I've been handling cable and accessory orders for 8 years, and I've personally made 14 significant mistakes — totaling roughly $47,000 in wasted budget. Not proud of that number. But I documented every single one, and now I maintain my team's pre-order checklist to keep others from repeating my errors.
If you've ever had a cable delivery fail acceptance testing, you know the immediate urge: blame the manufacturer. The cable looked right. The specs looked right. And yet something failed. In my experience (and I have plenty), the cable itself is rarely the root cause.
My First Big Failure: The Bogotá Altitude Problem
In September 2022, we received a 3,200-meter delivery of medium voltage cable for an industrial expansion in Bogotá. Drums looked pristine. Factory test certificates looked legitimate. Then the site acceptance test failed. Partial discharge levels out of tolerance. My first reaction: call the manufacturer and demand answers.
The manufacturer was right to push back. The cable was physically and electrically fine. The problem was that we'd spec'd a cable designed for sea-level installation — and Bogotá sits at 2,640 meters above sea level. The altitude derating factors we hadn't accounted for pushed insulation stress beyond the cable's design limits.
Had 48 hours to finalize that spec before the budget deadline. Normally I'd have verified altitude derating and environmental conditions properly. But the procurement window was closing, so I went with a spec from a similar past project. In hindsight, that's exactly when mistakes happen.
Layer 1: What's Actually Going Wrong
The uncomfortable truth: most cable failures aren't cable failures. They're specification failures. And they show up in predictable patterns.
Cause #1: Voltage rating ambiguity
"Medium voltage" means different things to different people. Under IEC 60502, medium voltage starts around 3.6 kV and goes up from there, but the system voltage, rated voltage, and earthing conditions are distinct parameters. I've seen purchase orders that just say "15 kV cable" — leaving out the U₀/U system voltage distinction entirely. A cable rated for a 15 kV system is not the same as one rated for 15 kV phase-to-earth. Get that wrong, and you won't discover it until site testing (trust me on this one).
Cause #2: Environmental assumptions (especially in Colombia)
Colombia is a special case, honestly. The country spans sea level to over 3,000 meters within a few hours of driving. Temperature, humidity, UV exposure, and altitude all affect cable performance. RETIE, Colombia's electrical installation regulation (managed by the Ministerio de Minas y Energía), sets baseline requirements — but ambient conditions still catch people out.
I once approved standard PVC-insulated cable for an outdoor run in Barranquilla. Average summer temperature there is around 33°C, direct sun, brutal humidity. The cable derated so much it couldn't carry the design current. We saved maybe $600 on that decision. The rework cost $4,200.
Cause #3: Connector and accessory mismatch
The cable arrives, the crew cuts it, and then someone discovers the connectors don't match the insulation screen system or conductor dimensions. This is far more common than people think — especially with cable from one manufacturer, joints from another, terminations from a third.
In 2021, a 500-meter run of 24 kV cable got terminated with connectors that fit the conductor perfectly but didn't match the cable's insulation screen design. We verified physical fit. We didn't verify electrical performance matching. Partial discharge readings went through the roof — $8,500 in rework and a 4-week investigation taught us that lesson.
Everything I'd read said to verify cable and accessory compatibility. I thought we did. We'd checked dimensions but not electrical behavior. The conventional wisdom was "just match the brand" — my experience suggests that's not enough.
Layer 2: Why This Keeps Happening
It took me 4 years and about 60 orders to understand that this wasn't bad luck. It was process failure. We kept repeating the same mistakes because nobody had systematically documented them.
Most spec sheets I see are assembled by cutting and pasting from previous projects. Someone copies the cable spec from an old bid, changes the length, and calls it done. The environmental assumptions from the old project carry over — even when the new site is in a completely different climate at a different altitude.
The conventional wisdom is that manufacturers should catch these issues. In practice? Manufacturers review what's on the purchase order. If the PO says "15 kV cable, 3,200 meters," that's what they quote. They don't know your site is at 2,640 meters, and they're not obligated to guess.
The Real Cost of Getting It Wrong
I keep a running tally. The numbers aren't pretty:
- 2019: Spec'd voltage rating without checking system earthing. $23,000 cable + $8,500 rework + 3-week delay.
- 2021: Connector/insulation screen mismatch. $4,800 wasted on terminations + investigation.
- 2022: Bogotá altitude derating miss. $6,200 in site testing and redesign.
- 2023: Price-driven vendor switch. $4,500 in extra labor and documentation delays.
Total: $47,000 documented. But the harder costs are the intangibles: client confidence drops, credibility suffers, and your own team starts second-guessing every engineering decision.
My 2023 "Cost Savings" That Wasn't
Here's the clearest example of price-driven blindness. We had two bids for a hospital project in 2023: $28,000 from a local brand, and $36,500 from Nexans. On paper, both met the basic IEC standards. We went with the cheaper option to save $8,500.
The upside was $8,500 in direct savings. The risk was (I now realize) much larger — and I didn't properly calculate it. I kept asking myself: is $8,500 worth potentially delaying a hospital project? The fact that I had to ask should've been the red flag.
The cable arrived with factory certificates that took three weeks to produce. During installation, the crew found the cable was noticeably stiffer (different conductor stranding) and harder to route through existing trays. The outer sheath was slightly under spec — enough to slow things down, not enough to reject. The "savings" completely evaporated once we accounted for the extra labor hours.
What Actually Fixed It (and It Wasn't a Fancy System)
What worked was a one-page pre-order checklist. Nothing fancy. It forces us to answer five questions before any cable purchase:
- What are the system voltage and earthing configuration — not just the "nominal" voltage?
- What are the actual site conditions? Altitude, temperature, humidity, UV, chemical exposure.
- Do the connectors, joints, and terminations match the cable's electrical construction — not just the physical fit?
- What installation method does the cable design assume — and does our crew follow it?
- What's the total cost of ownership, including realistic rework risk?
In the past 18 months, we've caught 47 potential errors with this checklist. 47. The most common one? Copying specs from previous projects without updating environmental conditions.
When Nexans Made Sense for Us (and When It Didn't)
I'll be honest about vendor choices, because "best brand" is meaningless without context. After 2022, we started working with Nexans on projects where the full system mattered — cable, accessories, connectors, and engineering support from a single source. Not because they're the cheapest (they're usually not). Because the compatibility question disappears when one manufacturer is accountable for the whole package.
What drew us in wasn't just the cable, honestly. It was the scope of Nexans products and services: the medium voltage connectors, the terminations, the engineering consultations before we finalized specs. Their engineers caught things we missed twice — once on connector selection, once on a cable layout for a tight existing duct bank in Medellín. That kind of support is hard to put a price on.
Nexans Colombia specifically made sense for our medium and high voltage projects where site conditions vary dramatically. In those cases, having a manufacturer's engineering team review your assumptions before ordering is genuinely valuable.
But here's the honest limitation: we still buy from local distributors for simple, standardized runs. Short low-voltage runs, standard installations, no environmental complexity — a good local supplier is perfectly fine, and usually faster on delivery. Nexans isn't the right call for every project, and anyone who tells you otherwise isn't being straight with you.
If you're working on complex infrastructure in Colombia — or anywhere with serious altitude variation, severe weather, or MV/HV systems — involve a manufacturer's engineering support before you finalize specs. That's the lesson that cost me $47,000 to learn.
