Electrofusion HDPE Fittings: How to Select the Right One

Choosing electrofusion HDPE fittings looks like a catalogue exercise until the first joint fails a pressure test. Search the term and you land almost entirely on product grids — couplers at 230 psi, reducers by the inch, tapping tees by outlet size — with almost nothing that tells you which of those parts belongs on your pipe, or what has to be true about the pipe before any of them will fuse. The part number is the easy half of the decision. The half that decides whether the joint holds is the match between the fitting, the pipe you actually have on site, and the tolerances the fitting silently assumes.

When electrofusion earns its cost premium

An electrofusion fitting costs several times what a compression coupler of the same diameter costs, and it needs a control box, a generator, a peeling tool and clamps before it will do anything at all. That price gap is the first thing a buyer notices and the last thing a catalogue explains. The honest answer is that electrofusion is not a general-purpose jointing method you default to; it is the method you choose when the alternatives are blocked by something physical.

The blocking condition is usually space. Butt fusion needs enough clear length on both sides of the joint to bring a machine in, clamp two pipe ends and hold them in alignment while they cool. In a congested trench, against an existing main, on a repair where you cannot move either pipe end, or on a branch connection where there is no pipe end at all, that clearance does not exist.

An electrofusion coupler works entirely from the outside of the pipe and needs only enough room to slide over the ends, which is why it dominates repairs, tie-ins and service connections. The trade-off is that electrofusion pushes the entire quality burden onto surface preparation, where butt fusion pushes it onto machine settings.

The second reason is far less discussed and matters more to procurement than to installers. Electrofusion tolerates a mismatch that butt fusion does not. Butt fusion requires the two pipe ends to have the same outside diameter and the same wall thickness, so joining a PE80 SDR11 pipe to a PE100 SDR17 pipe end-to-end is not a normal operation. Electrofusion fittings work on the outside surface only, so as the Plastics Industry Pipe Association's electrofusion guideline puts it, welding "can be used to join pipes manufactured from different PE material grades (i.e. PE80 and PE100) and Standard Dimensional Ratios (SDRs)."

On a network that has been extended in phases over fifteen years with whatever was available each time, that flexibility is not a nicety — it is the only practical way to tie the old work into the new.

That flexibility comes with a rule attached, and it is the rule this whole guide turns on. It is covered in the next section, and getting it wrong is how a buyer ends up with a technically sound joint on a system that no longer carries the pressure it was designed for.

SDR, grade and pressure class: the compatibility rule that governs everything

Electrofusion fittings are sized on outside diameter, and PE pipe of a given nominal diameter has the same outside diameter at every SDR — the wall thickness changes inward. That is why a single coupler part number can serve a range of SDRs, and why "will this fitting fit my pipe?" has a simpler answer than most buyers expect: if the outside diameter matches and the fitting's stated SDR range includes yours, it fits.

Whether it performs is a different question. PIPA states the governing rule plainly: "Where different materials or SDRs are used, the PN rating of the pipeline system is equivalent to that of the lowest-rated pipe or fitting component."

A PN16 fitting on PN10 pipe gives you a PN10 system. A PN10 fitting on PN16 pipe also gives you a PN10 system — and that second case is the expensive one, because the buyer paid for PN16 pipe and quietly de-rated the network at every joint. When you are pricing fittings against a bill of materials, the fitting's pressure class is not a specification detail to be traded down for margin. It is the ceiling on everything it touches.

The failure mode underneath this is a naming problem. SDR is a geometric ratio — outside diameter divided by wall thickness — and it is not a pressure rating on its own, because the same ratio in a weaker compound yields a lower class.

IFAN's own HDPE specification page states the mapping and then names the trap directly: in PE100, "SDR11 gives PN16, SDR17 gives PN10, SDR21 gives PN8 and SDR26 gives PN6.3", while "SDR11 in PE80 is PN12.5, not PN16," which IFAN describes as "the most common specification error we see on inbound enquiries." A buyer who orders "SDR11 fittings" believing they have ordered PN16 has specified nothing about pressure at all until the grade is named alongside it.

