Spring special ends 15 September: 30% off qualifying HACCP and FSSC 22000 courses. Code SPRING30 at checkout. See qualifying courses

Dead Legs in Food Processing Pipework: How to Find Them, Measure Them and Get Rid of Them

Home / Guides / Dead legs

HYGIENIC DESIGN · PIPEWORK · CIP VERIFICATION

A dead leg is a length of pipe with no flow through it: a capped sampling point, a blanked spare tee, a gauge or drain valve on a stub. Cleaning solution shears past the mouth at full velocity and never enters, so product sits there between batches, warm and wet, and is pushed back into the line on start up. You find them with a tape measure in about ten minutes a line. You size them with two measurements: L, from the wall of the main pipe to the face of the cap, and D, the bore of the branch. EHEDG closed equipment guidance sets a maximum L over D of 1. 3-A practice is commonly cited as 2, and never longer than 127 mm. EHEDG Guideline 8 itself gives no L over D figure at all, so quote the right document when you specify.

Key facts

  • What it is a branch or stub with no flow through it, on an otherwise cleanable line
  • How to size it L over D, where L is measured from the main pipe wall to the face of the cap and D is the branch bore
  • Commonly cited limits EHEDG closed equipment guidance, L over D of 1; 3-A, 2 and never longer than 127 mm
  • What Guideline 8 says no L over D figure at all, so do not cite it for this number
  • The permanent fix cut it out and weld the cap flush with the main pipe wall, or move the valve onto the pipe wall
  • Course this is Module 3 of Intermediate Hygienic Design, nine hours

Find every dead leg on your own line

Intermediate Hygienic Design takes you through pipework, couplings and welds, valves, tanks and CIP verification, with a practical application video that walks you through measuring a dead leg on your own plant. Nine hours, online, self paced.

See the Intermediate course →

What a dead leg actually is

Take a pipe carrying product from a tank to a filler. Anywhere a branch leaves that pipe and goes nowhere, you have a dead leg. The four that turn up most often on a South African plant are a sampling point that was removed and capped, a spare tee blanked off for a machine that was never bought, a pressure gauge or drain valve mounted on a stub, and an old transfer line that was isolated rather than removed.

The problem is not that the branch is dirty. The problem is that cleaning cannot reach it. Flow in the main pipe shears across the mouth of the branch at full velocity. A small eddy forms at the opening and the bulk of the solution carries straight on. Nothing travels up the stub, so the chemistry, the temperature and the mechanical action that clean the rest of the circuit stop at the mouth. Whatever is in that branch stays there, at room temperature, between production runs, and gets pushed back into the line the moment flow starts again.

That is also why a dead leg is invisible to your normal verification. A swab on the main line comes back clean, because the main line is clean. A CIP return conductivity trace looks correct, because the circuit reached temperature and chemistry everywhere the solution went. The stub is not on the trace.

How to find every one, in about ten minutes

You need a tape measure, a torch, a phone and somebody from engineering walking with you. Do not use the drawings. The drawings show what was designed, and dead legs are what was installed afterwards.

Walk one product line end to end and photograph every branch that does not carry flow in normal production. Cap, blank, gauge, sample point, drain valve, isolated line. Number each photograph. On a typical ready meals or dairy line you will find between four and a dozen.

While you are there, look for the two faults that travel with dead legs: the high point in a run that traps gas, so cleaning solution never touches the top of the pipe, and the low point that holds rinse water after the circuit drains, which then sits until the next run pushes it into product.

The two measurements, and the arithmetic

Measure L from the wall of the main pipe to the face of the cap. Not from the outside of the fitting, not from the weld: from the bore of the main pipe to where the branch ends.

Measure D as the bore of the branch, the inside, not the outside diameter of the tube. On a full bore tee this is the same as the line size, which is why the two get confused.

Then divide. A dairy in KwaZulu-Natal had an old sampling point on a 76 mm raw milk line, blanked with a cap 600 mm out. Six hundred divided by seventy six is about eight. On a sauce kettle, a ball valve sitting 120 mm off a 40 mm outlet pipe gives an L over D of 3.

