Wireless Systems for Measurement – Part 2

by Steve Liddle

In part 2 of Wireless Systems for Measurement, Steve Liddle will show the effects of changing the output signal level in the measurement chain. Three wireless systems were used in the test.

Introduction

In the second part of this short series you will see the effects of changing the output signal level in the measurement chain. Ideally, the response should remain the same regardless of the input level as long as it remains within the linear operating range.

Method

Dual channel FFT transfer function measurements were made as in part 1.

I used a signal that all the systems had previously handled without any issues, pink noise.

The output level of the sound card was then increased incrementally until the wireless systems started to show clipping either at the transmitter or receiver.

Results

The graphs below show the transfer functions of the 3 wireless systems under test.

The cursors show the levels at the indicated frequency and were previously adjusted so that they didn’t overlap when displayed on one graph – so don’t pay any attention to the fact that the Sennheiser system shows values around -4dB – it’s the differences between the response at the varying input levels that are of interest.

The dB levels are dB relative to Full Scale and are taken from the output level settings on the MOTU Ultralite Mk3 audio interface. The pink line is the reference response of a loop back cable with the output level at -10dB re full scale.

In alphabetical order:-

Behringer ULM2000

This system behaves relatively well with small variations of 0.1 or 0.2 dB until the clip level is approached and there is a larger change. However, 0.5dB does not present us with a huge problem and the changes are consistent across the frequency range.

Lectrosonics TM400

Surprisingly, this system shows an increasing attenuation at higher frequencies with increasing signal level. The red line shows the response once the overload indicator started to flash and demonstrates a limiting behaviour. However, there is a almost a 2dB difference between signals at -30dB and -15dB, something to consider if you are applying equalisation in that region.

Sennheiser EW100 G1

Although the response slopes off above 10kHz, the Sennheiser system is very linear until the maximum input level is approached.

Conclusion

All 3 systems show consistent linearity in the frequency range between 70 Hz and 10kHz.

Outside this range above 10kHz the Lectrosonics system has the best repsonse but showed the largest deviation in terms of linearity, which was surprising.

Generally the Lectrosonics system has the widest, flattest frequency response and it’s obvious why it has become and industry standard. I believe the Sennheiser system offers a more reasonably priced alternative as long you don’t expect it to perform outside it’s limitations. The Behringer system is surprisingly well behaved but the frequency response makes it necessary to employ a  reference frequency response compensation which isn’t available in all software platforms – for example, you need the pro version of Systune to do this.

Next up:

real world testing. It’s all well and good performing these tests in the comfort of the workshop environment but what happens when these systems are used in the real world? We’ll take a look at how the systems perform when making electroacoustic measurements in a large room.