I really don't think there is any preference as long as you think about the thrust developed and design for the correct transmission of the forces which will be generated in service, after all in really large mains pipe surges (water hammer) will in any event possibly reverse the thrust during transients.
30 years ago, when I first graduated as an engineer, we were on reservoir refurbishment job working to replace 3ft diameter 1939 vintage purge valves, which we had sealed on the 100 ft deep upstream reservoir side using divers to insert purpose made plugs, as we could not drain the entire reservoir just to repace the originals. These valves held back the thrust from the reservoir side in tension on the bolts. We were much alarmed before we took the pressure of the upstream side, to find that a number of the bolts on the stuffing boxes had failed completely and could be lifted out by hand! Perhaps the tension loaded bolts on the valve flange had been checked more assiduously over the years of use, but we never quite knew how close these big valves had been to blowing off due to flange bolt failure, which would have caused reservoir failure and a major disaster!
Since then, when a flange obstinately leaks I was always very reluctant to allow bolts to be overtightened, worrying that this overtensioning may, over time, cause bolt failure.
I was brought up in a design office (30 years ago) where we always anchored the valve on the upstream side so that the anchor (thrust pipe) flange and valve body would be in tension. I was never sure why but it is a practice I have stuck with
I assume that your pipeline is buried & the valve is in a valve pit with a flexible joint.
The design codes for sluice valves sometimes specify 'open end test' (valve on the end of a pipe in a terminal position) or 'closed end test' (valve is going to be located in a pipeline with pipe on both sides of the valve & pit). The 'open end test' valve is tested with the valve bolted to the test equipment with the gate closed & hence the body is subject to tensile forces during the gate water tightness test. This is the most stringent test. The 'closed end test' valve is placed in the testing machine with both flanges sealed by applying a compressive force.
An 'open test valve' can then be located in a valve pit with no flexible joint or any form of anchor because forces can be transferred safely by compression or tension. I have not used a flexible joint or an anchor for valves in valve pits (valves up to 1,200 mm diameter & nearly 300 m head) for over 20 years.
Constuction sequence is: -
* Pipe is laid to one side of the proposed valve pit
* Reducers bolted onto sluice valve
* Sluice valve & reducers lowered into position & supported (temporary - for construction)
* Reducer welded to pipe
* Next pipe laid & reducer welded on
* Temporary support removed
* Valve pit cast (last of all)
* Backfill around pipe & pit
* Permanent valve support constructed off pit floor
In Australia, most steel pipes are laid with rubber ring joints & consequently, the skin friction (anchorage resistance) from the pipe/soil interface of the pipes, must be assessed on both sides of the pit.
Brian (BRIS)
There is a culture with water utilities, that a pit is required for a valve (usually non negotiable). If you step back from the problem & look at the physical requirements from first principle, consider the following: -
* Flexible joints are really dismantling joints, to allow removal of the valve. If valves are only rarely removed, the cost of all of the dismantling joints not used in their lifetime, will more than compensate the cost of cutting a valve out of the pipeline. The only problem is that a dismantling joint is paid for at the time of construction & the cutting out of the valve later in the pipeline life, is paid for by the operations section.
* Keep reducers out of the pit & only have the pipe (equivalent to the valve diameter) in the pit, to keep the pit as small as practical.
* Consider two short spigots with two 150 mm valves on the bypass, that allows the 150 pipework to be replaced without closing the pipeline down (preserve the operation of the main pipeline). Bypasses can have the cement mortar lining stripped during bypass flow, to fill the downstream side of the valve (it depends on how often the bypass is used at full flow - the head loss across the 150 valve is very large resulting in extremely uncontrolled & high flows). Removal of the bends & the short section of 150 pipe is relatively simple with no effect on the integrity of the main pipe flow.
* Chambers are required to protect & maintain the gearing.
* This means that if you only want to operate the gearing, then there is no need for a full depth pit with a floor (see comment below on ground water). I have only on very rare occasions, been able to convince a utility that a 'dirt floor' is sufficient (utility culture).
* A half depth pit is adequate even for the bypass valves. Most bypass valves are about 150 mm diameter & are stock water utility 'buried service valves.'
* I once worked for a major water utility that had a large number of mainline valves from 600 mm to 900 mm diameter (up to 1500 mm pipes). Over a period of 20 years, very few mainline valves were removed & replaced, so it was decided to weld the valves in solid. If a valve is required to be removed, a 'gas axe' (oxycut) will do the job very quickly, especially compared to the time required to empty & refill the pipeline. Two cuts are required close together on one side of the pipe in the pit, remove the short ring, unbolt the valve & remove. A closing band is placed on the pipe, the new valve has the flanged spigot bolted on, lowered into position & bolted on to the flange left in position. The band is pulled back over the short gap & two temporary brackets welded on the band, a bolt used between the two brackets & tightened, the band welded on & a gouging rod used to burn off the temporary brackets. Usually a manhole has been located close to a major valve, allowing reinstatement of the lining.
Brian (BRIS)
I forgot to mention. If the valve does not have gearing, then I agree with you that no pit is required. This is the same situation as most of the valves used by utilities (150 to about 600 mm diameter). Gearing is mostly used because of high pressure situations (requiring high unseating loads) & in almost all cases requires a pit. If the valves are in low pressure pipelines, ungeared valves (manufacturers should be consulted) can be used up to 600 diameter (900 pipelines).
Gearing may also be required to artifically increase the time to close a valve & prevent water hammer, especially in very long pipelines (when used by untrained staff).
