Burst Water Hoses

[Route21]

Don't know if anyone else has had the same problem, but we've had a number of toilet/shower hoses burst, where the ferruled end blows out from the backing nut.

The basic reason is water pipes being closed off at 2 points and the water between expanding.

2 typical scenarios:

1. Water flow to shower controlled by wall-mounted valve in feed line from loft tank. Feedline closed. Trip-valve in shower unit automatically closes on low/no flow. Water between expands, hose blows. Cured by leaving feedline valve fully open, unless replacing hose.

2. Feed to pop's bungalow comes direct from pump. Pump has non-return valve in suction, for self-priming reasons. Pop's bungalow vacant for few days. Hose to toilet cistern ruptures. Wall valve isolates ruptured hose. Douche hose alongside ruptures. Wall valve used to isolte ruptured hose.

Has anyone else had similar probs and solved them?

[Litlos]

I think you may be right on the money with thermal expansion being the culprit. Normally there is a pressure relief valve after the non return valve to bleed off excess pressure and prevent this happening. Should be able to get something at a reasonable hardware shop, rule of thumb is take the Maximum Allowable Working Pressure (MAWP or sometimes max press etc) of the system and have the Press relief valve set about 80% or that. Does not really matter in this case as long as it is not running in normal operation and lets a little bit of water go to control the maximum pressure so the flexible lines do not burst.

Cheers

[sirineou]

Perhaps a expansion tank

[Crossy]

Water is essentially incompressible, so any expansion will cause a large increase in the strain on a closed system. The solution is don't have a closed system (bleed valve, expansion tank).

Does the pump have a pressure tank and is that tank still connected to the system when everything is shut down (it would act as an expansion tank)?

Any idea what the pump pressure is set to? (is it high enough to be stressing the hoses before any expansion starts).

It may be enough to kill power to the pump and bleed off the system pressure when vacating the house.

Or the obvious, cheap hoses?

EDIT A bit of research. The volumetric temperature expansion coefficient of water is about 0.000124 1/^oC so a temperature change of 10^oC will cause a volume increase of 0.0021 x original volume, 0.2%.

Is this sufficient to damage a flexible pipe? In an isolated length of pipe, I doubt it. In system where the majority of pipe is PVC (inflexible) with a small length of flexi on the end, possibly (all the extra volume will go to fill the easiest part to expand).

[Route21]

Perhaps a expansion tank

Pump already has an integral expansion tank on the outlet.

[Route21]

The pump, from the ground level tank (which is fed from the mains), feeds the indirect system in the house via 2 loft tanks but feeds the bungalow outlets/shower directly. The outside (Thai) loos are supplied under gravity (no pump required) from the outside tank.

Item 1's problem was, effectively, caused by isolation valves imediately at either end of the short hose.

We previously had the feed to the kitchen blow a joint, late one evening (in the dark) only to find that the only valve to stop the flow was under the kitchen floor at the far end of the house from the access to the underfloor void! That was cured the next day with a valve just inside the underfloor access door.

In item 2's case, the 1st we knew about it was the outside pump running for longer than normal. The line to the bungalow is a fair length and partially above/partially below ground.

Fortunately, we have isolation valves on each water feed that enters any room, so we could isolate it until the hose could be replaced, but the basic problem still exists. I'm fully familiar with thermal relief valves (and non-return valves) but haven't seen any out here in the sticks! Expansion tank(s) may be a solution, but the small one integral with the pump didn't cope.

Another related problem is that we have very hard water here, so cistern valves, shower heads and any potential relief valve/non-return valve is likely to fur up in a short space of time.

I would have liked to have used a non-return valve, as a backup, to enable us to easily convert to a pressurised direct feed from the pump in case there were any problems with the loft tanks that needed their temporary isolation. I wouldn't rely on someone inexperienced in M&E services to set the valves correctly and we could end up with the loft tanks overflowing and wrecking the upstairs ceiling - although we do have drainage from the "tray" that the loft tanks sit in.