Two practical consequences follow. First, put the PE grade, the SDR and the PN in every line of the purchase order, for pipe and fittings separately — "DN110 PE100 SDR17 PN10" leaves nothing to interpretation, and "DN110 SDR17" leaves the grade open.

Second, when a fitting is quoted with an SDR range rather than a single SDR, ask which end of that range the pressure class is stated against. PIPA's instruction on this is to "always check the fitting barcode or packaging and consult the manufacturer if unsure," which is a polite way of saying the range printed in a web catalogue is not a specification you should build a network on. The same discipline applies across the rest of the HDPE and PE product range: the grade and class belong on the paperwork, not in an assumption.

↑ Back to top

Coupler, reducer, tapping tee or branch saddle: pick by what the joint has to do

Most catalogues list these four side by side as if the choice were about shape. It is not. ISO 4427-3:2019 splits electrofusion fittings into two structurally different families, and the split determines the tooling you need, the cooling time you must schedule and whether you can work on a live main at all.

Socket fittings: couplers and reducers

ISO 4427-3 defines an electrofusion socket fitting as one that "contains one or more integral heating elements that are capable of transforming electrical energy into heat to realize a fusion joint with a spigot end or pipe." In plain terms, the pipe goes inside the fitting. Straight couplers join two pipes of the same diameter; reducing couplers join two different diameters in one part.

A reducer is the right call when the diameter change and the joint happen at the same place. When they do not, a straight coupler plus a reduced pipe run is often cheaper and always easier to source, because reducing couplers are the part most likely to be missing from a distributor's shelf in the size combination you want.

Saddle fittings: tapping tees and branch saddles

Saddle fittings fuse onto the outside of a pipe rather than around its end, which is how you take a branch off a main without cutting the main. Here ISO 4427-3 makes a distinction that most product pages blur, and it has direct cost consequences. A tapping tee is a saddle "which contains an integral cutter used for cutting through the wall of the main pipe, which remains in the body of the tapping tee after installation." A branch saddle is a saddle "which requires an ancillary cutting tool for drilling the hole in the adjoining main pipe."

Read that difference as a procurement instruction. The tapping tee carries its cutter with it, so every unit is self-contained and the hole is made after the fusion has cooled — which is what makes tapping under pressure possible on a live main. The branch saddle is cheaper per unit because the cutter is not in the box, but you must own the drilling tool and the main must be depressurised and drilled as a separate operation. Buying branch saddles to save unit cost, on a project where the main cannot be taken out of service, is a specification error that surfaces on site rather than at the quotation stage.

There is a second split inside the saddle family that governs tooling. PIPA describes under-clamp saddles as fittings that "fully encircle the pipe, with an integral clamp" and notes these "are often used with live tapping saddles", whereas top-loading saddles need "a manufacturer-specific mounting tool to securely position and restrain the fitting during welding", and that tool "is removed from the prepared joint following each welding and cooling process."

The phrase to notice is manufacturer-specific. If you buy top-loading saddles from two brands on one project, you need two sets of mounting tools, and a crew that grabs the wrong one loses the joint. Standardising the saddle brand across a project is worth more than the unit-price difference between brands.

Diameter range: where your size actually sits

Two different diameter ranges get quoted in electrofusion conversations and they are not interchangeable. One is the range of PE pipe a supplier can produce. The other is the range in which electrofusion fittings are commonly stocked and in which the published installation procedures actually apply. Buyers get caught when they read the first number and plan against the second.

On the pipe side, IFAN's published HDPE specification gives an outside diameter range of "DN20 – DN1600 (PE100)", supplied as "Coils to DN90; 6 m and 12 m straight lengths." That DN90 coil boundary is worth holding onto, because it predicts a problem rather than describing a product. Coiled pipe carries residual curvature and arrives out of round, which is the single most common reason an electrofusion joint needs re-rounding clamps and alignment work. Everything at DN90 and below is likely to reach site coiled; everything above it arrives straight. If your bill of materials sits mostly below DN90, budget for the geometry work described later in this guide.

