Why Acrylic Tube Crazes | Annealing Explained

That curl is not a bad cut. It is the tube going back to the shape it would rather be, and it is the plainest demonstration there is of what is locked inside an extruded acrylic section. How tightly it curls tells you how much stress is in there — and that is a scale, not a yes or no.

Residual stress is behind a large share of the failures people blame on "bad acrylic" — a bond that fails on one batch and holds on the next, cracks that appear weeks after the part was finished, a hole that splits the moment the drill breaks through. It is not the only cause, and we will come to the others. But it is the one that travels to you inside the tube, invisible, and it is the one you cannot inspect on arrival.

Where the stress comes from

Extrusion is a race between shape and temperature. Melt leaves the die at forming temperature and within seconds it is sized, pulled and cooled. The outer skin freezes first while the inside is still moving, so by the time the wall is solid, different layers through its thickness have set at different moments and want to be different lengths. They cannot all win, so the wall settles into a standoff — one layer in tension, another in compression, permanently pulling against each other.

Nothing about that is a fault in the process. It is what continuous extrusion does, to every extruder's tube, including ours. The only question is how much of it is taken back out afterwards — and by whom.

Stress is not a sign that the tube was made badly. Every extruded tube has it. What differs is how much is left by the time it reaches you.

The four ways it reaches your bench

Locked-in stress is invisible. It does not show on a dimensional report, it does not show in a photograph, and it does not show when the goods are inspected on arrival. It shows later, in four fairly specific ways.

1. Silvering along a solvent bond

Solvent cement works by softening the surface so two parts can fuse. Where that surface is already under tension, softening it lets the tension open the surface into a field of fine, bright cracks — silvering — running along the joint. You get it exactly where you needed strength, and you often get it on some parts of a batch and not others, which is what makes it so frustrating to chase.

2. A rough cut that turns into a split

A clean saw cut is a load the tube can carry. A rough one that chips the edge is a crack that has already started. On low-stress tube a chipped edge tends to stay a chipped edge. On high-stress tube it runs — sometimes immediately, sometimes as soon as the part is handled.

3. The slit tube that curls

Cut a tube open lengthwise and it pulls itself into a curl as the cut releases the balance. A heavily stressed tube can close up almost into a roll; a well-annealed one opens slightly and stays close to flat. This is the one test that needs no chemicals and no equipment, and it is the closest thing there is to seeing residual stress with your own eyes — provided you read it as a scale.

4. Spider-web crazing, weeks after cleaning

Someone wipes the finished part down with a cleaning fluid. Nothing happens. The job ships. Weeks later the surface is covered in a web of fine cracks.

The delay is why this one is so often blamed on the wrong thing. By the time the crazing appears, the cleaning is long forgotten and the tube gets the blame — or the resin does, or the weather does. The actual mechanism is stress plus chemical, and neither alone would have done it.

What annealing actually is

Annealing is not a quick bake. Every acrylic tube we ship goes into the oven and comes out roughly twenty hours later:

a controlled ramp up to temperature
about twelve hours held just under the material's heat-distortion temperature
a controlled ramp back down

Both the heating and the cooling are rate-controlled, and the slow cool is the half that does the real work. Held near that temperature, the molecular chains have enough mobility to let go of the arrangement they were frozen into. Cool them quickly at the end and you simply freeze them into a new one — you will have spent the energy and put the stress straight back. The twelve hours at temperature is the part people picture; the controlled descent is the part that decides whether it worked.

We do not publish our exact schedule. The principle is not a secret — it is in every resin maker's fabrication manual — but the specific temperature and ramp rates for our sections are something we worked out ourselves.

What annealing does not do

It does not take the stress to zero. No anneal does. What twenty hours in the oven achieves is bringing the residual stress down to a low level and, just as importantly, making it even through the wall instead of concentrated in layers. Some stress remains, and it will still show if you go looking for it.

We say this plainly because the opposite claim is easy to make and easy to disprove. Anyone who tells you their tube is stress-free is either not measuring or not telling you the truth. The useful question is never "is there stress?" — there always is — but how much, and is it low enough for what this part has to survive?

Why most extruders skip it

If it is this beneficial and the method is public, the obvious question is why it is not standard. Two reasons, and the second is the real one.

Cost. Twenty hours of oven time per batch is energy, floor space and lead time. It sits between production and shipping, so it makes every order slower. On a commodity price comparison it is pure margin given away for something the buyer cannot see in the box.

