I have inspected close to forty second-hand duct lines in the last decade, in Turkey, Vietnam, Mexico, Poland and a few places I would rather not drive to again. About one in three was a genuinely good buy. The rest were priced as if a respray and a wash had reset the clock. So before I talk about micrometers and test cuts, I want to start where the real information is: why the machine is for sale.
The reasons fall into four groups, and each one changes what you should look at. A shop closing down is usually the best case, because the line stopped working for commercial reasons, not mechanical ones. A shop upgrading from a Line 3 to a Line 5 is the second best case, and you can often buy the old line from the same dealer who is supplying the new one, which means service history exists. A line sold because it never hit the promised output is the dangerous category, because the seller has usually already tried the cheap fixes and failed. And a line that has been sitting under a lean-to for two years is a lottery: rust is visible, but seals, belts and control batteries are not.
Ask directly. If the answer is vague, walk the shop floor and look at what is still running. A fabricator who cannot tell you the year of the last hydraulic oil change will not have a maintenance file either, and you should price the machine accordingly.
I do this with the machine cold and locked out, because half the faults I find disappear once oil is warm and clearances have taken up. Twenty minutes, a torch, a 150 mm caliper, a straight edge and a notebook with no phone in sight. Nothing exotic.
Start at the decoiler. Look at the mandrel segments for fretting and at the coil car ram for bend. A machine that has handled 1.5 tonne coils for a decade will show a polished ring where the coil bore rubs; that is normal. What is not normal is a mandrel that will not expand evenly, because that means the coil walks and every blank is cut slightly skewed. Check the brake disc thickness and the pneumatic hose date stamps. Hoses older than eight years are hardened, and a burst hose on a decoiler with a full coil is not a maintenance inconvenience, it is an incident.
Then the straightener and feed rolls. Run a fingernail across each roll face. Fine circumferential scoring is acceptable; grooves you can feel with a fingernail are not, because they print through onto the sheet and show up later as marks on every visible duct face. Ask for the feed roll pressure readings if the machine has gauges, and check the servo coupling for backlash by rocking the roll by hand.
The shear is where most of the money hides. Open the blade holder and look at the cutting edge under a torch at a shallow angle. A blade that has been resharpened three or four times will show a narrow, bright land along the edge; that is fine. A blade with chipped sections or a rounded, shiny radius along its full length is finished, and a replacement set on a 1300 mm shear runs into real money.
On a lock-former station, the thing I check first is not the rolls but the side guides and the sheet path. Put a 500 mm offcut through by hand, slowly, and watch whether it tracks straight or climbs. Climbing means a roll is out of parallel or a bearing is collapsing, and both are fixable but neither is cheap.
For a TDF station, bring a strip of the thickness you actually intend to run, not the thickness the seller used. I carry 200 mm wide strips of 0.6, 0.8 and 1.0 mm galvanized for exactly this. Set the machine as the seller recommends, form a flange, and measure three things: flange height at both ends, the angle of the fold, and whether the pocket will accept a corner insert without force. If you are comparing that station against a modern duplex TDF flange former, remember the old single-sided unit needs a second pass on every piece, which is roughly forty extra seconds per blank. A flange that measures 33 mm at one end and 35 mm at the other will fight you on every joint for the next ten years.
Beading rolls are the most commonly worn item nobody mentions. Run a bead on a scrap piece and look at the profile: it should be a smooth, symmetric groove, not a sharp crease with a bright line at the root. A crease means the male roll has worn to a knife edge, which raises the risk of stress cracking in service, especially on 0.5 mm material.
On a Pittsburgh lock station, form a full lock and try to open it by hand. It should resist. If you can peel it apart with two fingers, the rolls are worn or the clearance is set for thinner material than you run, and every seam in your ductwork will need sealing compound to hold pressure.
Open the cabinet with the main breaker off and look for three signals. First, dust: a cabinet packed with galvanized dust has been running without door gaskets, and that dust is conductive. Second, re-terminated wires with mismatched ferrules, which tells you somebody has been chasing faults. Third, the PLC and drive model numbers, photographed and sent to the OEM before you pay anything, because a 2011 drive that is discontinued in your region is a five-figure surprise when it fails.
Then ask for the password. This sounds trivial until you are the third owner of a machine whose screen language is locked, whose parameters cannot be exported, and whose OEM has changed hands twice. If the seller will not provide it, that is a negotiating point worth several thousand dollars, not a detail.
Check the voltage and phase honestly against your own supply. A line built for 380 V 50 Hz running on 415 V 50 Hz is usually fine, but a 60 Hz machine on a 50 Hz supply runs every motor and the hydraulic pump about 17 percent slower, which quietly costs you throughput forever. I have seen buyers discover this after installation, when the promised 800 square meters per shift turned into 650.