On the procedure side, the scope limits matter more than the marketing ranges. The Plastics Pipe Institute's Municipal Advisory Board procedure — the document most North American utilities point their crews at — is titled for "Field Joining of 12 Inch and Smaller Polyethylene (PE) Pipe", and its cooling-time and out-of-roundness tables stop at 12 inch. Above that size you are outside the generic procedure and into manufacturer-specific instructions, which is exactly where a buyer should stop assuming and start asking the supplier for the written procedure that covers the diameter on the drawing.

The practical takeaway for sizing a purchase is that diameter changes which questions you have to ask, not just which part number you order. Small diameters bring coil-induced ovality. Large diameters take you past the generic procedure, add re-rounding tooling, and lengthen cooling times to an hour per joint. Neither is a reason to avoid electrofusion; both are reasons to price the labour and tooling honestly rather than assuming a fitting is a fitting.

↑ Back to top

The barcode is a quality control, not a convenience

Every electrofusion fitting carries a label with a barcode, and most buyers treat it as a convenience feature that saves the operator from typing. It is considerably more than that, and the standard behind it is one a buyer can name in a specification.

ISO 13950:2007 is the standard that governs it. Its scope covers "the characteristics of automatic recognition systems (numerical recognition by means of bar codes or magnetic cards, electromechanical recognition using implanted-resistor connectors and self-regulation systems) that enable the energy supply to be delivered automatically to the thermoplastic electrofusion fittings used in pipe jointing." It replaced the earlier technical report ISO/TR 13950:1997, so a supplier citing the 1997 document is quoting something that was superseded nearly two decades ago.

Two physical parameters in that standard are worth quoting back at a supplier, because they are checkable on a printed label rather than taken on trust. Both permitted formats are "2-in-5" interleaved codes with a specified thick-to-thin bar width ratio of 2,5 — a printing tolerance, and a badly reproduced label that misses it is a label a scanner may reject in a wet trench. And the code carries an internal checksum, so a mis-keyed digit fails rather than silently loading the wrong fusion time into the box.

The detail that belongs in your purchase order is the format. ISO 13950 permits exactly two, and they are not equivalent. The standard specifies "the 24 digit '2-in-5' interleaved type" and "the 32 digit '2-in-5' interleaved type, including traceability coding" — in both cases with a thick-to-thin bar width ratio of 2,5, and with an overall length where "one of these digits is composed by a control character (checksum)."

Only the 32-digit format carries traceability coding. If your project, your utility client or your own quality system requires per-joint traceability back to a specific fitting batch, a 24-digit barcode will not deliver it, and no amount of asking the installer to write things down will retrofit it. That is a question to settle with the supplier before the order, not a discovery to make during commissioning.

What the barcode does at the moment of fusion is the stronger argument for insisting on it. PPI's electrofusion user guide describes the control box performing "basic checks prior to the beginning the fusion to verify that the fitting coil has continuity, that the coil resistance matches the barcode information, and that the voltage being supplied is sufficient", and during the cycle detecting "certain assembly or fitting errors such as shorted heating coils or short-stabbed pipe ends."

A resistance cross-check against the scanned barcode catches a counterfeit or mislabelled fitting before any energy goes into the joint. Manual entry of fusion time and voltage skips that comparison entirely — the box does what it is told. PIPA adds that "barcode reading welding machines automatically adjust for variable ambient temperature conditions", which manual entry also forfeits.

That temperature correction is worth putting a number against, because it is the check most often lost when a crew keys parameters by hand. The qualified installation window runs roughly -10°C to 45°C, and a fitting fused at 5°C needs a materially different energy input from the same fitting fused at 40°C.

