Installation of a reduced Hepworth T-piece: step-by-step soldering process

We are still preparing products.

But you can look at other categories.

Installation of a reduced T-piece from Hepworth: soldering step by step

The reduced T-piece is one of the most commonly used fittings in the entire central heating system. Precisely in places where a branch to a radiator is taken off from the main pipe or where two different pipe dimensions meet, the quality of the joint determines the reliability of the entire system for decades to come. Soldering copper pipe with Hepworth fittings is one of the basic craft-technical skills of every heating technician, yet even experienced plumbers make mistakes here that only become apparent during the first filling – in the form of a leaking joint, corrosion, or mechanical stress in the piping system.

This article will guide you through the entire process of installing a reduced T-piece, from the preparation of tools and materials, through the soldering itself, to the inspection of the finished joint. You will find specific numbers, practical tips from real customer jobs, and an explanation of why each step is done exactly that way – not just mindlessly reciting the procedure, but truly understanding the physics and metallurgy behind it.

What is a reduced T-piece and how does it differ from a straight one?

Before we move on to the soldering itself, it is important to be clear about what we are actually installing. A reduced T-piece differs from a straight (equal) T-piece in that at least one of its branch openings is in a different dimension than the other two. In practice, this means that the fitting connects two different pipe sizes in one connecting element without the need for a separate reducer.

For example, the T-piece 22 x 22 x 15 has both running openings in the DN22 dimension and the branch is reduced to DN15. This is a typical combination in heating distribution, where the main loop runs in DN22 and a branch is taken off to a smaller radiator supplied by a DN15 pipe. Similarly, the T-piece 28 x 28 x 22 is used to branch off from a floor-level DN28 line to a connecting branch DN22.

For more information on how to read the three-digit designation of fittings and what the individual numbers in the name mean, see the article Dimensions of reduced T-pieces: how to read the designation 22 x 22 x 15 in this Knowledge Center. The differences between straight and reduced T-pieces in terms of hydraulics and selection are discussed in the article Difference between straight and reduced T-piece in copper piping.

Tools and materials – what you need to have ready before starting

Proper preparation is half the success. On jobs, we see that most problems with soldered joints do not arise during the soldering itself, but even before it – poorly cut pipe, insufficiently cleaned surfaces, or unsuitable solder. Prepare the following:

Basic tools

Copper pipe cutter – a special circular cutter (not a hacksaw, unless absolutely necessary). A cutter ensures a perpendicular, clean cut without burrs. For DN15–DN28, a standard compact cutter with one or two wheels is sufficient.
Internal burr remover (deburring tool) – most cutters have it integrated in the form of a conical tip. It removes the internal burr that forms during cutting and unnecessarily increases hydraulic resistance.
Emery cloth or wire brush for cleaning – grit 120–180. Some plumbers use a special steel brush (pad) specifically designed for cleaning copper fittings.
Gas torch – for dimensions DN15 to DN28, a standard propane or butane torch with an adjustable flame and a power of about 1.5–3 kW is sufficient. For DN28 and larger, a stronger torch is recommended, as the fitting has a higher thermal capacity and slow heating leads to overheating of the solder before the solder melts.
Soldering mat or heat-resistant plate – protects the surrounding structure from damage by the flame and reflected heat.
Pipe pliers or pipe holder – to prevent the assembled system from moving during soldering.
Wet cloth or damp sponge – for cleaning the joint after soldering, while it is still warm.

Materials – solder and flux

Soft solder Sn97Cu3 (or a similar lead-free type) – for heating, lead-free solder according to EN 29453 is mandatory. Melting temperature around 220–230 °C. It is supplied as a rod or coiled wire.
Flux – paste or liquid based on organic acids, specifically designed for copper pipe and drinking water/heating systems. Read the label – some fluxes are only suitable for drinking water systems and are not suitable for temperatures above 70 °C in heating systems.
Cleaning agent for degreasing – acetone or isopropyl alcohol spray, or acetone on a cloth. Grease and oils from hands are one of the main enemies of a quality joint.

Preparation of the pipe and fitting – a critical phase

A soldered joint works on the principle of capillary wicking – the molten solder is "drawn" into the gap between the inner wall of the fitting and the outer wall of the pipe. This gap should be in the range of 0.05–0.15 mm. If it is too large (for example, due to a deformed pipe), the solder will not flow evenly and cold spots will form. If it is zero (the pipe cannot be inserted), there is nowhere for the solder to flow.

Step 1: Cutting the pipe

Press the circular cutter perpendicular to the pipe and rotate it around the circumference while gently tightening the screw – only a small rotational step per one full rotation around the pipe. Too fast tightening deforms the end of the pipe into an oval shape. The resulting cut must be straight, perpendicular, and free of burrs.