Any more thoughts?

Regards

R21

Edited March 3, 2013 by Route21

[Route21]

One thought l had was to isolate the bungalow at the pump end of the supply and leave one of the taps on when no one's there, but it relies on aging parents/kids that may switch it back on remembering that the kitchen tap is open. Not the best of solutions?

[Crossy]

Does the bungalow have a flush toilet? If so turn off the water and flush the loo, no tap to forget just turn on the water and you're good to go.

[Litlos]

If relief valves are out of the question then how about the old style relief of a shepherds crook? Basically this is a line connected to the water line and open to atmosphere at the other end ,(20mm PVC) taken to such a height as to not leak under normal pressure, but bleed off under excess pressure situations. If you do not know the heights required have a look at Convert from Josh Madison which provides conversion from PSI/kPa to inches/mm of water.

Another off the wall solution is to install a stand pipe as an expansion tank. All this is is about 2 metres of 40mm pipe run vertically from the supply line with a cap and valve at the top. Ensure it is full of air and close the valve. Works by giving expansion room.

Crossy, almost perfect on the calculations, but the first thing you say is water is inincompressible As such if not relieved pressure will build in a closed system until something gives, in this case the bum gun line etc. I have seen systems locked in go from 200 to 600kPa in a very short time on a warm day exposed to sunlight.

Cheers

[Route21]

Does the bungalow have a flush toilet? If so turn off the water and flush the loo, no tap to forget just turn on the water and you're good to go.

Excellent suggestion, Crossy.

Many thanks

R21

[Route21]

If relief valves are out of the question then how about the old style relief of a shepherds crook?

Another off the wall solution is to install a stand pipe as an expansion tank. All this is is about 2 metres of 40mm pipe run vertically from the supply line with a cap and valve at the top. Ensure it is full of air and close the valve.

Many thanks for your 2 suggestions.

I quite like the low tech solution in #2 and may use it in situations where Crossy's empty cistern solution doesn't apply.

#1 doesn't work for the bungalow in my case, as the pump also supplies water to the loft tanks in the house and any shepherd's crook would therefore need to be higher than the house roof.

Again, many thanks for your suggestions.

Best regards

R21

[sirineou]

a properly installed. properly charged expansion tank should take care of most expansion in a system.

Have you checked the PSI charge on your expansion tank? if the psi charge in the tank is higher than the pressure required to blow the hose connection, then the hose connection is the week point rather than the bladder in the expansion tank , and will b;ow first. before the b;adder has a chance to expand/ Also it is possible that the bladder in the expansion tank is perforated and now both chambers in the tank are filled with water, check the air valve in the expansion tank , if water comes out of it , then you have a perforated bladder.

Final thought, is there a check valve (a valve that will allow flow on one direction only) installed between the expansion tank and the hose in question to prevent back flow? if so, such check valve will isolate the hose in question from the expansion tank.

[sirineou]

one more thought on the subject,

some spigots, have a build in check valve to prevent what ever is connected to the spigot from back feeding in to the house, if such spigot is used as an isolation calve then the portion of the system that is connected to that spigot is isolated from the rest of the system on that direction, and the expansion tank that is installed before the spigot will not protect that portion of the system from surge or thermal expansion.

That's why you some times see garden hoses burst when left in the hot sun.

Edited March 3, 2013 by sirineou

[Route21]

Background

The pump/expansion tank works perfectly correctly feeding the loft tanks in the house via ballcocks. It switches on when there is a demand and off when the demand ceases. Water is in regular use somewhere in the house during daylight hours.

The bungalow outlets are fed directly from the pump, without using loft header tanks. The bungalow was not occupied for a few days and hence the water outlets, which would have minimised any thermal expansion problems, during the daylight hours, were not used.

"Have you checked the PSI charge on your expansion tank?"

Don't understand your question.