A scanned barcode applies that correction automatically; a hand-keyed weld time applies whatever number the operator read off the label, at whatever temperature the trench happens to be. On a DN315 coupler carrying a 60 minute cooling clock, the cost of getting that wrong is not one joint — it is an hour of crew time per re-weld, and up to 24 hours before the assembly falls below 45°C if the joint has to be re-fused at all.

Two supporting details are worth knowing. The 24-digit value is also "printed on the label, either directly above or below the barcode", so a scuffed or unreadable barcode does not strand the joint — the operator can key the digits and the checksum still catches a typo. And the machine itself has a standard: PIPA specifies welding machines "manufactured in accordance with ISO 12176-2" that "operate with barcode-marked fittings according to ISO 13950", with a safety device that "should operate in less than 0.5 seconds." Those two standard numbers, ISO 13950 for the fitting and ISO 12176-2 for the machine, are the pair to write into a tender document.

↑ Back to top

Out-of-roundness, flat spots and toe-in: the geometry gate

An electrofusion fitting works by melting both surfaces and letting the expanding polyethylene close the annular gap between pipe and socket. That mechanism has a limit. If the gap is too wide before you start, the melt expansion cannot close it, and MAB-1 is explicit about the consequence: when pipe is out of round and not corrected, "the amount of gap between the pipe and fitting can be too large for the melt expansion to close." The fitting will still run its cycle. The pins will still rise. The joint will be hollow in places.

Two terms describe the same condition and get confused constantly. MAB-1 defines out-of-roundness as "the difference in the maximum measured diameter minus the minimum measured diameter" — a length, d1 minus d2. Ovality is the same difference "expressed as a percentage", calculated as (d1 − d2) / D_average × 100. North American procedures generally quote the length; metric and Australasian procedures generally quote the percentage. When a supplier's data sheet and your project specification disagree, check first whether they are simply using different units for the same measurement.

The acceptance limits, in both systems

MAB-1 gives imperial maximum out-of-roundness figures alongside a clear threshold for when re-rounding becomes necessary: "2″ IPS / CTS and smaller diameter tubing is flexible enough that the coupling and alignment clamps will provide the necessary rounding forces and no other re-rounding device is needed", while "for sizes equal to or larger than 3″ IPS / DIPS, re-rounding clamps may be needed on either side of an electrofusion fitting." PIPA gives the metric equivalent as a percentage that steps at DN315: below that, ovality at the fusion zone must not exceed 1.5% of DN or 3 mm, whichever is smaller; at DN315 and above, 1% of DN or 5 mm, whichever is smaller. Flat spots are capped separately at 3 mm depth.

The clamp timing is a detail crews get wrong. PIPA states that "re-rounding tools must remain in place during the welding and cooling phases" — not just while the fitting is being slid on. Pull them at the end of the fusion cycle and the pipe relaxes back toward its oval shape while the melt is still soft, which undoes the reason you fitted them.

The defects re-rounding cannot fix

Re-rounding is not a universal remedy. PIPA notes that "re-rounding tools may be unable to correct flat spots", and where pipe cannot be brought inside the acceptance criteria the options are to trim "back the end of the pipe until the flat spot is removed", relocate or rotate a saddle, or butt-weld pipe tails onto the end.

There is also a defect that is not about roundness at all: pipe end reversion, or toe-in, where residual manufacturing stress pulls the end inward and creates a gap the melt cannot close. PIPA's test for it is arithmetic a site engineer can run with a tape and a straight edge — measure DN × 5% back from the pipe end, and at that position the outside diameter must be no smaller than the DN. On a DN500 pipe that means measuring 25 mm back and confirming the OD is still at least 500 mm; if it is not, cut squarely at that point.

The warning attached to that test is the one that changes how a crew plans its day. PIPA cautions that residual stresses remain after cutting, so "freshly cut pipe can revert again, sometimes within a few hours." Preparing a run of joints in the morning to fuse in the afternoon is not the efficiency it appears to be — the prepared ends can toe back in before the fitting goes on.

