermal system is a closed system with no interconnection to the house plumbing. Even there was a connection or an RO recirc. issue, the check valve we installed should be catching this and the pressures on the 2 recorders should have been different. Something is causing the direct acting PRVs to operate outside of their range. Or I don't fully understand what could cause the PRVs to consistently move outside of their setpoints - I understand that this could happen if flow is going through the valve (the valve is open) and the pressure increases from the supply side but I'm having difficult understanding what could cause this when there is no flow (valve is seated/closed) on 2 PRVs separated by a check valve.

The water temp. should be fairly consistent with the ground temp. There is no seasonal variation in the complaints/calls.

The periodic pressure changes are associated with our booster station for this zone.

 

[utilengr]

The attached made me suspicious that something was generating pressure inside the house. Unfortunately, I didn't have the flow monitoring on the meter yet. The check valve was installed at this time but the data didn't remain consistent.

 

[bimr]

Several comments:

1. This is one of the reasons that almost all utilities require a separate service to each house or business.

2. Why is the customer watching pressure gauges? Very unusual behavior.

3. Most residential plumbing fixtures should be able to operate 110 psig although the service life will be lessened.

4. Is it possible there is a water heater malfunctioning?

5. Is there an elevation difference from the service to the house?

6. Is there a booster pump at the residence?

7. Why is this person using water in the middle of the night?

 

[itsmoked]

OK, thanks on the periodic.

Just to be clear is this exactly what the setup is?

Seems odd the house pressure is higher than the post meter pressure.

PRVs require working pressure to drive their functioning. If the after-meter PRV is set at or below the house PRV the functioning of the house PRV is undefined.

The check valve does seem to make having the house raise the pressure back to the meter less likely. Did you have the same result without the check valve?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[cvg]

is there some sort of illicit connection that the customer has installed which might bypass the prv or the meter? (that you are unaware of...)

 

[BigInch]

A better resolution of time on the chart would be nice. Can you plot a chart of only one day or less?

The relief valve is set at 65 psig.
Pressure reduction valves are set at 55 psig. Both PRVs have the same set point, right? I would tend to set PRV1 to 60 psig, just below the setting of the relief valve.

As it is now it looks like periods of high flow reduce supply pressure quite a lot and the supply pressure gradually increases with the mainline's delivery of more and more flow.

Eventually the supply pressure reaches a value near the PRV setting of 55, perhaps with a reduced flow at the house as well, maybe not. If the PRV is at 55 psig and the house is flowing, any reduction at all of the flow in the line will probably kick off the relief valve. The reduction of flow might be caused by the home owner, or the PRV, as it starts to close in its attempt to limit any higher pressure going to the house. If the flow capacity of the relief valve at its set pressure of 65 psig is small, the supply line pressure might increase further, but PRV1 might be then close, allowing only a small amount of flow through the relief valve while that section between PRV1 & 2 bleeds off a bit.
If the flow capacity of the relief valve at its set pressure of 65 psig is large, the supply line pressure could reduce significantly below the setting of PRV1, thereby opening it fully and even causing the mainline boosters to come on and the supply line pressure to increase too much. As you can see there might be a considerable amount of dynamics that is going on which we cannot really see because of the congested time resolution. Large pipe diameters would tend to reduce the dynamics, long lengths increase it and we don't know what pipe diameters are involved in the supply.

 

[utilengr]

1. This is one of the reasons that almost all utilities require a separate service to each house or business.
- Agreed. This is a separate connection but on a old service line (probably installed in the 60s or 70s).

2. Why is the customer watching pressure gauges? Very unusual behavior.
- Agreed. There is an issue here.

3. Most residential plumbing fixtures should be able to operate 110 psig although the service life will be lessened.
- Agreed. However, the Qest piping is a problem that he should deal with but has thus far refused. I have been involved in one other instance where the homeowners insurance company has gotten involved to remove this pipe. He hasn't pursued this.

4. Is it possible there is a water heater malfunctioning?
- I don't think so but don't know for sure. He has a small expansion tank mounted above it.

5. Is there an elevation difference from the service to the house?
- There is some elevation change between the meter location and house - this difference is picked up by and can be seen in our recorder data.

6. Is there a booster pump at the residence?
- There is a circulation pump on the geothermal system but we've been told there is no interconnection.

7. Why is this person using water in the middle of the night?
- Not sure about this one.

 

[utilengr]

@itsmoked:

The drawing you have is close - there is no tee and additional line running to the question mark. The Watts PRV is located inside the house but schematically, what you show is good.

It is difficult to know what is going on with the in-house PRV. I have no confidence that this one isn't being adjusted by the customer - he has a post-it note on the wall showing direction of rotation for raising and lowering the pressure. The one at the meter is the only one that I have some general confidence that it isn't being tampered with.

I'll look later today for data that is pre-check valve.

 

[utilengr]

@cvg

No illicit connection that I am aware of - he had the entire lateral replaced about 5 years ago when he had a leak between the house and our connection. We know the contractor and I don't have any reason to believe that he would have done anything like this.

 

[utilengr]

@BigInch

I'll look at the spreadsheets later today and see if I can get a better resolution graph posted.