| Inspection point | Accept | Negotiate / repair | Walk away |
|---|---|---|---|
| Decoiler mandrel and brake | Even expansion, brake disc above 80% thickness | Hoses over 8 years, worn pads | Cracked mandrel segments, bent coil car ram |
| Straightener / feed rolls | No grooves catch a fingernail, backlash under 0.05 mm | Light scoring, worn coupling | Deep grooves, seized bearing, servo alarm history |
| Shear blades | Bright land under 3 mm, no chips | One chipped corner, resharpening needed | Full-length rounded edge, cracked blade holder |
| TDF / TDC station | Flange height within 0.5 mm end to end | Roll wear requiring shimming | Pocket rejects standard corner inserts |
| Beading rolls | Smooth symmetric profile | Shallow wear, polishable | Knife-edge male roll, cracked shaft |
| Pittsburgh lock | Lock resists hand opening | Loose lock correctable by clearance | Rolls bottomed out with no adjustment left |
| Hydraulic system | Oil clear, filter date known, no weeping cylinders | Oil change overdue, minor seepage | Metal in oil, scored rod, pump noise under load |
| Control cabinet | Clean, gaskets intact, PLC password available | Dust ingress, one obsolete drive | No password, undocumented rewiring, burnt terminals |
| Spare parts and manuals | Printed manual plus electrical drawings | PDF only, drawings partial | Nothing at all |
Never accept a demo on the seller's favourite job. Bring a drawing. I use a simple 500 x 400 mm straight duct with one TDF flange on each end and a bead down the centre of the two long sides. It exercises the shear, the bead, the flange station and the notching in one pass, and it takes about four minutes.
Then measure. Length tolerance should hold within 0.5 mm over 500 mm on a healthy servo feed; 1.5 mm or worse means the encoder, the feed roll grip or the shear timing is tired. Squareness across the diagonal of the blank should be within 1 mm. Flange height within 0.5 mm. Run five pieces and check the spread, not just the average, because a machine that produces 500.0, 500.4, 499.6, 500.3, 499.8 is healthy while one that produces 499.0, 500.5, 498.7, 501.2, 499.3 will ruin your assembly line.
Listen during the cut. A clean shear produces a single dull thud. A double thud or a ringing follow-on sound means the blade is passing the material twice, usually a timing or clearance issue. On the hydraulic side, watch the pressure gauge during the flange stroke; a needle that jumps 20 bar above the rated figure is telling you about a worn seal or a relief valve that has been cranked up to compensate.
Finally, ask to run the line for a full hour, not five minutes. Thermal drift shows up in the second half hour, and so do feed problems caused by a coil that has been sitting with a set in it.
Buyers consistently underestimate the landing cost of a used line. My working rule is to add 18 to 25 percent of the purchase price for a machine that has been idle more than a year, and 10 to 15 percent for one that is still in production.
A 5-metre line is not a pallet. Expect to pay for a low-bed trailer, a crane at each end, and possibly a permit for an oversized load. On a recent purchase into Southeast Asia, freight and rigging came to 14 percent of the machine price, and the buyer also had to cut a door opening 300 mm wider to get the frame through.
Assume you will replace all hydraulic oil, all filters, every pneumatic hose older than five years, the shear blades if they are marginal, and at least one set of feed rolls. On a mid-size line that package typically lands between 4,000 and 9,000 US dollars in parts alone, before labour.
If you want a sense of what a new line includes and what a used one will therefore lack, compare the specification list on an auto duct line 5 against the machine in front of you. Features like automatic coil loading, servo-driven corner notching and a proper dust extraction connection are often absent on older machines, and retrofitting them rarely makes economic sense.
Ask for four documents and treat their absence as a price reduction. The electrical schematic, the PLC program backup, the hydraulic circuit diagram, and any record of repairs. A machine with drawings is a machine your local electrician can fix on a Sunday. A machine without them is a machine that depends on one retired technician in another country.
Also confirm ownership and lien status, especially where the machine is being sold by a dealer who took it as trade-in. I have seen a buyer pay in full and then spend eleven months arguing with a leasing company. It is a boring check and it has saved clients six figures.
There is a real case for buying second-hand, and it is not only about price. If your order book is 60 percent of what a new line would justify, a used auto duct line 3 at 45 percent of the new price can pay back in eighteen months rather than three years. Older machines are also simpler, and simplicity has value in a market where the nearest OEM service technician is a two-day flight away.
What I would not do is buy a used machine as a growth platform. If you expect to add a second shift within two years, the downtime profile of an ageing line will cost you more than the capital saving. The honest comparison is not used versus new at the same capacity; it is used at 70 percent availability against new at 92 percent. Run both numbers through your own order book before you sign.
One more thing worth checking: whether the tooling on the used machine matches the duct systems you actually sell. A line set up for TDF with 20 mm pockets is a bargain if your market uses TDF, and a liability if your contractors expect TDC or a simple slide-on flange. Retooling a station can cost as much as a third of the machine's value, and it is the single most common reason a cheap used line turns out not to be cheap.
How many hours is too many on a used duct line? Hours alone tell you little, because a line that ran two shifts of thin gauge can be healthier than one that ran one shift of 1.2 mm stainless. I care about the ratio of hours to tonnage. Above roughly 25,000 hours with heavy gauge work, expect to rebuild the shear and at least one forming station. Below 12,000 hours, wear is usually cosmetic.
Can I inspect a line without powering it up? You can catch perhaps 60 percent of the problems cold: blade condition, roll wear, backlash, rust, hose age, cabinet cleanliness. The other 40 percent, including servo faults, hydraulic pressure behaviour and thermal drift, only appear under load. If the seller refuses a powered test run, discount the price heavily or walk away.
Is it worth paying an OEM technician to inspect? Yes, if the machine is a brand the OEM still supports. A two-day inspection typically costs 800 to 1,500 US dollars plus travel, and it routinely finds issues worth ten times that. The report also gives you a written baseline for the negotiation.
What is a fair price for a ten-year-old Line 3? In the current market, a well-maintained 2014 to 2016 Line 3 with TDF tooling typically trades at 35 to 50 percent of the equivalent new price, depending on region and included spares. Below 30 percent usually means you are buying a project rather than a machine.