Fullscreen HTML5Med Res (??MB)HTML4

Hearing Loop Level Meter Last month, we described the circuit for the Hearing Loop Tester and gave the assembly details. This month, we show how to build a calibration coil and adjust the tester so that it gives accurate results. We also describe how the unit is used. Pt.2: By JOHN CLARKE H AVING COMPLETED the assembly, the Hearing Loop Tester is ready to be calibrated. As previously mentioned, the unit must be adjusted so that the 0dB LED lights for a magnetic field strength of 100mA/m. This is done by placing the meter in a known magnetic field and adjusting trimpot VR1. One possible method involves using a single 1-metre diameter turn of wire fed with 100mA at 1kHz. An amplifier set to deliver 0.82V RMS via an 8.2Ω 0.25W resistor could be used to drive the coil. However, to achieve correct calibration using this method, inductor L1 would have to be accurately located in the centre of the coil. That’s because the field strength varies depending on L1’s position relative to the centre position of the loop. A more practical calibration method involves using a Helmholtz coil (see http://en.wikipedia.org/wiki/Helmholtz_coil). A Helmholtz coil comprises two identical parallel on-axis coils that are driven by the same signal. siliconchip.com.au These two coils are separated from each other by the coil radius (Fig.6). A feature of a Helmholtz coil is that it gives a constant field along the axis between the two coils. This field remains constant to within 1% inside a central concentric area out to about half the diameter of the coil. The current required in each coil to give a field strength of 100mA/m is 0.1398 x R/n, where “R” is the radius in metres and “n” is the number of turns in each coil. In our case, we decided to design the coils so that they have only one turn each (to make construction easy) and can be driven by the headphone output of a PC. In practice, a 130mm-radius coil is suitable and this requires a coil current of 18.16mA to give 100mA/m. This is achieved by connecting the coils in series and driving them with a 600mV AC signal via a 33Ω resistor. Fig.6 shows the assembly details for our Helmholtz calibration coil. It’s built using 2.4mm diameter steel (fencing) wire, a 200 x 65mm length of timber, some hook-up wire and a 33Ω resistor. You will also need two screw terminals, a cable clamp, some shielded cable and a 3.5mm stereo plug. As shown, the 2.8mm steel wire is looped to form two 260mm-diameter coils. To do this, first cut two 836mm lengths and bend them down by 90° about 10mm from each end. That done, drill two sets of 3mm-diameter holes at each end of the timber to hold the wire ends. Each hole pair should be 8mm apart and the two pairs should be separated by 130mm (see Fig.6). The hook-up wire and 33Ω resistor can now be soldered to the ends of the steel wire. It’s then just a matter of bending the steel wires into loops and feeding the hook-up wires and the resistor down through the baseboard holes. The ends of the wire loops can be pushed into these holes to hold them in place. Use small cable clamps (if necessary) to hold the coils in place and make sure that the ends of each coil don’t short together. Use heatshrink to insulate them if necessary. December 2010  87 BOTH COILS 260mm IN DIAMETER, WOUND FROM 2.4mm DIAMETER STEEL WIRE CL 130mm DIA. 260 DIA. HOLD METER INSIDE THIS REGION BETWEEN COILS FOR CALIBRATION 90 19 8 130 10 3.5mm STEREO PLUG (INNER CABLE WIRE TO TIP, SHIELD BRAID TO SLEEVE. NO CONNECTION TO RING) 200 COILS (ABOVE BOARD) 65 33  RESISTOR UNDERSIDE VIEW OF BASEBOARD SHIELDED CABLE 4 x RUBBER FEET TERMINALS FOR MULTIMETER Fig.6: follow this diagram to build the Helmholtz calibration coil. The two loops are made from steel fencing wire and are connected in series and driven with a 600mV 1kHz sinewave signal via a 33Ω resistor – see text. Once the coils are in place, follow the wiring diagram of Fig.6 to complete the connections to the multimeter terminals and the stereo plug. Note that the ring terminal of the 3.5mm stereo plug is left open circuit. However, a mono jack plug can not be used since it would short out the right channel of a stereo socket. The cable to the 3.5mm stereo jack plug is held in place onto the timber using a suitable clamp. This clamp can be fashioned from some scrap aluminium or formed by soldering two solder lugs together. Finally, adjust the two coils so that they are vertical and parallel to each other and