What’s the INPUT IMPEDANCE of the TinySA Ultra? Find Out!

Educational teardown of TheSmokinApe Ham Radio

▶ 00:00 — The “panic” is… literally the published spec

What’s framed as “internet panic” is actually the manufacturer telling you a simple condition: the TinySA input behaves like a proper 50 Ω termination only in certain front-end settings. That’s not drama, it’s operating limits.

If the manual says “50 Ω when attenuation ≥ 10 dB” and your response is “people are panicking,” you’re not debunking myths… you’re debunking reading comprehension.

▶ 01:15 — “The mixer needs to see 50 Ω”… then source vs load gets swapped

The practical definition is boring (and that’s why it matters): in a 50 Ω measurement system, the analyzer input is the load. If the load isn’t ~50 Ω, then your assumed pad values, levels, and calibration chain can drift.

Yes, internally mixers and filters have nominal impedances. But the real-world question is: does the port behave like 50 Ω at the connector plane, across frequency, in the mode you’re using? If not, “30 dB” may not be 30 dB anymore (and your amplitude accuracy pays the bill).

▶ 01:30 — “With lots of attenuation, mismatch is a moot point” (close… then wrong conclusion)

A pad does improve the match seen by the source, because reflections get attenuated on the way out and back. That’s why pads are used as “match improvers” in RF labs.

But the conclusion “therefore the input is basically 50 Ω” doesn’t follow. Pads reduce the consequences of mismatch; they don’t rewrite the load’s impedance. You can still have level error, frequency-dependent ripple, and calibration drift when the termination isn’t controlled.

Saying “attenuation makes mismatch moot” is like saying “a seatbelt makes crashes moot.” No… it just reduces how hard your face meets physics.

▶ 03:20 — The setup quietly stops being an input-impedance test

The video drifts from “is the input 50 Ω?” into a different (valid!) experiment: “do spurs change when I change attenuation?” That’s a classic overload / internal-artifact diagnostic.

The problem is the implied conclusion. Spur behavior when you step attenuation says a lot about dynamic range and front-end nonlinearity. It does not directly prove the connector-plane impedance is 50 Ω in a given mode.

Congratulations, you discovered that spectrum analyzers get nonlinear when you hit them too hard. That’s not “input impedance.” That’s Chapter 1 of “don’t overdrive the mixer.”

▶ 06:20 — “Those are likely IMD3 artifacts” (with one tone…)

“IMD3” is not the correct label for every mystery peak. Intermodulation distortion is fundamentally a multi-tone mixing story. With a single swept tone, your suspects are usually: harmonics, images, mixer products, internal spurs, or source impurities.

The robust method is not “it disappeared, so it must be X.” The robust method is: step attenuation (and/or input level) and check whether the spur tracks the main signal like a real signal, or behaves like an internally generated product.

Calling every animal “a dog” works until you try to leash a squirrel.

▶ 08:50 — “Not meeting the 10 dB condition”… then testing one narrow VHF slice

Even if the measurement were clean (it isn’t), “I didn’t see much change around 145–148 MHz” does not disprove a conditional spec. A condition is not a promise of a dramatic effect at every frequency point.

On top of that, TinySA behavior depends on mode / input path (low input vs high input, pad state, LNA state). So one narrow slice of VHF is not a verdict. It’s a single data point with a lot of hidden configuration behind it.

Measuring ocean depth in a puddle and declaring “the sea is shallow.”

▶ 10:03 — “I set the input impedance to 10… to 20…” (No. You didn’t.)

That’s almost certainly a UI/wording confusion: you changed attenuation (10 dB, 20 dB) — not “input impedance.” If the control is labeled “attenuate,” it’s attenuation.

And even if internal attenuation settings also improve the effective match (which is the whole point of the published condition), you still need a measurement method with sufficient resolution and proper reference-plane calibration to claim “it’s 50 Ω now.”

The video description

“Input impedance is a critical parameter… deeper understanding… choose the right equipment… troubleshoot…” That reads like a toaster trained on LinkedIn.

And the “not a certified lab” disclaimer misses the point. Certification doesn’t decide truth. Methodology does. A home lab can do excellent RF work… and a confused test can’t.

What a real TinySA input-impedance test looks like

• State the exact mode/path (low input vs high input, pad state, LNA state). Different paths can behave very differently.
• Calibrate at the TinySA connector plane (SOLT at the end of the test cable; de-embed adapters if you care about small differences).
• Sweep wide (not just a 2–3 MHz sliver).
• Measure S11 / return loss at attenuation = 0, 10, 20, 30 dB (and with/without LNA if relevant).
• Don’t mix experiments: “spur visibility vs attenuation” is a distortion/overload test, not an impedance test.

The actual takeaway

• The “50 Ω when attenuation ≥ 10 dB” line is an operating condition, not internet drama.
• Pads reduce mismatch impact; they do not magically change a non-50 Ω load into a 50 Ω termination.
• Watching spurs change with attenuation is a valid nonlinearity diagnostic… just don’t pretend it proves input impedance.

Interested in more technical content? Subscribe to our updates for deep-dive RF articles and lab notes.

Questions or experiences to share? Contact RF.Guru ... we love discussing real-world RF behavior.