What your radio spec sheet doesn’t tell you about closing a LEO downlink. A practical guide for satellite engineers from a team that has closed a lot of links.

You picked a radio for your LEO mission. The spec sheet says 5 watts output. You plugged that into your link budget, added the usual margin, and the numbers closed.

Then you tested it with your actual modulation scheme at the edge of your operating band, and the link came apart.

If that sounds familiar — or if you want to make sure it never has to — this article is for you. We’ll walk through the five things a radio spec sheet won’t tell you that can quietly affect a link, and what to do about them before you’re on the integration bench wondering where your margin went.

The short version

A radio is one element of your link budget. A spec sheet is a marketing document. And the “5 watts” you’re counting on is probably not 5 watts in the configuration you’re actually flying.

Five realities every RF design engineer should know before committing to a design:

1. Your radio is only a single part of your link budget. Antenna gain, cabling, connectors, amplifier behaviour, modulation and bandwidth, noise figure, frequency selection, and atmospheric losses all carry weight in the budget.
2. Spec sheets show min/max — not what happens in between. Gain, power, return loss, and most other parameters drift across frequency and temperature. Your design point is almost never at the spec sheet’s best-case corner.
3. Power output is usually quoted at the simplest modulation. A “5W” radio can be effectively a 1.5W radio once you’re running 32APSK DVB-S2.
4. Adding an amplifier doesn’t simplify the problem. It compounds it. You now have two sets of drift curves, two sets of marketing-optimized specs, and a matching problem between them.
5. Bidirectional amplifiers add a second optimization you can’t skip. Get the receive gain wrong and you’ll saturate the radio’s front end.

Let’s take them one at a time.

1. Your radio is only one part of your link budget

A link margin calculation adds up every gain and every loss between your transmitter and your ground station receiver. The radio is one line item on a long list.

Here’s what the rest of the list looks like on a typical LEO X-band downlink:

• Transmit cabling and connector losses
• Transmit antenna gain (and off-axis beam behavior)
• Modulation, signal bandwidth, and peak-to-average power ratio of the waveform
• Feed losses, polarization mismatch
• Free-space path loss (frequency-dependent, distance-dependent)
• Atmospheric absorption and rain fade margin
• Horizon mask and line-of-sight blockage at low elevation angles
• Ground station antenna gain and pointing error
• Ground station receive cabling and LNA noise figure
• Receiver thermal noise floor at your operating bandwidth
• Required demodulation threshold at your target BER

Miss any of these by a decibel or two and your margin evaporates. The radio spec sheet gives you one number — transmitter output power — and says nothing about how it interacts with all of the above.

The working discipline is simple. Build the link budget first. Put every element into a table with a best-case and a worst-case value. Look at the worst-case sum before you pick anything. Then choose a radio, an amplifier, an antenna, and a cable run that give you the margin you need at the edge of your operating envelope, not at the corner of a data sheet.

2. Spec sheets show min/max — not what happens in between

Every radio and every amplifier has a spec sheet full of numbers. Gain at some frequency. Power at some temperature. Return loss across some band. Noise figure at some test condition.

The problem: design doesn’t happen at one frequency and one temperature. Your radio operates across a dynamic range. If that range is more than about 10 MHz wide — and for most space links it’s a lot wider — gain and power change across the band. Sometimes significantly. Add temperature drift from a satellite that cycles through sun and shadow and the curve moves again.

A min/max snapshot is a corner. Your actual operating point is somewhere on the surface between those corners, and for most missions it’s closer to the worst corner than the best.

What to do: ask for the full performance curves, not the summary. Gain versus frequency. Power versus temperature. Return loss across the operating band. Any supplier who has nothing to give you beyond the spec sheet is either hiding the data or doesn’t have it. Both should be a concern.

3. “Power output” is usually quoted at the simplest modulation

This is the one that costs people the most margin, and it’s the hardest to catch by reading a spec sheet.

Radio spec sheets go through marketing before they go to customers. The “output power” number on the front page is almost always the saturated output using the simplest modulation scheme the radio supports. BPSK. Maybe QPSK. Something with a constant envelope and a low peak-to-average power ratio.

That number is real. It’s just not the number you’re going to fly.

