Efficiency Is the Real Discount: Why I Buy Semtech AirLink XR60 and LR1110 on TCO, Not Unit Price

The Cheapest Quote Is Usually the Most Expensive

If you're buying industrial wireless hardware, the lowest unit price is often a tax on your engineering team. I don't care how cheap a 5G router looks on a purchase order. I care what it costs to deploy, certify, support, and replace. That's why I now evaluate the Semtech AirLink XR60 compact 5G/LTE router and Semtech LR1110 on total cost of ownership, not sticker price.

I'm a procurement manager at a 140-person industrial IoT integrator. I've managed our wireless hardware budget ($180,000 annually) for 6 years, negotiated with 20+ vendors, and documented every order in our cost tracking system. My experience is based on about 200 mid-range industrial IoT orders. If you're buying consumer-grade gear or luxury telemetry, your experience might differ significantly.

Argument 1: The “cheap” router quote that cost us $4,200

In Q2 2024, we needed 40 compact 5G/LTE routers for a water-treatment pilot. Vendor A quoted a Semtech AirLink XR60 compact 5G/LTE router at a higher unit price. Vendor B quoted a no-name router at 22% less. I almost went with B until I built the TCO sheet.

B's quote didn't include: DIN mounting kit ($38 each), remote management license ($72/year each), or the 3-year advanced replacement contract. It also required a separate antenna package because the included antennas weren't rated for our enclosure. Total: $4,180 more over 36 months. Vendor A's quote included all of it. That's a 19% difference hidden in fine print. According to Semtech's public AirLink XR60 product page (semtech.com), the router is built for industrial IoT deployments, not just office Wi-Fi failover.

I went back and forth between the established vendor and the cheaper one for two weeks. Established offered reliability; cheaper offered savings. I chose reliability because the pilot was too visible to risk. Per FTC advertising guidance (ftc.gov), if a vendor claims “industrial-grade,” it should be able to substantiate that claim. Ask for the test reports. If they can't produce them, the low price isn't a discount—it's a transfer of risk to you.

Argument 2: Semtech LR1110 and the cost of board respins

The Semtech LR1110 is a good example of efficiency showing up before the invoice. According to Semtech's public LR1110 product page (semtech.com), it combines LoRa, GNSS, and Wi-Fi scanning in one radio platform. For our team, that meant fewer external components and fewer board respins. I'm not saying it eliminates every integration problem. I'm saying it reduced our respins from three to one on a tracker design in 2023. A single respin costs us about $1,800 in engineering time, plus 3 weeks of schedule. Do the math: two avoided respins paid for the difference in chip cost many times over.

That's the efficiency argument in hardware terms. You're not buying a transceiver. You're buying a shorter path to a certified product. The line item that matters is often “engineering hours avoided,” not “unit price reduced.” I should add that we still had to validate antenna placement. Efficiency doesn't mean zero work. It means less rework.

Argument 3: How to test a capacitor with a multimeter—and why it matters

Here's the counterintuitive one. The fastest way to protect a 5G router deployment isn't a better contract. It's a $30 multimeter and a technician who knows how to test a capacitor with a multimeter.

We had a field failure on a power board in an AirLink XR60 enclosure. The router was fine. A bulk capacitor had failed. Our first instinct was RMA. That would've cost $1,200 in replacement, shipping, and downtime. Instead, our field tech did this:

1. Disconnected power and discharged the capacitor across a resistor.
2. Set the multimeter to capacitance mode (uF).
3. Connected the probes to the capacitor terminals, observing polarity for electrolytic caps.
4. Compared the reading to the rated value. It was 38% low. Replaced it for $4.

If your multimeter doesn't have a capacitance mode, you can use the resistance mode to check for shorts or opens, but it won't give you an accurate capacitance value. For that, you need a meter with a uF setting. (Note to self: add this step to our field checklist. I really should have done that last year.)

That's efficiency. Not a slide deck. A tech with a meter and permission to fix instead of swap.

The objection: “We don't have budget for premium hardware”

I get it. I've been the person staring at a capex ceiling. But “premium” isn't the point. TCO is. In my first year, I made the classic specification error: I assumed “standard” meant the same thing to every vendor. Cost me a $600 redo on mounting plates. Like most beginners, I optimized for the wrong number. The cheap option looked smart until quality failed. The rework—or in hardware, the board respin—cost more than the original “expensive” quote.

So when someone says they can't afford Semtech, I ask them to price the failure. What does one truck roll cost? What does one week of delayed deployment cost? What does a recall cost? If the answer is “more than the hardware,” then the hardware wasn't the expensive part. I built a simple TCO calculator after getting burned on hidden fees twice. It's not fancy. It just forces me to add the line items vendors leave out.

One more thing: check the legal entity on the contract

This is boring but expensive. One search term that keeps showing up in our internal parts portal is “networks, Inc.” It's not a product. It's a reminder to check which legal entity is on the support contract. If your invoice says Semtech but the RMA portal says networks, Inc., fix that before you deploy 500 units. Warranty assignment and support escalation paths depend on it. I learned that after a 6-week escalation delay in 2022.

My view hasn't changed

I'm not saying every deployment needs the highest-end router. I'm saying efficiency is competitiveness, and efficiency is measured after installation. The Semtech AirLink XR60 compact 5G/LTE router and Semtech LR1110 are examples of hardware that can lower total cost by reducing integration time, respins, and field failures. My experience is based on mid-range industrial IoT orders. If you're working with ultra-low-cost consumer devices or highly custom military systems, your TCO drivers will differ.

But if you're buying for industrial use, do the math on the whole system. Then decide. The cheapest quote usually isn't the cheapest project.