are aligned along the same axis. However, while the construction needs to be reasonably accurate, it does not have to be perfect. Small variations in the coil radius and position do not 88  Silicon Chip affect the field strength by much, so this should be well within 3dB of the theoretical value. Driving the coils The coils can be driven using a 1kHz signal generator and a suitable amplifier to deliver a 600mV AC signal. Alternatively, you can use a software sinewave generator and the soundcard output from a PC to drive the coils. The latter method will be the most used, so we’ll concentrate on that. We tested two free software generators. The first comes from BIP Freeware and can be downloaded from http:// www.electronics-lab.com/downloads/ pc/005/index.html It’s available as a compressed file named sine30.zip. To use this program, unzip the files to c:\program files\sine30 and create a shortcut to sine.exe on the desktop. The controls are easy to use. Make sure the mute is switched off and on again after every change in frequency, other­ wise the signal becomes corrupted. The output level can be varied over 255 steps using the volume control – see Fig.7. The second recommended sinewave generator is available at http:// www.diffusionsoftware.com/sinegen. php Download and run the SineGen _V1_0_setup.exe file. The relevant files will be placed into c:\Program Files\Little SineGen and you should create a shortcut to SineGen.exe on the desktop. Now run the program and select the soundcard driver. That done, set the output frequency to 1kHz by dragging the Frequency and the Divide sliders (the latter must be set to 1) – see Fig.8. Depending on which program you siliconchip.com.au Fig.7: this software sinewave generator from BIP Freeware can be used to generate the 1kHz sinewave signal. choose, the output level is adjusted using either the level control or the volume control. If there are sound problems with either sine generator, go to the sound properties dialogs (eg, in Control Panel) on your PC and check the various audio level adjustments. Setting the signal level With the sinewave generator now operational and set to 1kHz, plug the Helmholtz coil assembly into the PC’s audio output socket (green). The applied signal level should now be checked and adjusted using a multimeter that’s accurate for readings up to 600mV at 1kHz. If the meter is not accurate at this frequency, then set the generator to the highest frequency that the multimeter can accurately measure and adjust the level to 600mV. The output frequency should then be set back to 1kHz for the calibration. As an example, the multimeter we used has a claimed accuracy of 2% from 45-500Hz on its lowest AC voltage range (3.2V). As a result, we set the sinewave generator to 500Hz, adjusted the output level for a reading of 0.6V AC on the multimeter and then siliconchip.com.au Fig.8: another suitable sinewave generator program is Little Sinegen from Diffusion Software. set the generator back to 1kHz. If your DMM is only accurate up to 50Hz, then it is not sufficiently accurate to set the level from a computer sound card. That’s because most sound cards do not have a flat frequency response down to 50Hz, ie, the output level at 50Hz will be less than at 1kHz. Note, however, that you can adjust the level at 50Hz if you are using a “standalone” signal generator and an amplifier, provided the amplifier has a flat response down to 20Hz. During calibration, make sure that any equaliser settings on the computer (or tone controls on the amplifier) are set for a flat response. The calibration procedure is as follows: (1) Set the driving signal level to 600mV AC and the frequency to 1kHz as described above; (2) Hold the Hearing Loop Tester (without its lid) between the two coils. The unit should be held horizontally (ie, with the LED bargraph horizontal) and with its pick-up coil (L1) centred within the measurement area. (3) Adjust trimpot VR1 so that the 0dB LED (LED3) just lights. (4) Check that both coils are working by moving the Hearing Loop Tester along their axis. The signal strength should remain consistent at 0dB over the 90mm range depicted on Fig.6 and should be 3dB down (LED4 lit) if the pick-up coil is directly centred inside each coil. If the signal strength varies along the axis (ie, within the 90mm range), it’s probably because a coil is not working. In that case, check for shorts at the bottom of the coils, where they attach to the timber. That completes