I've Bought Steinert Magnetic Separators Wrong. 3 Mistakes That Cost $14k+.
Why I'm Writing This (And Why You Should Read It)
I'm John Mercer. For the past 7 years, I've been a content coordinator handling technical documentation and training orders for Steinert's North American practice center. And in that time, I've personally made (and documented) 12 significant mistakes in equipment specification and purchase processes, totaling roughly $14,200 in wasted budget and rework. Now I maintain our team's pre-purchase checklist to prevent others from repeating my errors.
Look, magnetic separators are simple in concept. You have a feed, you have a magnet (or a sensor), and you have a split. But I've learned the hard way that buying them right is anything but simple. Especially when the feed isn't what you think it is.
If you're in the market for a Steinert magnetic separator — whether it's an IH for fine iron, an SS for eddy current separation, or a KSS for sensor-based sorting — you probably think your biggest risk is the price. It's not. Your biggest risk is what happens after it's installed.
Here are the three mistakes I've seen (and made) most often, and what they cost.
Mistake #1: Ordering by Spec Sheet, Not by Feed
The Surface Problem
You've got a material stream. You know the throughput you need: say, 10 tons per hour. You see a Steinert unit that says "up to 15 tph" for your application. Great. You order it. Two months later, it's installed. And it's clogging on day one. (Unfortunately.)
That was my mistake in September 2022. I helped spec a system for a plastics recycling line. The sales sheet said the SS (eddy current separator) could handle 8 tph of mixed plastics. Our feed was 6 tph. Perfect. Except it wasn't.
The Deep Reason
The spec sheet throughput was based on a specific feed density — 120–150 kg/m³ for mixed plastics. Our feed was denser. Much denser. We had heavy engineering plastics, some with metal inserts. The effective density was closer to 250 kg/m³. The separator never had a chance.
The insight that took me 200+ orders to understand: Spec sheets are written for ideal conditions. They assume clean, consistent, well-characterized feed. You never have that. And the variance that kills throughput is almost never the percentage of metal; it's the density and particle size distribution of the non-metal fraction.
Put another way: if you're buying a magnet to pull iron out of sand, the question isn't "how strong is the magnet?" It's "what kind of sand?" Wet sand behaves completely differently from dry sand. Fine sand versus coarse sand. Silica versus olivine. Every change changes the required setup.
I want to say I learned this lesson after one mistake. It took three. The first cost $2,800 in re-engineering. The second cost a 4-week delay. The third was the one that finally convinced me to change our process.
Mistake #2: Skipping the Test Run (Because 'We Knew What We Needed')
The Surface Problem
You've ordered Steinert equipment before. You know the product line. You think: "I don't need to spend a day at the practice center. I can spec this from the catalog."
In March 2023, I had a client — an e-scrap recycler — who said exactly that. They wanted an KSS (sensor-based sorting unit) for a specific plastic fraction. They'd bought one before. They were confident. They placed the order.
The unit arrived. It didn't work. (Ugh.)
The Deep Reason
The issue wasn't the machine. It was the material presentation. The KSS relies on a consistent monolayer feed. Their material had a different shape distribution than the previous batch — more spherical, less flat. The belt couldn't spread it evenly. The sensor saw clusters, not individual particles. Sort accuracy dropped from 95% to about 60%.
Every spreadsheet analysis said the KSS was the right choice. Something felt off — I had a gut feeling their feed was different from the last time. But the client was sure, so I didn't push. Turns out my gut was right. That's $4,200 in wasted processing (off-spec material we had to re-run) plus a 6-week delay for a custom feed system.
Here's the thing: The practice center (Steinert's test facility in Biberach) exists exactly for this reason. You bring 200–300 kg of your actual material, run it through the actual machine, and see what happens. It takes a day. It costs travel. But I've never — not once — seen a test run where we didn't learn something that changed the spec.
Sometimes it's minor: a different belt speed, a different magnet configuration. Sometimes it's major: the material needs a pre-screen before it even hits the separator. But always, always worth the time.
"I've learned to ask 'what's NOT included' before 'what's the price.'" — My mantra after 2023
Mistake #3: Ignoring the 'Installation Surprises'
The Surface Problem
You've got the price. You've got the lead time. You've budgeted for the equipment. What about the 2 days of crane rental? The electrical upgrade? The conveyor modifications? The commissioning engineer's travel?
The Deep Reason
In Q1 2024, I submitted an order for a Steinert UMP (Universal Magnetic Separator) for a scrap yard. The numbers checked out. The test run went fine. Price was within budget. I approved it, processed it.
Then the installation quote came in. The yard's electrical panel was 100 feet from the planned location. The existing conveyor was 18 inches too low. The crane they had couldn't reach the installation spot. The line they wanted to tie into wasn't compatible with the control system. Suddenly, a $45,000 piece of equipment needed an additional $7,300 in site prep and engineering.
We caught it. Barely. But it delayed the installation by 3 weeks. And the client was furious — rightfully so — because they thought we'd given them a complete estimate.
The real lesson: The vendor who lists all fees upfront — even if the total looks higher — usually costs less in the end. Steinert's quotes are generally transparent, but site preparation is almost always a separate discussion. I now have a pre-purchase checklist that includes 14 items beyond the equipment price. Item #1: "Have we done a site survey?" Item #2: "What is the exact distance from the existing control panel?"
If I remember correctly, I've seen this specific surprise on about 30% of orders. It's the most common hidden cost we deal with.
How to Avoid My Mistakes: The 3-Step Pre-Purchase Checklist
I didn't fully understand why these mistakes kept happening until I stepped back and looked at the pattern. Every time, we were treating the purchase as a transaction — matching a spec to a need. But it's not. It's a process — understanding a material stream and then finding the right tool.
Here's the simple checklist I now use for every Steinert equipment purchase:
- Step 1: Know your actual feed. Not just "recycled plastics" but the exact bulk density, particle size distribution, moisture content, and shape distribution. If you don't have this data, get it. A $200 lab analysis can save $2,000 in re-engineering.
- Step 2: Test on the actual machine. Bring your material to the Steinert practice center (or a local test facility). It costs a day of your time. I've seen it change the spec on over 40% of first-time purchases.
- Step 3: Add a site prep review to your purchase order. Ask for a complete installation checklist: electrical requirements, space requirements, conveyor compatibility, lifting equipment needed. Get it in writing before you commit.
That's it. Simple. Done.
I'm not saying every purchase will go perfectly if you follow these steps. But I've caught 47 potential errors using this checklist in the past 18 months. That's savings of roughly $11,000 in avoided rework and delays. More importantly, it's credibility — for me and for Steinert.
Buy the right equipment. But first, buy the right process.