For information: for DN22, the outer diameter of the pipe is exactly 22.0 mm and the inner bore diameter of the Hepworth fitting is 22.15–22.25 mm, which corresponds to the mentioned capillary clearance of 0.075–0.125 mm on each side.

Step 2: Deburring and checking the end of the pipe

Remove the internal burr all around the circumference using a conical tip or deburring tool. Then visually and with your fingers, check that the end of the pipe is not oval. If the pipe has been bent during installation or transport, you can carefully correct it with calipers for small diameters (DN15) – for DN22 and DN28 it is more difficult and it is recommended to cut the section again.

Step 3: Cleaning with emery cloth

Grind the outer surface of the pipe end (to a depth corresponding to the depth of the fitting socket, usually 15–25 mm) with emery cloth to a clean, shiny copper surface. Move longitudinally and in circular motions. The inside of the fitting socket can be cleaned with a rolled strip of emery cloth or a special cylindrical cleaner of the appropriate diameter.

Warning: do not touch the surfaces with bare hands after grinding. Finger oil (human sweat contains chlorides) will degrade the cleaned surface and create a barrier for the flux. If you touch the pipe, clean it again.

Step 4: Applying the flux

Using a brush or a piece of cloth, apply a thin, even layer of flux paste to the outer surface of the pipe end and the inner surface of the fitting socket. The flux must cover the entire contact area. Do not overdo the amount – excess flux flows out during soldering, burns, and forms black oxide deposits that are difficult to remove later. A thin layer is sufficient.

Flux serves two purposes: first, it chemically removes the thin oxide layer from the cleaned copper (copper oxidizes very quickly in air, even without visible changes), and second, it reduces the surface tension of the molten solder and allows it to capillary penetrate the gap.

Step 5: Assembly and fixing

Insert the pipe or pipes into the socket of the fitting until they reach the stop and slightly rotate them (about 30–45°) to evenly spread the flux. Check that the system is properly fixed – either supported by clamps or a bracket – to prevent movement during soldering. In a reduced T-piece with three sockets, fixing is more complicated because the fitting is not symmetrical. Make sure all three pipes are inserted and positioned at the correct angle (90° between the straight axis and the branch axis).

The soldering itself – procedure for a flawless joint

This is the core of the entire process. Soft soldering of copper takes place at temperatures around 220–250 °C – significantly lower than hard soldering (brazing), where temperatures exceed 600 °C. Soft soldering is not demanding on the torch, but it requires precision and patience.

Step 6: Heating the fitting

This is the most important principle: the torch is held on the body of the fitting, not at the point where the solder is applied. The solder melts due to the heat of the metal, not the flame. If you apply the flame directly to the solder, the flux will burn, the solder will form into balls, and it will not flow into the joint.

For the T-piece 22 x 22 x 15, heat the body of the fitting using circular movements of the flame, moving slowly around the entire circumference. Watch the color change of the flux – it first bubbles and sizzles (moisture evaporates from the paste), then turns black (organic components burn), and finally becomes a shiny liquid. This is your signal that the temperature is close to the target.

For a larger T-piece 28 x 28 x 22 or T-piece 28 x 22 x 28, you will need more heating time, as the body of the fitting has a higher thermal capacity. Be patient and slowly rotate the torch, do not rush.

Step 7: Temperature test and application of the solder

When the flux is in a shiny liquid state, bring the end of the solder wire close to the edge of the fitting socket (to the line where the fitting meets the pipe). If the temperature is correct, the solder will immediately melt and appear to be "drawn" into the gap – you will see the solder as if it were self-penetrating. One or two centimeters of solder (depending on the diameter) is sufficient for one joint; further dosing is unnecessary waste and creates thick droplets on the surface.

With a reduced T-piece, you have three sockets. Solder them one by one – first the two straight openings (DN22), then the reduced branch (DN15 or DN22, depending on the specific piece). Solder each socket separately. After soldering the first socket, let the joint cool slightly (15–20 seconds), then continue with the next one.

Step 8: Cooling and cleaning the joint

After applying the solder, let the joint cool down slowly. Never pour cold water onto a hot copper joint – thermal shock can cause microcracks in the joint, even if they appear undamaged from the outside. Let it cool for about 1–2 minutes. While the fitting is still warm (but not hot), wipe the joint with a damp cloth – it removes oxide residues from the flux and the shine of the final joint is more convincing. After it has fully cooled, it is advisable to wipe the joints with a weak solution of baking soda (neutralization of residual acidic flux), especially if it is a piping system where the flux is not otherwise flushed away.