↑ Back to top

Scraping is a specified dimension, not a cleaning step

Polyethylene develops a thin oxidised skin from manufacture and UV exposure, and that skin will not fuse. Removing it is the step that decides whether the joint works, and it is where MAB-1 places the overwhelming majority of failures: contamination from poor pipe preparation, geometry problems, and alignment errors together "account for more than 95% of all fusion failures." Not fitting quality. Not machine settings. Preparation.

The dimension is fixed and smaller than people expect. MAB-1 states that "an adequate minimum amount of material that must be removed is just seven one-thousandths of an inch (0.007″)", a thickness "approximately the same as two sheets of ordinary paper."

The oxidation layer does not deepen meaningfully with storage time, so the requirement is the same for pipe delivered yesterday and pipe that has sat in a yard for two years — a useful fact for an importer holding stock.

What almost no product page mentions is that there is also a maximum. Scraping is not a case of more being safer: remove too much and the pipe outside diameter drops, the annular gap widens, and you have engineered the same failure that out-of-round pipe causes. MAB-1 tabulates both bounds by size.

The tool matters as much as the depth. MAB-1 states that "sandpaper, emery cloth, or other abrasives should never be used to prepare a pipe surface for electrofusion", that "wood rasps and metal files are not acceptable peeling tools", and that hand scrapers "are not recommended due to inconsistent surface preparation."

Abrasives polish contaminants into the surface instead of removing a layer, and they leave no evidence of how deep they went. A rotary peeler produces a continuous curl the crew can measure — which is the point. If a supplier or subcontractor is proposing to prepare joints with abrasive cloth, that is a reason to stop the work, not a cost saving.

Two more numbers close this out. PIPA specifies the peel length as "half the fitting length plus 20 mm from the end of the pipe" — and the reason for the extra 20 mm is an inspection feature, because one of the post-weld checks is that "the peeling is visible beyond the ends of electrofusion fitting." That turns an invisible step into something an inspector can verify after the joint is made.

And for damaged pipe, MAB-1 sets a replacement threshold: "if the gouge exceeds 10% of the pipe wall thickness, that pipe section should be cut out and replaced." Deeper gouges cannot be scraped away without breaching the maximum peel depth. If the surface is re-contaminated after preparation, the specified remedy is "a minimum 90% concentration of isopropyl alcohol" — and denatured alcohol is explicitly ruled out because its additives can prevent fusion.

↑ Back to top

Temperature, power and the conditions that void the joint

Electrofusion is more weather-tolerant than butt fusion, which is one of the reasons it gets specified. It is not weather-proof, and the operating window is narrower than most buyers assume in both directions.

The temperature window

PPI puts the typical qualified installation range at "14°F minimum to 113°F maximum", while noting that manufacturers state their own limits and some go wider. PIPA, working in metric and describing the machine rather than the fitting, gives electrofusion welding machines "upper and lower temperature operating limits of -10°C to 45°C." Those two windows are the same window: 14°F is -10°C and 113°F is 45°C. Two independent industry bodies converging on identical limits is a strong signal that this is a real physical boundary, not a conservative manufacturer's disclaimer.

The heat end catches more projects than the cold end, and the reason is solar gain rather than air temperature. PIPA warns that black pipe and fittings in direct sunlight "can absorb solar energy and may reach well above ambient environmental temperature", with "pipe surface temperatures as high as 70°C."

A 32°C day in West Africa or the Gulf is comfortably inside the ambient window while the pipe lying in the sun beside the trench is 25 degrees outside it.

The remedy is shade, and PIPA specifies the target: use sun protection such as a portable gazebo "to reduce the pipe's surface temperature to 45°C or less." For a project buyer this is a line item — the shelter is equipment, not an optional comfort, and a crew without one will either stop work at midday or make joints that should not have been made.

