I Bought a Used Steinert Separator and It Cost More Than a New One: A $58,700 Lesson
Let me start with a number: $88,000. That was the asking price for a used Steinert sensor-based separator I bought in March 2021. The same model, based on a new quote we got in early 2021, was around $140,000. My quick math said I was saving $52,000.
I've been handling used equipment orders for our recycling operation for eight years. I've personally made—and documented—nine significant mistakes, totaling roughly $120,000 in wasted budget. This deal was not going to be mistake number ten. It was. I just didn't know it yet.
The surface problem: a bargain that makes sense
In our industry, 'Steinert' is one of those names that makes you relax. German engineering. Magnetic separation, sensor-based sorting, eddy current systems—they've been at it for decades. When you see a used unit at nearly half the replacement cost, your brain starts celebrating before your spreadsheet has finished loading.
The seller had photos, a maintenance record binder, and a way of talking that made the machine sound like it had spent its life in a museum. The price was negotiable, which felt like a gift. I negotiated another $3,000 off. That $3,000 would become a very expensive win.
The deeper issue: serial numbers, history, and parts stock
Our senior engineer, Dr. Robert Steinert—no relation to the company—moved to our team from Dresden in 2019. Before he worked in mineral processing, he spent years restoring old pianos. One day, while we were standing in front of the new-used machine, he told me something I should have needed to hear only once.
When I see a Steinert piano, the serial number is the first thing I check. It tells me the real age, the factory, and whether parts are still in stock anywhere. Industrial machines are not that different. The nameplate is a starting point, not a guarantee.
I didn't check the serial number. I didn't call the manufacturer. I looked at the year in the listing ('2018 model, low hours') and accepted it. When I finally did check—after the first breakdown—the actual production date was 2014. Maybe the seller believed it was a 2018. Maybe the data plate had been switched at some point. I'm not here to accuse anyone. The point is that the difference between 2014 and 2018 mattered more than I expected, mostly because of spare parts.
That leads to the second thing I missed: parts stock. For a current model, wearable items like sensor boards, belts, and control modules are easy to order. For a machine that's six years older than you thought, the manufacturer's stock gets thin. Some components are discontinued. Third-party suppliers exist, but they charge accordingly and don't offer the same support.
And the third thing? Performance assumptions. A Steinert separator is not a generic machine that works the same for every feed. Recovery rates and sorting quality depend on material size, composition, moisture, and contamination. I assumed that because the machine had worked at its previous site, it would work at ours. The fundamental physics were different.
What the inaction actually cost
The first sensor failure appeared after about 40 operating hours. The machine would sort, then suddenly skip a series of objects, then resume. Our maintenance team swapped connections, rebooted the control system, and cleaned the sensor window. No change. The service call alone was $3,800.
The diagnosis: a failed sensor board. No, wait—it was actually the processing unit, but the quote for the replacement came back as one assembly. $9,200. Not in stock at the manufacturer. Two weeks lead time. We needed it in five days. The expedited freight from a third-party supplier cost $4,100. The line sat idle for part of that time, and the throughput loss came to about $31,000. Our own electrician added another $5,000 in overtime. When I added it up: $52,000 saving, $53,100 in extras on the conservative side. My final spreadsheet, including the service contract we bought afterward, says $58,700. Either way, the 'bargain' Steinert cost approximately as much as a new one, without a warranty, without commissioning support, and with a maintenance team that stopped speaking to me for a week.
I want to be clear: I'm not saying every used Steinert is a bad purchase. I've bought three more since then. But that first one taught me the difference between a price and a total cost.
Why the cheapest quote usually isn't
In my experience, the lowest quote has cost us more in about 60% of cases. That's not because the vendors lie. It's because purchase price is only the first page of a very long invoice. Freight, installation, commissioning, downtime, spare parts availability, and eventual resale value all belong in the same equation. Most of those costs are invisible until the machine is on your floor.
Per FTC advertising guidelines (ftc.gov), specific performance claims like '99% recovery' need to be substantiated. If a seller says a used machine will hit a certain rate, ask for test reports—not just testimonials. The same discipline should apply to your own due diligence. If the seller can't produce a service history, don't fill the gap with assumptions. A black hole of information is not a data point.
The short fix: a three-item checklist
After that purchase, I put together a pre-purchase checklist. It's deliberately short, because long checklists get ignored. Three items, in order:
- Verify the serial number with the manufacturer before you pay anything. One email. It tells you the actual build year, original configuration, and whether service parts are still supported. Do this even if the machine looks spotless.
- Ask for price and lead time on the top ten wearable parts. Not 'are parts available?' Ask for a quote. If the seller can't get you a quote for a sensor board or a belt, assume replacement will hurt.
- Run a sample of your material through the machine before closing. If that's not possible, get a written performance guarantee tied to your feed specifications. A nameplate doesn't know what your material looks like.
Using this checklist over the past 18 months, we've flagged 47 potential issues—some as simple as a serial number that didn't match the paperwork, others as serious as a main component that had been discontinued. Every one of those issues was cheaper to fix before the purchase than after.
Look, I'm not saying cheap equipment is always a mistake. I'm saying cheap equipment needs to prove itself before you let it into your plant. And the only way it can prove itself is through serial numbers, test reports, and parts quotes. The work is boring. That's why so few people do it.
One of the younger guys in the plant asked me the other day how to get 'the Wise' in Blooket. I told him: you answer enough questions, you stop making the same mistakes, and eventually you upgrade. That's not a bad description of buying used industrial equipment either. You don't get wise by finding a cheat code. You get wise by checking the serial number first.