---
Title: Turntable through a USB sound card – AudioLab
---

# Turntable through a USB sound card, no separate interface

The usual way to measure a turntable needs a phono preamplifier and a measuring interface's line input ([Turntable and test record](turntable.html)). A USB sound card with its own phono input does both jobs at once: choose **Windows audio device (no ASIO)** as the audio interface, and no separate interface or ASIO driver is needed at all. This page connects and sets it up; the alignment itself is the same procedure as on any interface – [Turntable setup, step by step](turntable-setup.html).

## What you need

- A **test record**: the Ortofon Test Record or The Ultimate Analogue Test LP.
- A **USB sound card with a phono input**, switchable between PHONO and LINE, with a ground terminal. The pitfalls and the numbers below are from a **Behringer UFO202** ([its full protocol](measured-behringer-ufo202.html)); another card's figures will differ, but the kind of thing to watch for should not.
- A **USB isolator** between the computer and the card (see the pitfalls below).
- The cartridge maker's **tracking force**.

## Connect

Turntable → the card's **PHONO** input directly, the ground wire on the card's GND terminal. The card → a USB isolator → the computer. There is nothing to connect to an interface's outputs: AudioLab plays nothing, the record does.

## Set

1. **CONNECTIONS** → *Audio interface*: **Windows audio device (no ASIO)**. *Wiring* locks itself to **External ADC (USB recording), no reference** – nothing to choose there.
2. *ADC (recording device)*: the card's input (↻ lists the devices again if it is not offered yet).
3. **Set to 0 dB** next to the volume line: on a USB card the Windows input volume is a gain in the card, also in exclusive mode.
4. *Sample rate*: the card's own rate, commonly **48 kHz**.
5. **MEASUREMENT** → Turntable → **Test record**.

Only the tasks that listen are offered: **Test record**, **Test tape**, **Player**, **Spectrum**, **Level meter**, **Speed & flutter** (with *Play test signal* off). Everything that needs AudioLab's own generator is greyed out – there is no output to play it through.

## Measure

From here it is the same as with a full interface: press **RUN**, play the record, and align it with the **Setup** view – [Turntable setup, step by step](turntable-setup.html) goes through every step. What follows here is what changes with just a card.

## The pitfalls

**1 – Check PHONO, not LINE, before every session.** The card's LINE position has no RIAA preamplifier built in: a cartridge's few millivolts then read tens of decibels lower, and what little comes through reads a worse THD, since the signal sits closer to the card's own noise. Picking LINE by mistake on a Behringer UFO202 read **33 dB lower** and distortion that looked a third worse – purely from the switch, nothing to do with the turntable.

**2 – Use a USB isolator.** The card and the computer share a ground over the USB cable; without an isolator the loop adds mains hum and extra noise onto the cartridge's small signal – on a line input it matters less, on a phono input it can be the difference between a usable and an unusable measurement (see also [Measuring a USB audio card, pitfall 1](step-usb-card.html#the-pitfalls)).

**3 – Get the gain right before trusting a track that shows heard.** The tracks between tones are matched by what they sound like, not by where the stylus physically is: a second or so of broadband surface noise or rumble, loud enough against a weak reference tone, can be mistaken for the record's own pink-noise band. This is what pitfall 1 going wrong looks like on the track strip – with PHONO correctly set it did not happen; if a track shows heard that was not played, check the gain before anything else.

**4 – A red Input overload is the card's headroom, not the turntable's fault.** A phono input has no gain control: the loudest tracking tracks can overload it well before the cartridge itself mistracks. Watch for AudioLab's own warning – a red **Input overload** message, and the **Tracking force** step reading *not judged* rather than a distortion figure – and do not read either as a setup problem. On the UFO202, clean tracking held to about 70–80 µm and overloaded above it; the rest of the setup (force, anti-skating, azimuth) still read correctly from the tracks within range.

**5 – Don't quote the card's own crosstalk figure elsewhere.** Set the turntable up by what actually moves – speed, tracking force, anti-skating, and above all the **azimuth** (the *difference* the crosstalk figures make, not their absolute level): these read the same through the card alone as through a full measuring interface, within the normal spread between two plays. The absolute crosstalk level does not: it read several decibels worse through the card. Fine for setting the turntable up, not for a number to quote anywhere else.

## What we checked

The same turntable, the same test record, measured once through a calibrated interface and once through nothing but the card:

| | Through a measuring interface | Through the card alone |
| --- | --- | --- |
| Crosstalk 1 kHz, L→R / R→L | −27.5 / −29.7 dB | −27.5 / −29.6 dB |
| Azimuth (the difference) | +2.3 dB | +2.2 dB |
| Channel balance | +0.58 dB | +0.59 dB |
| Response, largest deviation | −8.8 dB | −8.2 dB |
| Arm/cartridge resonance | 27.9 Hz, +3.7 dB | 28.0 Hz, +3.1 dB |
| Rumble, A-weighted | −58.3 / −55.9 dB | −57.8 / −54.2 dB |
| Wow & flutter, weighted RMS | 0.092 % | 0.091 % |

Close enough, on every figure, to call the card as good as the interface for setting a turntable up – which is what pitfall 5 is really saying: it is the adjustable figures that matter here, and they agreed.
