September 2026
The short version: whatever is stuck to it, not what it is made of
The easiest way to know whether this method will work on your problem is to ask one question. Is the stuff you want gone bonded to the surface, or is it part of the surface?
Bonded means it arrived later and attached itself. Grease, carbon, adhesive, paint, ink, resin. All of that is a separate layer sitting on top of a substrate, held there by heat, chemistry or time. Thermal shock breaks that bond and the layer releases.
Part of the surface means there is no interface to break. Rust is iron that has become iron oxide. Scale on a boiler tube has grown into the metal. Corrosion pitting is an absence, not a deposit. Dry ice does not help you there, and we will say so.
Everything below is on the first list.
Grease and oil film
This is the most common single thing we take off, and it is the layer people misjudge. Grease is soft, so the assumption is that anything will shift it. What actually happens is that the oil film bonds dust, swarf and carbon against the metal and holds them there, which is why a part can look clean and still feel dirty two minutes after you wipe it.
Engine bays, gearboxes, hydraulic packs, machine bases, chain drives, bearing housings. The thermal method lifts the whole loaded layer rather than smearing it around, which is the failure mode of every rag-and-solvent approach.
Carbon and soot
Carbon is the material this process is best at, because heat put it there and cold takes it off.
You find it on blower housings, industrial ovens, furnaces, weld cells, exhaust components, turbine casings and anything that has been in a fire. It goes on as a hard crust, packed into every radius and corner where a brush cannot reach. Under thermal shock it is usually the first thing to let go.
Fire and smoke work is its own category, and it turns into soot removal rather than routine maintenance. The advantage on a damaged building or piece of equipment is that you are not adding water to something that has already been soaked by a fire department.
Adhesive, sound deadening and tape residue
Anyone who has pulled factory sound deadening out of a car floor knows what comes next: a brown film that laughs at a heat gun and a scraper. Same story with old badge adhesive, vinyl wrap residue, label glue on production equipment and the tape that held a mask in place during a repaint.
Adhesive is a good candidate because it is brittle when cold. It goes from a stretchy smear to something that fractures and flakes off. On the printing and packaging side that same property is why adhesive removal is one of the routine uses of this method.
Release agent and resin
Mold release builds up on tooling every cycle, and eventually it starts showing up as a defect on the part instead of a film on the tool. Solvent soaking works and takes the tool out of production; manual scrubbing risks the surface.
Cured resin, sealant and overspray on tooling and fixtures behave the same way. This is the work behind release agent removal, and it is why injection molds, composite tooling and press surfaces get cleaned in place between runs.
Paint and coatings
This one comes with a caveat, and we would rather give it to you now than on site.
Failing coatings come off well. Brittle, chalked, cracked or poorly bonded paint has already lost most of its grip, and thermal shock finishes the argument. Thin single-stage finishes and old undercoating are usually straightforward.
A healthy modern two-part coating properly bonded to sound steel is a different job, and honest answer, dry ice is not always the right tool for it. We test a patch and tell you what we are seeing rather than promising a result over the phone. When it is the right tool, coating removal has the large advantage of not putting a profile into the substrate you wanted to keep smooth.
Ink, paper dust and coating build-up
Press work is precision work. Anilox rollers, plates, doctor blades and dryer sections foul with dried ink, coating and paper lint, and the usual options are solvent soaking or a brush, both of which risk the cell structure that determines how the press lays ink down.
Non-abrasive matters here more than almost anywhere. Change the roller surface and you have changed the print.
Food residue
Carbonized product, fat, starch, sugar and burnt-on build-up on ovens, conveyors, mixers and processing equipment. The reason this method is used in food production is not speed, it is that dry ice leaves no chemical residue on a food contact surface and no standing water in a place where standing water is a sanitation problem in its own right.
Mildew staining and general household grime
On the residential side the list gets more ordinary: garage floor grime, grill and smoker carbon, mildew staining on masonry, old finish on furniture and trim, soot on a fireplace surround. Household surfaces are usually softer than industrial ones, so the non-abrasive part of this matters more rather than less.
Salt and corrosion product
South Florida gives everybody this one whether they asked for it or not. Salt does not stay outside. It comes in through louvers, settles on engine rooms and bilges, crusts onto trailer frames and outboard brackets, and then holds moisture against metal for the rest of the year.
Salt itself comes off easily; it is a deposit, not a bond. The value is in what you do not lose while removing it. Abrasive cleaning takes the protective coating along with the salt, which leaves bare metal in a marine environment, which is a worse position than you started from. That is the whole argument on marine work: remove the contamination, leave the coating that was holding corrosion back.
Loose corrosion product on top of sound metal will come away with it. Pitting that has already eaten into the metal will not, because there is nothing left to remove.
Things we get asked about that are a no
Rust, as covered. Heavy mill scale. Deep corrosion pitting, because the metal is already gone. Surface profiling for a coating spec, which needs an abrasive by definition. And loose mud on heavy equipment, which is a hose and five minutes, not a specialist visit.
Saying no to those is not modesty. Every job we take that was wrong for the method costs us more than the job was worth. If you are not sure which list your problem is on, describe it through the contact page and you will get a straight answer either way.
Questions we get on this list
How do you know before starting whether it will come off?
We test a small, inconspicuous area and watch what happens. Thirty seconds of blasting tells you more than any description of the contaminant. If it is not releasing, you find out on a patch nobody sees instead of halfway across the part.
Can you get one layer off and leave the one underneath?
Often, yes, and it is one of the more useful things about the method. Undercoating off a painted frame, overspray off a finished panel, carbon off a coated housing. It depends on how different the two bonds are, which the test area shows.
Does it clean inside passages and blind holes?
Line of sight matters, the same as any blasting. What helps is that if a pellet does find its way into a passage, it leaves as gas rather than staying there as grit, which is the reason people are nervous about abrasive near an assembled machine.
If your contaminant is not on the list
Describe it and send a photo. We look at fouled equipment across South Florida every week and can usually tell you on the spot whether this is your method. Call 786-302-6553 or use the contact page. If dry ice is the wrong tool we will tell you on that call, not after taking your money.