The moment you move to a higher-order modulation — 8PSK, 16APSK, 32APSK, the DVB-S2 schemes that actually deliver the data rates LEO missions need — you have to back off from saturation to preserve linearity. That back-off is frequently 5 to 8 dB.

A “5 watt” radio at 32APSK DVB-S2 is effectively a 1.5 watt radio for link-budget purposes. A “25 watt” radio drops to roughly 8 watts.

If you built your link budget assuming the spec-sheet number, your margin is already gone before you start.

What to do: ask your radio vendor for the output power specification at the exact modulation and coding you plan to fly. If they can’t give it to you, ask for the output third-order intercept point (OIP3) and back-off curves so you can calculate it yourself. The specific phrase to use when you call: “What’s the linear output power at my actual modulation?”

4. Adding an amplifier doesn’t simplify the problem — it compounds it

Let’s say you’ve done all the above, the radio alone can’t close your link, and you need to add a power amplifier.

You now have a second component with all the same issues:

• Gain that drifts over frequency and temperature
• A spec sheet written by a marketing department
• A saturated power number that doesn’t describe behavior at your actual modulation
• Linearity that degrades as you approach compression

And a new problem: matching. The drive level the amplifier wants is almost certainly not the level the radio wants to output. If you over-drive the amplifier, you compress it and kill your linearity — which kills your higher-order modulation — which kills your data rate. If you under-drive it, you’re not getting the gain you specified it for and your link closes with no margin — or doesn’t close at all.

Noise is the other half of the matching problem. Every stage adds noise. An amplifier that’s marginally linear at your operating point will turn a clean 32APSK constellation into a fuzzy one, and your demodulator’s bit error rate will climb.

The fix is not picking the biggest amplifier on the catalog page. It’s selecting an amplifier designed to operate linearly at the drive level your radio actually produces, at your actual modulation, across your full operating temperature. That’s an engineering conversation, not a procurement one.

5. Bidirectional amplifiers add a second optimization you can’t skip

If your link is bidirectional — if your amplifier both transmits to the ground and receives from it — there is a second optimization that a lot of engineers either rush or skip entirely.

The receive side needs its own gain plan. If your amplifier’s receive gain is too high, you’ll saturate the radio’s receive front end. Too low and you lose signal into the noise floor. The window is narrower than most engineers expect, and it’s frequency-dependent.

On a unidirectional amp you only have to worry about what you’re transmitting. On a bidirectional one you have to optimize the transmit path and the receive path and the switching behavior between them, and verify that all three work across temperature and frequency without degrading one another.

This is usually where an off-the-shelf bidirectional amp will disappoint you: the transmit side is characterized to a spec sheet, and the receive side has a single gain number with no context.

So what do you do with all this?

Three practical moves, in order.

First — build your link budget with every element in it, and review worst-case sums before you pick any component. If the worst-case budget closes, you have a mission. If it closes only in the best case, you have a risk.

Second — stop trusting front-page spec-sheet numbers on radios and amplifiers. Ask for performance across frequency, across temperature, and at the specific modulation and coding you plan to fly. If a supplier can’t give you those curves, that tells you something important.

Third — when your link requires something the standard catalog doesn’t cover, work with someone who designs to your mission, not to their product line. A good RF partner will walk through the link budget with you, identify the specification gaps before you commit, and build the amplifier (or the bidirectional subsystem) that makes your actual modulation close with margin — at your actual operating conditions, not the corner of a data sheet.

That partnership matters more than any single number on any single spec sheet. A link that has to work the first time — on a launch window that doesn’t move — is not a catalog-shopping exercise. It’s an engineering conversation.

Want a second set of eyes on your link budget?

Tell us your frequency, target data rate, satellite altitude, and modulation scheme. We’ll walk through the budget with you and tell you what amplifier configuration actually closes it — and what we’d build if nothing off the shelf fits.

Or talk to an engineer — five inputs, one real conversation with the Triad RF team.

---

About the author: Pat Sherlock is the New Space Growth Lead at Triad RF Systems. He leads the company’s go-to-market strategy for LEO, SmallSat, and CubeSat RF amplifier customers and works directly with engineering teams on custom RF front ends for new space missions.

Triad RF Systems builds custom RF subsystems for LEO constellations, satellite IoT networks, and new space missions. 600+ payloads delivered. Flight heritage across UHF through Ka-band.