Moisture is absolute rather than graduated. The welding surfaces must be clean and dry, the assembly must be sheltered from rain, and PIPA's instruction is that "welding should not be undertaken if protection cannot be provided." Trenches that collect mud and water have to be dewatered first. There is also a workspace minimum that gets designed out of tight excavations and then discovered on site: "in trenches, a minimum clearance of 150 mm is required around the pipe", and more as fitting diameter grows, because the crew needs room to run a peeler, fit re-rounding clamps and attach leads.

Power is a procurement decision, not a site detail

The control box has to deliver a precise energy input, and unstable power is a leading cause of aborted welds. PPI notes control boxes are "typically available in 110v or 220v versions" and must operate "within a frequency range of 50 to 60 Hertz" — which is a genuine trap on export projects, because a box bought for one market may not match the generator fleet in another. PIPA specifies the supply more tightly: nominal "230V-240V" unloaded, held "stable at 230V±15%" under load, stable frequency of 50-60Hz under load, and extension cables that "shall not exceed 50m" and must be completely unrolled.

Generator sizing follows the fitting, and PPI publishes the figures: sockets from 1/2 inch to 2 inch need a 2.5 kVA minimum generator on a 15 amp breaker, sockets from 3 inch to 12 inch need 5 kVA with a 30 amp breaker at 110v or 20 amp at 220v, and saddles of all sizes need 2.5 kVA. Extension cords are specified by gauge, #10/3 at 25 feet and #8/3 at 50 feet. A crew arriving with a 3 kVA generator for 8 inch couplers will get intermittent faults that look like defective fittings and are not.

Cooling time is schedule, not slack

Polyethylene is a good insulator, so the joint stays soft long after the current stops. MAB-1 publishes the most conservative cooling times across all listed manufacturers, and the list of things that must wait is longer than "pressurising": the cooling time must be observed "prior to removal of clamps, movement, backfilling, pressure testing, tapping, or placing the fitting in service." For couplings up to 250 psi that is 30 minutes at 3/4 inch to 2 inch, 45 minutes at 3 inch to 8 inch, and 60 minutes at 10 inch and 12 inch.

Saddles follow a separate table and it is keyed to outlet size, not to the diameter of the main — 30 minutes for 3/4 inch to 4 inch outlets and 60 minutes for 6 inch to 12 inch outlets. A 2 inch service connection on a 12 inch main therefore clears in 30 minutes, which is a scheduling advantage worth knowing when planning a day of service tie-ins.

One caution on re-welds: if power is lost mid-cycle, some manufacturers allow a re-weld once the fitting has cooled to ambient, but PIPA notes the time for a complete assembly to fall below 45°C "could be up to 24 hours depending on fitting size and environmental conditions." That is not a lunch break.

↑ Back to top

Which standards to write into your purchase order

Electrofusion fittings sit under different standards depending on what the pipe carries and which market it is going into. Naming the right one is what makes a specification enforceable; naming a withdrawn one tells a supplier you copied it from an old document.

For water, the current international reference is ISO 4427-3:2019, which covers PE fittings "intended for the conveyance of water for human consumption, raw water prior to treatment, drainage and sewerage under pressure, vacuum sewer systems, and water for other purposes", for systems up to a maximum allowable operating pressure of 25 bar at a 20°C reference temperature. It is also the document that supplies the tapping tee and branch saddle definitions used earlier in this guide. In Europe the parallel is EN 12201. IFAN's published HDPE specification names both, listing the product standard as "ISO 4427 (water) / EN 12201 (Europe)" with "DIN 8074/8075 for dimensions and testing."

For gas, be careful with ISO 8085-3:2001. It genuinely is the polyethylene electrofusion fitting specification for "the supply of gaseous fuels", applicable to fittings "designed to be fusion-jointed to PE pipes conforming to ISO 4437" — but it has since been superseded by the ISO 4437 series. Citing ISO 8085-3 in a 2026 tender is not fatal, though a supplier who quotes it back to you as their current conformance is working from an old data sheet, and that is worth a follow-up question about what else on the sheet is out of date.

