What a Support Has to Do Besides Hold Weight

Why this matters

Almost no support in the field fails by dropping the thing it holds. They fail by holding it perfectly while the run does something else: slides, rotates, sags between points, hammers, sweats, saws itself in half against the saddle, or pumps compressor noise into a bedroom wall. The clamp is still on. The weight is still carried. The complaint is real. If your mental model of a support stops at gravity, every one of those shows up as a mystery, because the part you were watching is fine.

The useful reframe is that a support is a statement about motion: at this point, the run may move in these directions and may not move in those. Weight is one of the directions.

Before you touch an existing support

A run that is currently held is holding energy. Release a clamp on a horizontal line and the segment either drops or springs, depending on whether it was sagging or bowed, so rig temporary support under both sides of the point before the clamp comes off, and stand out from under the run rather than beside it. On a suspended run this is stored mechanical energy under 29 CFR 1910.147, and where the release is part of a broader isolation, the run gets locked and tagged in its supported state first.

Two more that this article's own example requires. Roof work above 4 feet in general industry (29 CFR 1910.28) or 6 feet in construction (29 CFR 1926.501) needs fall protection in place before you carry a clamp up there, and the two Parts trigger at different heights, so know which one your job falls under. And a refrigerant, steam or fuel line is a pressurized path: never drill, cut, or drive an anchor where the bit or the fastener can reach it, and never take a clamp off a pressurized line to reposition it without confirming the line is restrained elsewhere.

If the run is insulated with thermal system insulation in a building constructed no later than 1980, that material is presumed asbestos-containing under 29 CFR 1926.1101 for construction work, with the general-industry duties at 29 CFR 1910.1001. Do not cut, tear or brush it back to get at the pipe. That is an inhalation hazard and a survey question, not a glove question.

The six duties, which are a list and not a ranking

At any given support point, some subset of these applies. They are not ordered, because which one governs is entirely a function of the point.

1. Carry the gravity load, meaning the run plus its contents plus its insulation plus anything that lands on it outdoors.
2. Set and hold position, including slope for drainage or condensate, clearances from other trades, and alignment into a piece of equipment.
3. Decide axial movement, meaning either restrain thermal growth at this point or deliberately let it slide.
4. Take lateral and transient loads, which is wind and seismic outdoors and water hammer, thrust at a change of direction, and torque reaction at a valve or a motor anywhere.
5. Break the vibration and noise path into the structure, where a rotating machine is upstream.
6. Protect the run from the support, meaning no point loading into thin wall, no abrasion, no crushed insulation, no dissimilar-metal contact.

The last one catches people because it runs backwards. Most trades think of a support as protecting the building from the pipe. It also has to protect the pipe from the clamp.

The gate

Before you pick hardware for a support point, answer two questions in writing.

Which of the six duties does this point owe? Not the run, the point. Two points twelve feet apart on the same line can owe completely different sets.

Do any two of them conflict? This is the question that separates a support that works from one that was chosen by habit. Duty 3 and duty 4 conflict at any point that must both slide and be held down. Duty 5 conflicts with almost everything, because isolation wants compliance and restraint wants stiffness. When two duties conflict at one point, the resolution is almost never a cleverer clamp. It is to move one of the duties to a different point.

Run that gate against two points on the same installation and it resolves in opposite directions.

Point one: the interior branch

A refrigerant line set running horizontally through a conditioned corridor, strapped to a channel strut every few feet along its length.

Duties owed: gravity, position, and protection of the run from the support. That is three of six. There is no wind, no rotating machine within reach, negligible transient load, and the thermal movement at this point is small and shared across a long row of hangers.

The right hardware is a clamp sized to the outside of the insulation with a rigid insert or shield under the strap so the clamp bears on the insert and not on the insulation. A bare strap pulled down on the insulation, which is what happens when a tech grabs a clamp sized for bare tube, satisfies duties 1 and 2 and fails duty 6.

