Why I Stopped Picking Danfoss VFD and Relay Suppliers on Unit Price Alone

I'm the procurement manager at a 240-person systems integration company. I own our automation components budget — about $328,000 a year — and I've spent the past 8 years negotiating with more than 40 vendors across 1,200+ orders in our cost tracking system.

Here's the number that made me rebuild our entire supplier scorecard: 6 out of every 10 dollars we labeled as "budget overruns" didn't come from price hikes. They came from downtime, rework, expedited shipping, wrong parts, and hidden fees we never saw at quote time.

So let me save you the audit: if you're buying Danfoss VFDs, PLCs, relays, or timers, and you're comparing unit prices, you're leaving money on the table — and the money is walking out the back door where you can't see it.

How I got those numbers

In Q4 2023, I ran an audit on everything in our ERP — purchase orders, freight invoices, RMAs, return records, maintenance tickets. Three years of data. Around 460 transactions — maybe 450, I'd have to check the system — across 14 vendors. Not a huge sample, but it matches our actual operations.

Here's where TCO landed for us, averaged by vendor:

• Unit purchase price (including inbound freight): about 43% of total cost
• Expedite and rush surcharges: about 14%
• Downtime from late or wrong deliveries: about 26%
• Returns, rework, and replacements: about 11%
• Poor documentation that caused downstream issues: about 6%

So the line item everyone compares — unit price — accounts for less than half the money that actually left our account.

What a Danfoss VFD alarm 13 actually costs

In Q2 2024, we had a line running for about six weeks when it started throwing alarm 13. If you're not familiar — that's a DC bus overvoltage fault on Danfoss VLT drives, usually tied to deceleration, braking, or input voltage. It's a diagnostic code, not a broken part.

The VFD we spec'd came from the lowest-price vendor in a 6-quote batch. Saved us about $380. We went with it.

Then the drive alarmed. I called them. Their tech support was out — "back tomorrow." When we got someone, it took two more calls before anyone could explain that the fix wasn't in the drive itself.

Two half-days of two engineers on site, with the customer's line down. When I tallied it up, that single alarm cost us north of $4,300 in downtime and labor. The $380 we saved on the drive rounded to nothing.

Here's the part I got wrong: I thought I was buying a box with specs. I was actually buying the response time behind it. A VFD is only as good as the person who picks up when it faults at 2 a.m.

PLC catalogs and relay suppliers — you're buying documentation, not hardware

We don't buy relays by spec sheet anymore. What actually separates vendors:

Does the PLC catalog list firmware revisions, I/O mapping, and wiring per model — or just a download link? Are contact ratings given as a single number, or broken out by load type and cycle life? A spec that says "10A" is not the same as one that says "10A at 250VAC, resistive, 100,000 cycles."

Same logic with relay suppliers. When I'm evaluating one, here's what I actually look for:

• Is contact life data tested at rated load, or in ideal lab conditions?
• Do different ratings in the same series share a housing and pinout — or does swapping a model mean rewiring?
• How are MOQs and lead times quoted — standard price list, or "call us"?
• What's their RMA policy when a part fails in the field, versus when a part fails on the shelf?

None of that shows up on a quote. But over a three-year production cycle, it matters more than an 8% price gap.

The timer manufacturer that changed how I evaluate vendors

Timers are where I got surprised. Our assumption had always been that the premium brand would be more accurate, and the cheap one would be, well, cheap.

We bought a batch from a budget timer manufacturer. Not for anything precision-critical — a high-volume assembly station where a couple hundred milliseconds of drift wouldn't kill us. Their pricing was about 30% lower than the brand we'd been using.

The surprise wasn't the accuracy. It was how they handled our first quality complaint.

We sent an email. Three days, no reply. We called. "Check your wiring" — that was their answer. Our wiring was fine. It turned out to be an internal batch problem on their end. And it took them almost three weeks to figure that out, because they apparently couldn't trace their own production lots.

We had said "standard lead time." They had heard "whenever it ships." We discovered that when the replacement units showed up 11 days late — and the customer's install date was already locked in.

We switched to a different supplier. About 7% higher unit price. But in three years, they've never taken more than 24 hours to respond, they flag batch issues proactively, and they ship test data per lot number with every order.

That's not "cheap is bad." That's "we now price in how fast a vendor can solve a problem when it's their fault."

The industry moved. Our scorecard had to move with it.

Five years ago, our supplier scorecard weighted unit price at 60%, lead time at 20%, and quality history at 20%. Today, unit price is 30%, lead time is 20%, and the other 50% goes to tech support, documentation quality, and problem response.

Not because we got bigger. Because the lines we support got more complex — VFDs, PLCs, relays, and timers are all connected now — and the cost of a single slow link has gone up more than the cost of any single part.

The principle still holds: you can't save your way to a good supplier. But how you measure "good" has shifted, and it's not going back.

When picking the cheaper supplier is actually the right call

None of this means chase TCO on every purchase. That's a mistake too.

Commodity spares, drop-in replacements, short-cycle builds — if the brand and model match and nothing custom is involved, sure, take the lowest price. Documentation barely matters, you can call the OEM for support, and lead time is within tolerance. The TCO gap is small and not worth your time.

But if you're commissioning a line where the customer measures downtime in dollars per hour — stop comparing unit prices. Compare response times and documentation first. Then compare price.

This has been my experience, and my sample size is one company. It might not fit every industry or every budget. But if you've ever run a TCO split on your own vendor data, I'd bet the costs hiding outside the quote are eating more than half your budget too.