For anything entering the United States, ASTM F1055 is the specification to name, and there is a current regulatory reason to get the edition right. PHMSA published a direct final rule on 24 April 2026 updating the edition incorporated into the federal pipeline safety regulations to "ASTM F1055-16a (Reapproved 2022) ... approved November 1, 2022", replacing the 2016 edition, with the incorporation by reference approved "for Sec. 192.283(a); appendix B to this part."

The rule is effective 1 January 2027, and compliance after 23 June 2026 is already authorised — so as of now both editions are live in different places, and a purchase order that simply says "ASTM F1055" without an edition is ambiguous at exactly the moment the reference is changing. MAB-1's screening instruction is the practical version of this: "all electrofusion fittings should be marked to indicate that they meet the design and performance requirements of ASTM F1055 before being considered for use."

Requirements vary by market, by application and by the utility's own adopted specification, so treat the list above as the starting point for a conversation with your local authority or consulting engineer rather than as a universal compliance answer.

Potable-water approvals in particular — certification marks such as NSF/ANSI 61 or AWWA C906 that MAB-1 mentions as additional markings — are separate from the product standards above and have to be confirmed for the specific market and the specific product, not assumed from a general ISO conformance claim. For a wider view of how these families fit together, our guide to HDPE pipe standards covers the pipe side, and pipe certification standards covers what the marks themselves mean.

↑ Back to top

A worked selection: DN110 branch off a PE100 main

Take a common job: a DN110 PE100 SDR17 distribution main, already in service, and a new DN63 service branch that has to come off it without shutting down supply. Here is how the decisions above resolve, in the order you actually make them.

The branch cannot interrupt supply, so the fitting family is settled first: this is a tapping tee, not a branch saddle, because only the tapping tee carries the integral cutter that lets the hole be made after fusion on a live main. Choosing a branch saddle here to save unit cost would require depressurising the main, which is the thing the job was defined to avoid.

Pressure class comes next. The main is PE100 SDR17, which is PN10. The tapping tee must be PN10 or better, because the system takes the rating of its lowest-rated component. If the only tee in stock is PN16, that is fine and costs nothing in performance — the main still governs at PN10. If the only tee in stock is PN8, the branch connection has just de-rated that point of the network below the pipe it is fused to, and it should be rejected regardless of price.

Then the pipe gets measured, because the fitting will not forgive geometry. DN110 sits below the DN315 step, so the ovality limit is the smaller of 1.5% of DN or 3 mm: 1.5% of 110 is 1.65 mm, so 1.65 mm is the limit.

Using PIPA's own worked figures, a pipe measuring d1 = 111.0 mm and d2 = 109.5 mm has an ovality of 1.5 mm, which is inside the 1.65 mm allowance and passes. Had it measured 112.0 and 109.5, the 2.5 mm result would have failed and a re-rounding clamp would be needed on the main before the saddle went on. At DN110 the pipe may well have arrived coiled, which makes that measurement more likely to matter, not less.

Preparation follows the fixed dimensions. Scrape at least 0.007 inch from the saddle footprint, using a rotary peeler and never abrasive cloth, and check the pipe for gouges — anything deeper than 10% of wall thickness means that section comes out rather than gets scraped harder. Fit the manufacturer's mounting or clamping device for that specific saddle, scan the barcode rather than keying the parameters so the box can cross-check coil resistance and correct for ambient temperature, and confirm the machine is fed by a supply holding within tolerance under load.

Finally, the schedule. The outlet is DN63, roughly 2 inch, so the saddle cooling table gives 30 minutes — keyed to the outlet, not to the DN110 main. Clamps and any re-rounding tools stay on for all of it. Only after that 30 minutes has elapsed can the tee be tapped, the joint moved, backfilled or pressure tested. One crew, one joint, from arrival to tapped branch, is comfortably an hour's work — and pricing it as fifteen minutes because the fusion cycle itself is short is how electrofusion labour gets under-quoted.