Write both numbers on the job card, not just the ratio. A ratio with no measurements behind it cannot be checked by the next person, and it will be questioned.

Which limit applies, and who says so

This is where most specifications go wrong, because the number people quote and the document they credit it to do not match.

SourceWhat it sets
EHEDG closed equipment guidanceA maximum L over D of 1
3-A practice, commonly citedL over D of 2, and never longer than 127 mm
EHEDG Guideline 8No L over D figure at all. It covers hygienic design principles, radii, surface finish and drainability, and it is routinely miscited for this number

The practical consequence: if you write an L over D limit into a purchase specification and credit it to Guideline 8, a competent supplier will eventually check, and the rest of your specification loses authority with it. Name the document you are actually relying on, with its edition.

What the fix is, and what it costs

There is only one permanent fix for a stub that is too long: remove it. Cut it out and weld a cap flush with the wall of the main pipe, or move the valve so that it sits on the pipe wall with no stub at all. Anything else manages the problem rather than solving it.

Interim controls, while the money and the shutdown are found: increase flow velocity through the main line during CIP where the circuit allows it, add a dedicated flush on the branch, or manually strip and clean the branch on a stated frequency with a record. All three are temporary by definition, and each one belongs on the risk register with an owner and a date.

The economics usually favour the fix. A welder and a day of downtime is a known, single cost. A dead leg costs you a recurring risk that no amount of cleaning time removes, and it is the kind of finding that turns a positive environmental result into a difficult conversation with a customer.

Getting it funded

Findings do not get funded. Scored findings with owners and dates do. Put every branch you measured on a hygienic design risk register: the location and the photograph reference, L and D and the ratio, a severity, likelihood and exposure score, the band that produces, the interim control in place today, the permanent fix, its cost, its owner and its date.

Then write three sentences to management naming the worst three, what is holding the line safe today, and what the fix costs and when it happens. That is the document that gets a welder booked.

Frequently asked questions

What is a dead leg in food processing?

A length of pipe with no flow through it, branching off a line that does have flow. Capped sampling points, blanked spare tees, and gauges or drain valves on stubs are the common ones. Cleaning solution sweeps past the mouth rather than through the branch, so product sits in it between batches and is pushed back into the line on start up.

How do you calculate L over D on a dead leg?

Measure L from the wall of the main pipe to the face of the cap, and D as the bore of the branch. Divide L by D. A 600 mm stub on a 76 mm bore gives about 8. Record both measurements, not just the ratio, so the figure can be checked.

What is the maximum acceptable dead leg length?

It depends which document you are working to. EHEDG closed equipment guidance sets a maximum L over D of 1. 3-A practice is commonly cited as 2, and never longer than 127 mm. EHEDG Guideline 8 gives no figure for this, so it should not be cited as the source.

Can a dead leg be cleaned by increasing CIP flow or time?

Not reliably, and not as a permanent control. Higher velocity in the main line changes the eddy at the mouth of the branch but does not produce flow through the stub. Longer times and stronger chemistry cost money every night without solving the geometry. Treat these as interim controls with a date on them.

Does a dead leg show up on a swab or on the CIP record?

Usually not. The main line swabs clean because the main line is cleaned, and the CIP record shows temperature, flow and time everywhere the solution actually went. The stub is not on the trace. This is why dead legs are found with a tape measure rather than with verification data.

Who should be measuring dead legs, QA or engineering?

Both, walking together. QA knows which lines carry which risk and what the standard asks for, and engineering knows what can be cut out and when. A walk done by one without the other produces a report rather than a job card.

Find every dead leg on your own line

Intermediate Hygienic Design takes you through pipework, couplings and welds, valves, tanks and CIP verification, with a practical application video that walks you through measuring a dead leg on your own plant. Nine hours, online, self paced.

See the Intermediate course →

Sources and further reading

Leave a Comment