Special situations and difficult cases during installation

Soldering in confined conditions

A real customer job rarely looks like a laboratory demonstration. A T-piece is often installed in a ceiling, under the floor, or in a technical shaft, where there is no space for comfortable work. In such cases:

Use a burner with a longer nozzle extension or a bent nozzle.
Ensure sufficient ventilation – fumes from the solder are not healthy even in open space, let alone in an enclosed flue.
Thermal-resistant sheeting or boards are mandatory, not optional.
Never solder directly near flammable materials (wooden beams, thermal insulation) – always separate the working area with a thermal barrier.

Connecting to an existing pipe run

If you are inserting a T-piece into an existing pipe (typically when expanding the heating system with a new radiator), you must drain the system of water and purge it before starting work. Residual moisture inside the pipe prevents reaching the correct soldering temperature – water turns into steam and actively cools the joint. Before starting work, blow out open sections with compressed air or let them drip freely for a few minutes. Sometimes the hidden issue during replacement or soldering is micro-moisture in the wall of the old pipe – for old pipe runs, this can be resolved by longer preheating.

Combining three different diameters

Certain situations require a T-piece where all three openings are in different dimensions. For example, if you are running DN28 as a floor-level route, a DN22 branch is taken off from it, and you also need to connect a DN15, you must either use two T-pieces in succession or choose a correctly dimensioned combined reducer. For this scenario, the T-piece 28 x 15 x 28 is suitable, where the running branch remains in DN28 and the branch is DN15, thus saving one fitting. The issue of combining different diameters is discussed in more detail in the article Combining different diameters in a heating circuit using reduced fittings.

Pressure test and checking the joints

After completing all joints and before closing the structure (false ceiling, plaster, floor), a pressure test is a mandatory step. There is no visual inspection that can reliably replace a pressure test. Sometimes a joint may look fine, but the solder may mask a fine, discontinuous solder line – and it is precisely there that a leak will appear when heated to operating temperature.

Pressure test procedure:

Fill the system with cold water and purge all branches of air.
Using a hand pressure pump (manometer pump), increase the pressure to 1.5 times the operating pressure – for standard home heating, the operating pressure is 1.5–2.0 bar, so the test pressure should be 3.0 bar.
Let the pressure stabilize for at least 30 minutes and monitor the manometer. A drop of more than 0.1 bar indicates a leak.
Look for leak locations visually (drips) or use a foaming leak detection spray.

If a leaky joint is found: drain the system, let the pipe dry completely, and do not solder while it is wet. Remove the joint, apply fresh solder, and resolder. If there are doubts about the integrity of an old joint, it is sometimes faster to replace the fitting with a new one rather than repeatedly adjust the old one.

Most common mistakes – what we see in customer jobs

After years of experience in customer jobs, the same mistakes keep repeating. A detailed analysis can be found in the separate article Common mistakes when installing copper T-pieces and how to avoid them, but here are the most critical ones:

Insufficient cleaning: The most common cause of discontinuous, porous joints. Solder applied to an unclean surface will not work properly.
Overheating: Too long or too intense heating burns the solder before it can flow into the joint. Result: black, brittle solder residue and a joint that does not fully penetrate.
Moving the joint during solidification: If you touch the joint or move the pipe in the first 30 seconds after applying the solder, the joint loses its crystalline structure and becomes brittle. It may hold tightly, but it will loosen during a pressure test or the first heating.
Soldering a wet joint: Water inside the pipe actively cools the joint and prevents reaching the working temperature. The system must be thoroughly drained and purged of air.
Wrong solder (with lead content): Lead-containing solders are not acceptable for potable or heating systems according to current standards. Always check the product labeling.
No solder on the fitting's socket: Many apply solder only to the pipe and forget to apply it inside the fitting's socket. The joint is only partially covered.

What to do after completion – long-term care of joints

Copper soldered joints are extremely durable when done correctly – a well-made joint can last for decades without any intervention. However, a central heating system requires regular inspection, which includes checking visible joints. What to look for:

White or greenish deposits around the fitting's socket – a sign of slow leakage (carbonates from water).
Corrosion on the outer wall of the fitting near the joint, which may indicate the presence of moisture from condensation or micro-leakage.
Mechanical stress in the pipe – if the pipe "works" (clicking when heated), check whether the joints are exposed to tensile or torsional stress. A soldered joint is not a structural element and must not catch forces from thermal expansion of the pipe.

A detailed procedure for regular inspection and maintenance is covered in the article Inspection and maintenance of copper fitting joints in a central heating system.