You are correct. The relief valve is set to about 65-66 psi and PRV1 (at the meter) is set at approx. 55 psi. This arrangement worked for a while but now it seems that something isn't holding. The relief valve is constantly running a little now (fast drip). If we lower the setpoint on the PRV to say 45, the relief valve stops flowing but as soon as flow stops (we set the valve while flowing a little water) the pressure exceeds the setpoint of both the PRV and the relief valve to the point that the relief valve constantly drips.

The supply line is definitely a problem - it is old, long (for the diameter), and small. I haven't ruled out the possibilities of this line contributing to the issues. The line is approx. 400 ft. long to the customer meter. This line connects to a larger 8" main that shouldn't be a factor - supplied by a 500,000 gallon tank that is periodically refilled by a booster station as the level falls.

We're looking to replace the line but this isn't a fast process. There are 2 options - replace this customer's service line with a separate, new line or replace the line for everyone with a new main (we have a possibility of looping this line to an adjacent subdivision, which would eliminate the dead end issue as well). I prefer the second option but this is a larger scope. I'd also like to have confidence that I would resolve the problem with either of these..... I also haven't ruled out the possibility of something going on inside the home that even the customer isn't aware of.

If you look at the August graph, this made me think that there was definitely something going on in the house the pressure went up on the house recorder while the pressure went down at the meter. However, the September graph shows that this didn't continue. Debris from the old line getting in the check valve seat?

 

[rconner]

As to <<<...I have been involved in one other instance where the homeowners insurance company has gotten involved to remove this pipe.>>> I am curious, are you saying the insurance company is paying for damage and even that re-pipe? If so, sounds like good insurance!

[Of course some paid-for pipe replacements (result of that particular/prior huge plastic pipe lawsuit) played out several years ago, but I am aware the failures continued long after the industry was aware of problem and to date. The decided remunerations eventually ended after only a few years, though many homeowners are now aware they were handed problematic pipe (probably reason Customer is aggravated and watching gauges etc.)]

 

[BigInch]

Some flow data would help as well.

 

[hydrae]

Utilengr

I see 2 possible causes

You mention there is a bladder tank on top of the water heater,
has this device been checked to see if is not flooded. The bladders do fail and eventually the tank fills up with water providing no protection. The water inside the heater expands upon heating which in homes without PRV or check valves a small amount of water is vented back to the system.

Is it possible there are bits of rust from the service line breaking loose and lodging inside the PRV's causing the PRV's to bleed through? have the PRV's been disassembled and inspected to verify they are working properly?

My bet is on the first one.

Hydrae

 

[itsmoked]

I can see hydrae's flooded expansion tank scenario fitting this well. A fast drip is exactly what you'd get on reheating water in a closed system, (check valve). Next, his water heater will fail and he'll probably blame it on the 'high pressure'.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[coloeng]

Utilengr,

I would agree with Hydrae on both points. The spike at 11:53 PM looks like something got stuck in the meter PRV and kept it from shutting tightly. The long period that it took the pressure to drop back down and the usage from 11:53 to 3:00 AM might be explained by the PRV being stuck open and the pressure relief discharging the excess pressure. The constant drip is probably from the expansion tank not operating properly.

If this is indeed the case, the only true solution in my opinion would be to replace the 1" galvanized iron line that serves this and the other properties. Once that is done, the issue of the PRV not functioning properly should be resolved. Once you can resolve the PRV at the meter not working properly, the remainder of the problem is the customers. Although it is commendable that you are trying to help the customer solve his issues, you sometimes have to step back and remember that you are only responsible for the system to a certain point. Don't take any actions that will make you or the utility responsible for the customers system.

Good Luck

 

[LittleInch]

If the PRV leaks only a very small amount I was wondering if wouldn't be easier to install a small expansion tank downstream of your PRV to absorb the PRV leakage issue.

Looks to me like the PRVs are not sealing 100% any more, but a potentially malfunctioning bladder sounds like a good place to start in the house itself.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[cvg]

I would second coloeng, replacement of 400 feet of 50-year old, 1 inch underground galvanized pipe should be priority one. I wonder how much flow you can actually get through this old, corroded, undersized line.

 

[itsmoked]

It's simply the expansion tank pure and simple not the PRVs.

Reasoning: Notice the time it takes to pressurize? What's the typical time it takes a water heater to re-heat? 10 to 20 minutes. The rises take 12 minutes.

Notice the periodicity of the these rises?
About as frequently as a water heater cycles in a typically un-busy home.

Notice the pressure rise is always with the same rate-of-rise.
Sometimes it doesn't reach clear up to peak pressure seen because during the water heat up water is used in the home; a toilet flushed, hands washed, washing machine uses again, etc.

Why would a PRV only occasionally leak? Virtually every problematic water device I've seen from PRVs, valves, check-valves, water heaters, facets, etc, etc, down to a hose spray nozzle, if they have a problem they almost always keep the problem or the problem grows with time.

If somehow the PRV was faulty and leaking why would it leak at the same exact rate? If debris were sticking in it they'd be different sizes causing different leak rates.

It's the blocked system having a tank filling a volume considerably larger than the rest of the plumbing then heating that volume 60 or 70F°.

Keith Cress
kcress -

http://www.flaminsystems.com