That failure has a number attached to it. Crush a 3/4 inch insulation wall down to about 1/4 inch under the strap and you have cut the insulating thickness at that spot to roughly a third, so the thermal resistance there drops to roughly a third as well, and the outside surface of the insulation at that strap now runs closer to line temperature than anywhere else on the run. On a cold line that patch is the first place in the building to reach dew point, so it sweats, drips, and stains a ceiling tile two floors of complaint later. The crushed cell structure does not spring back when you loosen the strap, so the fix is replacing the insulation at that point, not adjusting the clamp.

Point two: six feet off the condensing unit

Same line set, same clamp in the van, but on the roof, six feet from a running compressor.

Duties owed: all six. Gravity and position obviously. Wind uplift, because a roof line set is a lifting surface and the local design wind speed is set by the adopted building code for your jurisdiction, not by judgment. Thermal movement, because the roof surface temperature swing across a day dwarfs anything indoors. Vibration, because there is a compressor six feet away. And protection from the support, because roof membrane, UV and grit are all present.

Three of those six pull against each other. Duty 4 wants the line held down against uplift. Duty 3 wants it free to grow and shrink along its axis. Duty 5 wants a soft path so compressor vibration does not run into the curb.

Resolving the conflict by splitting it across points

The resolution is to give each duty its own point rather than asking one clamp to be three things.

The uplift restraint and the axial freedom go to the same clamp, but as different directions: a clamp that captures the line vertically and laterally while bearing on a wear pad that lets it slide along its axis. That is a guide, and it satisfies duties 1, 2, 4 and 6 at once without touching duty 3, because it never claimed to hold the line axially.

The vibration break does not go into any clamp. It goes into the line itself, close to the machine, as a deliberate compliant element, and the first rigid support after that element is placed far enough away that the compliant element gets to work. A soft-mounted clamp under a line that also has to resist uplift ends up stiff enough to transmit vibration and soft enough to let the line lift, which is the worst of both. What a compliant connection can and cannot do for you has its own article; the point here is only that duty 5 is a location decision before it is a hardware decision.

What getting duty 3 wrong costs, measured in wear

Say the line at that guide point moves about a third of an inch each way over a daily cycle, which is the kind of figure the movement-budget method produces for a run of that length on a roof. That is about 0.7 inch of relative sliding per full cycle. One cycle a day is roughly 255 inches of sliding per year at that one contact patch, which is a little over 21 feet of the pipe wall dragged across the same spot every year.

Put that against a bare copper line resting on a galvanized channel edge with no wear pad and the outcome is not in doubt: it wears a groove, the groove thins the wall, and the failure arrives as a refrigerant leak at a support, which is a location nobody inspects because nothing is connected there. Put a wear pad or a shoe under it and the same 21 feet a year of movement wears a sacrificial part you can replace in minutes.

The other direction is just as instructive. Clamp that point rigidly instead of guiding it and the movement does not disappear, it goes somewhere else: into the next joint, into the equipment connection, or into the clamp's own fasteners, which back out over a season of being worked. The run does not stop moving because you decided it should.

Reading a support somebody else installed

Walk the run and ask, at each point, what it is restraining. Three field cues do most of the work.

• A shiny arc or a groove on the pipe at a support means that point is sliding and was never meant to. Something upstream is restraining what should have been free, or the guide was installed without a wear surface.
• Fasteners backed out at one point in a row of tight ones means that point is taking a load the others are not, which is usually a thermal or thrust load arriving at the one rigid clamp in a line of loose ones.
• A sag between two supports with sound hardware at both is a spacing problem rather than a support problem, and spacing has its own article, because span and anchorage together decide what a support point actually carries.

References

• 29 CFR 1910.147 for isolating and controlling stored mechanical energy before releasing a loaded support; 29 CFR 1910.28 (general industry, 4 feet) and 29 CFR 1926.501 (construction, 6 feet) for fall protection on roof work.
• 29 CFR 1926.1101, construction, which presumes thermal system insulation in buildings constructed no later than 1980 to be asbestos-containing, with general-industry duties at 29 CFR 1910.1001.
• The mechanical and plumbing codes and the building code adopted by your authority having jurisdiction, which own support spacing, seismic restraint and design wind loading in the editions adopted.
• See related: How Spacing and Anchorage Decide What a Support Carries; How to Support a Run That Has to Move; What a Vibration Isolating Connection Can and Cannot Do.