↑ Back to top

Is electrofusion right for your project?

Electrofusion is a genuinely excellent jointing method with a narrow set of conditions in which it is the wrong choice. Being honest about those conditions saves more money than negotiating the fitting price.

The operator-skill row deserves emphasis because it is the one buyers discount. PIPA states that "the biggest single contribution to a successful electrofusion weld is the competency of the welder and their dedication to correct surface preparation and weld procedures", and in Australia that is formalised — installers must be trained and certified to a specific competency unit and hold a current certificate, with PIPA recommending reaccreditation every 2-3 years. Most markets have no such requirement, which does not make the underlying dependency go away.

If you are buying fittings for a crew that has not done electrofusion before, the training is part of the cost of the method, and the alternative is discovering the 95% preparation-failure statistic on your own project. Our guide to HDPE joint types sets out the alternatives if this table points you away from fusion.

↑ Back to top

What to check before you accept an electrofusion shipment

Most of what goes wrong with electrofusion fittings can be caught at goods-in, before anything reaches a trench. This is the documented inspection sequence, drawn from the published procedures rather than from a supplier's marketing sheet.

• Packaging integrity first: fittings should arrive individually sealed in transparent bags inside cartons. PIPA is explicit that the packaging protects the fusion surface from contamination and sunlight and that "the fitting must be stored in its original packaging until immediately before use." A carton of loose, unbagged fittings is a rejection, not a discount negotiation.
• The ASTM F1055 marking: per MAB-1, fittings should be marked to show they meet its design and performance requirements before being considered for use. Check the marking exists on the fitting, not only on the invoice.
• Barcode format: confirm whether the label carries the 24-digit or the 32-digit ISO 13950 code, and check that the human-readable digits are printed alongside it. If your contract requires traceability, the 24-digit format will not satisfy it.
• Stated SDR range and pressure class: both should appear on the fitting or its label with the PE grade named. A fitting marked only "SDR11" without a grade has not told you its pressure class.
• Storage conditions on arrival: PIPA specifies storage undercover between 0°C and 50°C. For an importer, that is a warehouse specification — a container standing in the sun on a quayside can exceed it.
• Matching tooling: for top-loading saddles, confirm the manufacturer-specific mounting tool is included or already owned, and that it matches the brand actually shipped rather than the brand quoted.
• Material documentation: ask for the certificate covering the shipment. IFAN issues a material certificate per shipment on its HDPE line, and any supplier should be able to do the same.

One warning belongs here rather than in an installation guide, because it changes what a buyer should accept as evidence of a good joint. Fusion indicator pins are widely read as a pass mark — they rise, the joint is good. PIPA states the opposite in plain terms: "the melt indicator pins confirm that some melt pressure has developed in the weld. They are not a confirmation of a successful weld." Pins can rise on a joint made over an unscraped or oval pipe.

If a contractor's quality regime consists of photographing risen pins, it is not a quality regime. The checks that carry weight are the fusion record from the control box, the absence of error messages, no molten polyethylene having escaped the socket, no displaced heating wires visible in the annular gap, the insertion-depth mark still where it started, and the scraped band visible beyond both ends of the fitting.

↑ Back to top

Conclusion

Selecting electrofusion HDPE fittings well comes down to three questions the product grids do not ask: what does this joint have to do, what is the lowest-rated component in it, and is the pipe in good enough geometric condition for the melt to close the gap. Get those right and electrofusion delivers a joint as strong as the pipe, on sites where nothing else would work. Get them wrong and the fitting will still complete its cycle and still raise its pins over a joint that was never going to hold.

If you are specifying a PE network rather than buying a handful of parts, it is worth settling the grade, SDR and pressure class on paper before you compare fitting prices, since the cheapest fitting that de-rates your main is the most expensive item on the order. You can review the full HDPE range and its published specifications, or send your project's grade and size schedule to our engineering desk and we will tell you what we can supply against it.

Frequently Asked Questions