Welding HDPE pipe and pressure drop

[KevinNZ]

When PE pipe is thermal welded the weld leaves a bead inside the pipe.

How much does this bead effect the pressure drop down the pipe?

We have line (90mm OD 10mm wall) that does not flow any where near the flow it should but the pressure drop calculation did not include the effect of the weld bend.

 

[CRG]

I don't know how the bead affects the pressure drop; however, you might look at it like an orifice. I suspect that it is not significant unless it is a very long line with lots of joints. If it is a buried line, I would look into the possibility that the pipe was deformed from an external load. Was the line kinked from too tight of radius?

How long is the line, and what is the fluid? What was the flow you expected, and what are the upstream and downstream pressures?

 

[KevinNZ]

The line is 1300m long with welds at 12m.

The flow we want is 5.5 l/s DP = 5.5 bar
we get 1 l/s DP = 12 bar.

fluid is water @ 48C

The line is above gound and the run is flat.

thanks
Kevin

 

[CRG]

If the numbers you have given are correct, you have more than the normal weld bead inside the line causing the restricted flow.

1. Try reversing the flow and see if any debris comes out.
2. Run a sizing pig through until it stops; replace the defective segment/segments.

 

[stanier]

Try looking at the manufactuirng standard. You may be surprised by the wall thickness tolerance. Do your calculations again with the minimum bore and you may find that the flow will match the calculation.

Some less competent manufacturers keep to the highest tolerance wall because they cannot control the extruders well enough. They tend to be back yarders who use lots of regrind, have low overheads, old machines and die heads and no technology back up. The additonal resin used is countered by their low caost of operation.

Dont buy PE on price buy on technology. The cost of a failure will soon outweigh the savings you have made.

As far as the weld bead is conserned you could always cut it out. The Daisy Gorseline cutter is one such tool but I believe there are others out there.

 

[hydrae]

I would consider each bead to be a sudden increase in pipe diameter, which is where the minor headloss would be.
He = K*v1^2/2Gc
K= (1- (d1/d2)^2)^2

He = headloss
K = K factor
d1 = effective diameter of pipe at bead
d2 = diameter of pipe
Gc = gravity
v1 = velocity at d1

Hydrae

 

[stanier]

Hydrae, What about pressure recovery downstream of the orifice?

 

[KevinNZ]

Thanks for the input

I did a calculation treating the welds as an orfice plant and got a very small orfice to give the measure drop.

After a bit more site work we found some rocks on the line. all OK now

Kevin

 

[hydrae]

Stanier

The equation is derived as the resultant pressure recovery down stream of the orifice using conservation of momentum and energy. The energy goes into the small eddies downstream of the expansion as heat. Look up in your fluids book of all the minor losses a 'sudden expansion' is the only one determined with an equation, all the other losses have a lookup table for the K value based upon experimental results.

Kevin
Where the rocks on or in the line?

Hydrae

 

[KevinNZ]

The rocks were in the line, but think they were hanging up on the welds as it took a lot of pressure to move them.

Kevin

 

[plasguy]

Some older literature on HDPE used t say that the bead causes the pipe C value to drop to 140 from 150 but it is common tday to have the bead removed after installation to keep the flow at 150. Plasguy

 

[BRIS]

There is a tendency to get a little too hung up on theory -Where do the friction factors such as "C" or "Ks" come from in the first place?. In general they come from tests on existing pipelines and include for the minor loses of beads, bells and sockets etc. You are therefore only looking for the additional loss from the oversized bead.

For the orifice loss it would be reasonable for this simple approximation to assume that the slight increase in velocity is not recovered thus Hl = (Vo - Vp)^2/2g. Vo = V orifice will be Q/(cc.do).
Vp = average pipe velocity.
You would then need to reduce this by a factor (say 50%) to account for the fact that the friction factor is for real pipe and already includes bead losses.

If losses are more than anticipated it usually means that there is an obstruction in the pipe as is the case here.

 

[rconner]

While I realize the promotions/claims/promises? of the plastic pipe industry, I think a Hazen-Williams "C" of 150 may be an EXTREMELY optimistic assumption particularly for very small diameter piping in some practical applications. In this regard I noticed a table in Appendix "C" on page 640 of a recent Haestad/Bentley publication (that may now be available online), "Advanced Water Distribution and Management" now advocates a "C" of 140 and 147 for "Smooth" (including some plastics) pipes in 1 inch and 3 inch diameters, repectively, and a "C" of 134 and 142, respectively in the same diameters for "PVC wavy-clean". It may thus perhaps be adviseable for designers to use realistic or proven by experience values for "C" along with the flow diameters involved, as opposed to perhaps unrealistically high values, for practical application. Of course it should proably also be noted as well that if pipes are used for wastewater applications, many pipe material flow coefficients are reportedly eventually effected by sliming and/or other efects etc. as explained in the prior thread at

http://www.eng-tips.com/viewthread.cfm?qid=73633&page=19.

 

[ytse]

Following is a technical note published by CP Chem, a major manufacturer of HDPE pipe, on this issue:

Technical Note PP 828-TN
Internal Fusion Bead Flow Resistance
For pressure piping applications, tests show that polyethylene pipe with internal fusion beads has a surface roughness of 7 x 10-5 feet, which corresponds to typical surface roughness values for the design of smooth pipes and to a Hazen-Williams C-factor for water at 60°F of 150. AWWA and Factory Mutual Research both recommend a C-factor of 150 for the design of polyethylene water pipe using the Hazen-Williams equation.
When these surface roughness or C-factor values are used in design, the effects of internal fusion beads are automatically factored into the results, and flow resistance due to internal fusion beads does not need to be considered separately.

Hope this helps