Barry - I worked for a UK water utility in the 1960's. All of the valves were then direct buried. One of my first jobs was to design and supervise the construction of valve chambers around some of the larger cros-over valves.
The following year I was asked by the operations department to demolish the chambers - their view was it was much easier to exacavate and cut out the valve than to work in a chamber.
Nowadays I never see designs with direct buried valves. In fact I have recently seen designs were even the tees are shown installed in chambers!!..
Barry – I am in general agreement with your comments - I worked for a UK water utility in the 1960's. All of the valves were then direct buried. One of my first jobs was to design and supervise the construction of valve chambers around some of the larger cross-over valves.
The following year I was asked by the operations department to demolish the chambers - their view was it was much easier to excavate and cut out the valve than to work in a chamber.
Nowadays I never see designs with direct buried valves. In fact I have recently seen designs were even the tees are shown installed in chambers!!..
Now that the trend is for utilities to be 'corporatised' there is a new group of operations people who are going to the 'buried service' valves for increasing larger diameters. As I said in my previous thread, almost every valve in the reticulation is a 'buried service' valve - so where is the cut off point? 150 dia is OK, 200 is OK, 250 is OK, etc. What size is not OK & why?
It really depends on practical problems of the high unseating loads on sluice valves (high retic pressures - but this is discouraged by operating & design staff because of the economics of high pressure pipes vs pressure reducing valves or tanks, & higher unaccounted for water or its corect term, leakage), & possible large & unacceptable variations of retic pressures leading to consumer complaints - hence the move towards butterfly valves. Lower unseating & operating loads (the water load on the gate is theoretically balanced) & cheaper valves are driving the change. 'Robust' sluice valves were a belt & braces solution but now we have the experience of other (acceptable?) solutions.
There appears to be a history for long operating lives of butterfly valves (especially in dam outlets & controls) but generally the only caution for larger butterfly valves in reticulation, is the short times to open & close (assuming operation by unskilled staff) causing pressure transients in long pipelines or damage to reflux valves in pumping installations.
I am the one that posed the original question. We are based in South Africa but we do all our work in Botswana where the practise is to put all valves in concrete chambers/pits. This applies to air release valves, scour/washout valves and in-line isolation valves. In fact, we have on several contracts, combined the scour and isolating valve chambers (this sometimes cuts the number of chambers required in half at a considerable saving to the client). We usually have steel, PVC or GRP main lines coming into the chamber. We build a double flanged thrust pipe into the chamber wall on the incoming side. The plain ended mainline pipe is joined to the thrust pipe by means of a Viking Johnson type flanged adapter. Inside the chamber, an all flanged scour tee is bolted to the thrust pipe and the isolating valve to the other side of the tee. A VJ flanged adapter goes onto the other side of the valve to take the plain ended main pipe out of the chamber. The VJ flanged adapter provides a flexible joint to allow removal of the valve. On the branch of the tee (100mm dia up to 400 main line and 150mm dia up to 600mm man line)the configuration is as follows: scour valve, thrust pipe through chamber wall and a flap valve on the end of the puddle pipe outside the chamber. The flap valve lifts under the pressure of the water and drops again to close off the thrust pipe to keep out vermin etc. Sometimes the washout pipe is taken to ground level and sometimes it goes into another chamber. Any thoughts on this configuration?
Wayne.
My only comment is why do you build a double flange thrust pipe into the wall instead of a flanged spigot pipe?. (I assume it is because you need to change material and changing at a flange is easier than using a stepped coupling). The solution gives you a buried flange which is not a good idea because flanges leak. A stepped coupling between two spigots would be a better solution it would also allow some flexibility.
Otherwise your arrangement is as used by 90% of designers that I have seen in recent years. (Personally I would use ductile iron bolted gland fittings rather than VJ couplings and adaptors. VJ have done excellent marketing over the last 20 years and now their fittings are used as standard - I think I am the only one left not using them).
On the last two contracts we have actually switched to what we call a "Ranger" VJ coupling instead of the flanged VJ for the very reason you mentioned. For instance, a 300 NB Ranger coupling can join pipes with outside diameters that vary from 315mm to 332mm. The most common joint is PVC 315mm to steel 324mm, so we no longer use a stepped coupling for 300 NB pipes. The rubber ring in the coupling takes up the different OD's and the coupling is torqued up to about 90 Nm so it seals on both materials without a problem.
To Iputpipe
I designed a pipeline in Botswana several years ago. A 1200 pumping main from a dam to a mining town. I didn't see the final construction but I was warned to put air valves & scour valves in pits (with locked covers) to reduce theft of water. I have designed many above ground pipelines in Western countries with exposed air valves with no problems. Air valves on buried pipelines usually require a pit. I usually recommend a 'standard' circular pit that is readily available - say 1200 dia X 600 deep, or a depth to accommodate the sluice (or butterfly) valve & an air valve.
I usually design the scour, with a short & horizontal 150 dia pipe into a vertical, circular pit (from standard well or septic tank liners). The pipe terminates in the pit with a 150 sluice valve. An extension rod is supported off the side of the pit to operate the SV from the surface. The cover is lifted, the valve opened & the water allowed to rise in the pit & pour over the side. The last part of the water can be removed with a simple pump operating from the surface with the suction hose positioned to the bottom of the pit. If there is a reason to pump without using a loose suction hose, the scour pipe can be taken to the surface & terminated with a flange just below the ground surface (to allow a pump suction to be bolted on direct).
In Botswana, the utility required a second pit - I cannot see the requirement for this. Their reason was that the sluice valve was always in 'the dry' but for the few times a scour is operated, an occasional 'drowning' should not be a problem.
