I almost approved a $38,000 used Andritz centrifuge
I manage procurement for a 120-person water treatment and separation company. For the past seven years, I've handled an equipment and services budget of around $1.8 million a year, negotiated with 40+ vendors, and documented every order in our cost tracking system. So when a used Andritz decanter centrifuge came into our candidate list in Q2 2024, I did what I usually do: I compared prices.
The first quote was $38,000. The second was $46,500. The third was $52,000. My first instinct was to approve the $38,000 option and move on. I almost did. Then I spent a week digging into what that price actually covered—and we cancelled the purchase.
I'm not a mechanical engineer, so I'm not going to pretend I can judge scroll wear or bowl geometry from a photo. What I can tell you, from a procurement perspective, is where the hidden costs live in a used Andritz centrifuge deal.
The problem: 'used Andritz' sounds like a standard product
Andritz is a well-known name in separation equipment—decanter centrifuges, dewatering systems, pumps, complete process lines. The Andritz centrifuges used in these plants are usually decanters, but the duty can be very different from one application to the next.
Two machines with the same model number can be completely different animals. One may have run in a food processing plant for three shifts a day for over a decade. Another may have come from a municipal sludge plant with lower hours but harsher duty cycles. The seller doesn't always know which one they're selling. If the seller doesn't know, the buyer certainly won't.
One listing described the unit as 'woolly bear.' I assumed that was a nickname for the bearing housing. No, wait—the seller meant the outside was covered in dried polymer fibers, kind of like a caterpillar. It took two or three calls to get that translated. I should have asked for photos of the inside of the feed inlet instead of guessing.
The deeper problem: quotes don't include history
When we asked every seller for the same information—serial number, original application, operating hours, maintenance records, reason for removal—the differences appeared fast.
- One seller sent a one-line description: 'Serviced and ready to run.'
- Another sent photos of the casing and nothing else.
- A third sent a full folder: operating history, bearing replacement log, and a note that the soft starter should be reviewed before startup.
The cheapest machine had the thinnest folder. 'Serviced' turned out to mean they had replaced an oil filter and washed the outside. Actually, they had also repainted the frame. No test run, no vibration report, no loading information.
The most expensive machine had the strongest documentation. That note about the soft starter saved us from a $6,000 integration surprise. You don't catch that by looking at the price column.
What missing history actually costs you
I don't have hard data on industry-wide used-centrifuge failure rates. I wish I'd tracked that more carefully since 2019. What I can say anecdotally is that in our experience, roughly one in three used-equipment purchases has some post-sale surprise within 90 days. Sometimes it's a sensor. Sometimes it's a seal. Sometimes it's the bearing.
A neighboring plant bought a used decanter centrifuge in 2023. It ran for eleven days, then the main bearing failed. The bearing kit, labor, and lost production added up to around $18,000. The machine itself cost $22,000. I heard that from the plant manager at a trade event, and it stayed with me.
At the second congress I attended in the water sector, another operator said something I've repeated ever since: 'You're not buying a used machine. You're buying someone else's maintenance history.' I didn't fully understand that until I ran the numbers on our three quotes.
Here's what the comparison looked like after I included everything I could think of:
- Purchase price: $38,000 vs. $52,000
- Freight and rigging: $3,200 vs. $1,800
- Electrical and controls: $8,300 vs. included in the higher quote
- First-year repair reserve: $7,000 vs. $5,000
The one-year total on the cheap machine was about $56,500. The higher-priced machine was about $54,800. Wait—let me be careful. Those totals came from our tracking spreadsheet, and I might be off by a few hundred dollars on freight. But the ordering was clear: the $38,000 quote was not cheaper.
The price difference was $14,000. The total cost difference was less than $2,000 in the other direction. That's the hidden part of used equipment pricing.
What we did instead
We didn't buy the $38,000 machine. We bought the $52,000 one, but only after a few conditions were met.
First, we took the serial number and sent it to Andritz Geldrop (the Andritz facility in Geldrop, Netherlands) to check whether the machine was still supported and service records were available. They came back with a yes. That simple confirmation removed the biggest unknown in the whole deal.
Second, we built a TCO spreadsheet. It's not fancy. It's just purchase price plus freight, installation, controls, and a repair reserve. For used industrial equipment, I usually set aside 15 to 20 percent of the purchase price for the first year. The right percentage depends on the documentation, the application, and the seller, but it's better than pretending nothing will break.
Third, we put the maintenance history in the purchase order. If the seller couldn't provide it, the deal changed. The seller of the $52,000 machine had no problem signing it.
The real lesson from this purchase wasn't 'buy the most expensive quote.' It was 'let the full cost structure, not just the list price, make the decision.'
The $38,000 machine might have worked out fine. I don't know. But the fact that I couldn't verify its history made it a gamble, and the odds weren't in our favor.
If you're looking at used Andritz centrifuges, don't worry only about whether you're getting a fair price. Worry about whether you're getting a true picture of what the machine is—and who's going to help you when it isn't.