There are 2 pressure switches on the pump dicharge. One shuts down the pump on a specific "high" pressure in the expansion tank. As the pressure in the expansion tank falls, when water is used at one or more of the outlets, the second pressure switch turns on the pump to meet the demand. The pump continues to operate until the first pressure switch shuts it down again. That's the way they work i.e. the tank pressure fluctuates between 2 set points depending on the demand. There's no single "psi charge" figure.

"if the psi charge in the tank is higher than the pressure required to blow the hose connection ...."

then the hose connection would have blown 6 months ago when the system was first installed!

"Also it is possible that the bladder in the expansion tank is perforated and now both chambers in the tank are filled with water, check the air valve in the expansion tank , if water comes out of it , then you have a perforated bladder."

The top section of the expansion tank is filled with air. Only when all the air has been vented and the pump was operating would any possible water appear and only if any bladder has been perforated. If no water appears, you have just disabled a working expansion system. As was explained by a previous poster, even a simple air column tee'd off from the pump discharge line (without any bladder) would protect the hose from simple thermal expansion.

If there was no air buffer in the expansion tank, the pump would be constantly tripping in and out, when, as soon as it switched on, it would hit the shutoff pressure and shut down. On shutting down, it would trigger the low pressure switch that would turn it on again.

"Final thought, is there a check valve (a valve that will allow flow on one direction only) installed between the expansion tank and the hose in question to prevent back flow?"

As explained in my original post, the only check valve in the system is on the pump suction, to enable the pump to maintain prime. It is upstream of the discharge expansion tank and so does not isolate the hose from the expansion tank. The suction NRV does, however, effectively act as an upstream isolation valve which, when combined with a closed cistern ballcock (full cistern) and closed taps/douche valve, provides no room for thermal expansion.

I hope this more clearly explains the hydraulics of automatic water pump systems such as the off-the-shelf one that we use.

Regards

R21

PS The "empty cistern" and "bladderless air expansion pipe" suggestions appear to be excellent practical ideas.

[sirineou]

an expansion tank, Aside from providing positive pressure in a closed system, also provides surge and thermal expansion relief ,

the positive pressure is determined by the amount of charge in PSI (Pounds per Square Inch)of the air chamber of the expansion tank.

If the air in the tank is set too high then it will take a lot of back pressure to compress it, and rather than the air being compressed the pipe could - burst. or you could get pipe banging when the pump kicks in, or when there is thermal expansion.

I don't know the system you have , but if I was a betting men , I would bet that if you have pipes bursting, unless there is something wrong with the pipes the problem resides in either the expansion tank set up,or the low and high setting of your pump.

again I am not there to survey your system, and I could be wrong, but... it is quite possible that the offending bursting pipes are not protected from expansion , because there is some sort of a check valve in the way, that would allow pressure to travel one way to the pipe, but no way for excess pressure to return to the expansion tank. Some spigot valves have such check valves built in them, to protect from contaminants back feeding in to the system,such as spaying fertilizers in your garden. There could be other devises that act as a check valve between the bursting pipe and the expansion tank.

Remember, it is called an expansion tank and not a positive pressure tank for a reason.

At least that is the way I understand it,

I hope my input helps

[Route21]

Please read both my post and those of others.

Background

The pump/expansion tank works perfectly correctly feeding the loft tanks in the house via ballcocks. It switches on when there is a demand and off when the demand ceases. Water is in regular use somewhere in the house during daylight hours.

There is nothing, repeat nothing, wrong with the pump or the expansion tank system - full stop! As already advised, it works perfectly correctly on the (below ground and hence not exposed to daytime sunlight) supply to the house.

The problem only arises on the direct from the pump, partially above ground and hence exposed to daytime sunlight, supply to the bungalow and even then only when the bungalow outlets are not used during daylight hours.

There are no, repeat no, NRV's of any kind on the supply to the bungalow (except, as already advised, on the suction side of the pump which works perfectly correctly for the house supply).

I regard this topic as now closed, with many thanks again to those who have provided constructive suggestions.

Regards

R21