The oven. This is the part that stops most factories before they start. A tube can only be annealed standing upright — lay a long tube down for twenty hours at that temperature and it sags under its own weight. So the oven has to be taller than the longest tube you intend to anneal. For 2-metre and 3-metre lengths that means a very tall, purpose-built chamber, and you have to build it before you can anneal a single tube.

This is not a matter of trying harder or caring more. It is a capital decision a factory makes once, years before any particular order — and most never make it.

What changes once it is done

Solvent bonding succeeds far more often. The surface no longer opens up when the cement softens it.
Machining gets far more forgiving. Drilling, tapping, turning, routing — the rate at which a crack runs out from the cut drops sharply. Not to zero: a blunt tool in a hole too close to an edge will still find a way.
The part stays sound for longer. This is the one people miss. Residual stress does not stay constant in service: every heating and cooling cycle — every hot afternoon and cold night — adds a little more. A part that starts with high baseline stress has less margin before it reaches the point where something gives. Annealing does not stop thermal cycling and does not remove the baseline, but it lowers it — and over an application that runs for years, that margin is the difference between a part that ages and one that fails.

It is also the reason our extruded round tube can be used in contact with media that would craze stressed acrylic on sight. Stress and chemical attack are the same mechanism; remove one and the other has much less to work with. That is a longer subject and we will write it up separately.

Stress is not the only cause

It would be convenient for us to tell you that annealed tube does not crack. It is not true, and anyone who has worked with acrylic for a while knows it is not true. Cracking and crazing have several causes, and in the field they usually arrive in combination rather than one at a time.

Applied stress. Over-torqued fasteners, a part forced into a frame that is 0.5 mm too tight, a clamp holding it square. This is frequently larger than anything left inside the material, and no amount of annealing protects against it.
Restrained thermal movement. Acrylic expands roughly seven times as much as steel for the same temperature rise. Bolt a tube rigidly into a metal structure with no clearance, put it in the sun, and something has to give — it will be the acrylic.
Chemical contact. Solvents and cleaners, obviously, but also adhesives, the plasticisers that migrate out of some tapes and gaskets, and a surprising number of marker inks.
Tool and cut quality. A blunt blade, the wrong feed, or a cut that overheats leaves a damaged edge that behaves like a pre-started crack — whatever the state of the material.
Design. Holes drilled too close to an edge, sharp internal corners, no clearance for movement. These concentrate whatever stress exists into one place.
Age and weathering. Years of UV exposure make the surface less tolerant than it was on day one.

Annealing does not make the tube uncrackable. It removes one variable from the list — the one you cannot see, cannot measure on arrival, and did not choose.

That is the honest version of the claim, and it is still worth a great deal. The other six are yours to control: you can specify a clearance, change a blade, pick a different cleaner, move a hole. The stress locked into the wall at the die is the only one that arrives already decided, by someone else, before you ever see the part.

How to check the tube you are buying now

You do not need us to tell you whether your current supply is annealed. Take three offcuts from stock and find out:

Slit one lengthwise — and slit a second one from a different supplier the same way. Both will move; what you are reading is how much. Side by side, the difference between a well-annealed tube and an unannealed one is not subtle.
Wipe one with your cleaner — whatever your line actually uses — and leave it on a shelf where you will see it. Check it in two weeks and again in a month. Fine surface cracking is your answer.
Drill one near an edge. Not in the middle where it is easy — close to an end, the way a real bracket hole often has to be. Look at the hole wall with a light behind it.

Run the same three on a sample from anyone you are considering, including us. It is a better supplier comparison than any certificate.

Read the results for what they are: these tests isolate one of the variables above. A tube that comes through them well can still crack in service if it is clamped too tight or cleaned with the wrong thing. What they tell you is whether your supplier handed you a problem before the part even reached your bench.

What we do

Every acrylic tube we ship is annealed. Not as an option, not as a line item you have to know to ask for, and not only on the orders where a customer specified it. It is one of the reasons our lead time is what it is, and we would rather explain that than quietly skip the step to win a week.

What we will not tell you is that this makes the part immune. It lowers the odds — that is what it does, and that is all it does. If you tell us the medium, the fixing method and what the part gets cleaned with, we will tell you which of the other causes we think you should be worrying about instead.

After the oven, the tube is inspected again before it can enter stock — checked for out-of-roundness, residual stress and oil mist. A tube that went into the oven and came out wrong does not quietly rejoin the batch.

The stock sizes are published: 211 round tube sizes from Ø4 to Ø400 mm, plus square, rectangular and hexagonal sections. If you want to see how the rest of the process is controlled, the inspection plan is written out too.

Why acrylic tube crazes — and what annealing actually does