Dimension comparison – approximate sizes of Hepworth T-piece sockets

T-piece designation	Running branch	Branch	Typical use
T-piece 22 × 22 × 15	DN22 / DN22	DN15	Feeding a smaller radiator from a DN22 main branch
T-piece 22 × 15 × 22	DN22 / DN22	DN15 (middle)	Branch from the center of the T-piece in DN15, both end openings DN22
T-piece 28 × 28 × 22	DN28 / DN28	DN22	Branching off a DN22 branch from a floor-level DN28 route
T-piece 28 × 22 × 28	DN28 / DN28	DN22 (middle)	Central branch in DN22, the running route remains DN28
T-piece 28 × 15 × 28	DN28 / DN28	DN15	Direct feeding of a small radiator from DN28 without an intermediate fitting

Safety at work – clearly explained

Flame soldering is work involving an open flame, and although the temperatures involved are relatively low compared to welding, there is still a risk. A few basic rules that should always be followed:

Never leave the torch unattended. Always close it after setting it down.
Flux vapours are irritating. Work in a well-ventilated area or wear a respirator for organic vapours.
Store gas cylinders upright and away from heat sources.
Hot pipes and fittings cannot be visually distinguished from cold ones. Always assume that a fitting may be hot and work with heat-resistant gloves.
When working in enclosed spaces with propane, monitor whether gas is escaping from the torch connection – propane is heavier than air and stays close to the ground.

Most frequently asked questions (FAQ)

Do I need to degrease the pipe before soldering, if I have already sanded it with sandpaper?

Yes, it is recommended. Sandpaper mechanically cleans the surface, but it can become contaminated again during handling from fingers or from the oil of the sandpaper itself. A quick wipe with acetone or isopropyl alcohol before applying flux is quick and significantly increases the reliability of the joint. This is especially important in industrial applications, where many joints are soldered in succession and hands inevitably get oily.

How much solder is enough for one DN22 joint?

For a standard DN22 socket with a depth of approximately 16–18 mm, 1.5–2.0 cm of 3 mm diameter solder wire is usually sufficient. Excessive solder forms thick droplets on the surface, which serve no sealing function and only visually "hide" possible poor areas. For DN15, use 1.0–1.5 cm, and for DN28, use 2.0–2.5 cm.

Can I solder a reduced T-piece vertically if the branch is pointing downward?

Yes, soldering in all positions is common and soft soldering of copper allows it. Capillary force is sufficient to overcome gravity when the gap is properly sized at 0.05–0.15 mm. The only difference is that with a downward-facing joint, excess solder may drip – so apply it more carefully and do not add extra if capillary action is working well.

How long do I have to wait before pressure testing after soldering?

For soft soldering, it is sufficient to let the joints cool completely to room temperature – this usually takes 10–15 minutes after completing the last joint. There is no need to wait longer, as soft solder solidifies immediately after removing the heat source and reaches its strength within a few minutes. Do not speed up cooling with cold water (thermal shock).

Can I use PTFE tape or hemp fibre as an additional sealing material for a soldered joint?

No, and this is a serious mistake. A properly executed soldered copper joint is fully hermetic on its own and does not require any additional sealing. PTFE or hemp fibre cannot fit into the socket and will not improve the joint functionally – they would only cause problems when heated. If you feel the need to "add" sealing material to a joint, it means the joint was not properly made and should be redone.

What should I do if flux is running out and burning during soldering – is this normal?

Partial burning of flux on the outer surface is a normal occurrence – the flux performs its function inside the joint and the excess on the outside oxidises. Small visible smoke clouds and darkening of the pipe surface around the tip are normal. A problem arises if the flux burns off in large amounts and solder droplets also escape – this indicates overheating or excessive flux. Slow down the heating and reduce the amount of flux for the next joint.

Conclusion

Soldering a reduced T-piece from Hepworth is a skill that can be learned and refined. Summarised into a few principles: surface cleanliness is the foundation, heat is applied to the fitting (not to the solder), flux indicates the correct temperature, solder is applied sparingly, and joints are never moved during solidification. If you follow these principles, the resulting joints will be reliable for the entire lifetime of the heating system – and that can be easily several decades with a well-designed system.

To choose the right T-piece for your specific case, use the accompanying articles in this Knowledge Centre – especially How to choose the right reduced T-piece for a heating system, When to use a reduced T-piece instead of a straight one in a heating distribution, and How to correctly choose the branch reduction when connecting a radiator to the main pipe. The full range of fittings can be found in the Hepworth reduced T-piece category.

Do you have a question about this topic?

Not sure or dealing with a specific situation in your home? Write to us – we are happy to help.