The Day I Learned ‘Same Specs’ Isn’t the Same Thing
I‘m a project manager handling large-scale hydro and process equipment orders for Andritz. I’ve been doing this for about 10 years now. In my first year (2017), I made a classic mistake that still haunts me when I see certain project files.
It was a Wednesday in September 2022. A project for a new run-of-river hydropower plant in Latin America. We had the contract for the main turbine assembly and the governor system. The client, a well-known power utility, had their own specifications—thick documents, lots of international standards. Everything looked solid.
The problem started with a simple assumption: “Same specifications” meant identical results across our global supply chain.
It didn’t.
The Assumption That Cost Us a Week
We sourced a critical sub-assembly—the guide bearing—from a secondary supplier we’d qualified earlier that year. Their spec sheet matched perfectly. I checked it myself, approved it, and told my team we were good to go. (Honestly, I was patting myself on the back for saving a couple of weeks on lead time.)
Turned out each vendor (Andritz’s own workshops, our suppliers) had slightly different interpretations of the “same” tolerance. The bearing from the new supplier was within spec—technically—but it didn‘t fit the shaft seal assembly we’d built in-house. The clearance was 0.2mm off. Doesn‘t sound like much, right? On a turbine spinning at 300 RPM, that’s a vibration risk we couldn't take.
We caught it during a routine pre-assembly check in our workshop (thanks to an alert technician who’d seen this before). The result: a 1-week rework delay, $12,000 in wasted machining hours, and a very uncomfortable meeting with the client’s chief engineer. I had to explain that the spec was correct, but the execution was inconsistent.
Learned never to assume “same specifications” means identical results across vendors after that incident. That's when I created our internal pre-check checklist for incoming supplier components.
The Hidden Cost of ‘Good Enough’
This experience taught me something deeper about industrial procurement. The total cost of ownership isn’t just the base price—it’s the cost of verification, rework, and lost time. We started tracking these hidden costs after that project.
In Q3 2023, we audited 10 critical component suppliers against our own master tolerance database. The variation was shocking. Parts were all within the printed spec, but the combined stack-up tolerance was often 15-20% wider than what our internal design assumed. That's a recipe for field failures.
Here’s what we now do:
- Pre-shipment fit checks: Every critical sub-assembly gets a dry fit in our workshop before release.
- Supplier tolerance audits: We sample incoming lots and compare actual distributions to nominal specs.
- A master tolerance database: Internal, not shared externally, that accounts for real-world supplier variation.
The Myth of ‘Global Standards’
This is where I see a common misconception in the industry. Many buyers assume that if a supplier follows ISO or IEC standards, the parts are interchangeable. That was true 20 years ago when manufacturing tolerances were looser (and everyone had bigger margins for error). Today, with tighter designs and higher efficiency targets, the margin for error is much smaller.
The myth of “global standards” comes from an era when a 0.5mm clearance was fine. That’s changed. Modern hydro turbines run at peak efficiency with clearances measured in microns. A component that’s “within spec” but at the wrong end of the tolerance band can kill performance.
Honestly, I'm not sure why some suppliers consistently hit the nominal target while others drift to the edge of the allowed range. My best guess is it comes down to their internal quality culture—whether they're aiming for “just good enough” or genuinely trying to hit the middle of the band.
What I‘d Do Differently (and What We Fixed)
If I could go back to 2017, I’d tell myself: the spec sheet is the start, not the end. Verification is not optional. And the supplier who says “we‘ll send you the same part as last time” deserves a skeptical follow-up question.
We’ve caught 47 potential errors using my checklist in the past 18 months. 47. That‘s 47 delays, reworks, or field issues we avoided. The ROI on that checklist is absurd—it paid for itself in the first three months.
On a recent project for a pumped-storage hydro plant in Europe, we used the checklist and found a bearing housing alignment issue at the supplier’s shop, before shipment. Saved about $45,000 in potential field rework and a 3-week installation delay. The client never even knew there was a problem. (Which, honestly, is how it should work.)
The Takeaway: Specialization Over Generalization
This whole experience reinforced a view I've held for a while: I‘d rather work with a specialist who knows their limits than a generalist who overpromises. When a supplier tells me, “This isn’t our core strength—here‘s who does it better,” I trust them for everything else.
Andritz’s strength is integrating complex systems across energy and mining. We’re not a bearing manufacturer, we‘re a turbine and process equipment integrator. Knowing when to rely on our own global workshops and when to partner with specialized vendors is the difference between a smooth project and a costly fix.
“The vendor who said ’this isn't our strength—here‘s who does it better’ earned my trust for everything else.”
If you’re sourcing critical equipment, don't just compare spec sheets. Ask for proof. Ask for tolerance distributions. Ask what happens if two parts from two suppliers don‘t fit. Because that question will tell you more about the vendor’s engineering culture than any ISO certificate.
Prices are for reference only (based on Andritz project data, Q3 2024; verify current rates with your local team). Regulatory information is for general guidance—consult official sources for current project requirements.