the calibration procedure. The Hearing Loop Tester is now ready for use. Checking background noise Checking the background noise prior to installing a hearing loop is important. This will help ensure that the loop is not affected by excessive noise due to mains wiring and/or any nearby equipment. According to Australian Standard AS60118.4-2007, environmental noise should not be any more than -20dB Aweighted with respect to a 100mA/m field (or -40dB with respect to a 1A/m field strength). At this level, the -21dB LED on the meter should either be off or just beginning to light. December 2010  89 Instead, we found that if the environmental noise is only just below -20dB with respect to the 100mA/m field, then the noise is too high for acceptable loop performance. In short, any signal from the hearing loop will be dominated by noise. It seems that the measurement standard for background noise is not stringent enough. And the reason for this is that an A-weighted measurement response masks out the major source of noise which happens to be at 50Hz and 100Hz. A-weighting rolls off these frequencies at -30dB and -19dB respectively, before the noise measurement is taken. Ditching A-weighting The Hearing Loop Tester is calibrated by holding it horizontally inside the centre-region of two wire loops and adjusting VR1 for a 0dB reading on the bargraph. The loops are driven with a 1kHz 600mV sinewave signal. Note that if a hearing aid loop is already installed, it must be switched off when making environmental noise measurements. Note also that the unit must be held vertically when making both noise and field-strength measurements. In order to make the A-weighting measurement, jumper LK1 must be out of circuit. However, before pro- ceeding, we should comment on the AS60118.4-2007 environmental noise standard and the A-weighting used for the measurement. Basically, the standard assumes that if the measured background noise level is 20dB below the 100mA/m reference level, then the area will be suitable for a hearing loop. However, our tests don’t bear this out in practice. We found that doing away with the A-weighting gives a better indication of background noise levels. By installing link LK1, the meter has a much better response over the 50-100Hz region and this gives a better correlation between noise measurements and the noise that is actually heard in a hearing aid (or hearing loop receiver) when picking up loop signals. As a test, we set up a loop, fed it with an audio signal and monitored it with the SILICON CHIP Hearing Loop Receiver described in the September 2010 issue. We also monitored the signal levels using the Hearing Loop Tester. We then introduced various noise sources to the loop (eg, a mains cord connected to a fluorescent lamp) and checked the audible noise levels. By then turning the audio signal off, we were also to check the noise levels on the tester. This showed that the indicated noise levels on the tester were well matched to any audible noise from into VIDEO/TV/RF? Television & Video Technology – by KF Ibrahim New edition has a full and compre-hensive guide to NEW LOW PRICE! video and TV tech-nology including HDTV and DVD, $ 58 starting with fundamentals. $ 70 DVD Players and Drives $ 95 NEW LOW PRICE! 85 $ – by KF Ibrahim DVD technology and applications - ideal for engineers, technicians, students, installation and sales staff. Practical Guide To Satellite TV – by Garry Cratt The book written by an Aussie for Aussie conditions. Everything you need to know – including what you cannot do! 7th ed. $ 49 Hands-On Zigbee – by Fred Eady $ 9650 NEW LOW PRICE! $ 75 An in-depth look at the clever little 2.4GHz wireless chip that’s starting to be found in a wide range of equipment from consumer to industrial. There’s something to suit every RF fan in the SILICON CHIP reference bookshop: see the bookshop pages in this issue $ 75 RF Circuit Design – by Chris Bowick A new edition of this classic RF text - tells how to design and integrate RF components into virtually any circuitry. NEW LOW PRICE!design 74 $ Practical RF H’book – by Ian Hickman A reference work for technic90 ians, engineers, students and NEW LOW PRICE! the more specialised enthusiast. Covers all the key topics in $ 73 RF that you need to understand. $ ! Audio ! RF ! Digital ! Analog ! TV ! Video ! Power Control ! Motors ! Robots ! Drives ! Op Amps ! Satellite 90  Silicon Chip siliconchip.com.au the receiver but only when LK1 was installed (ie, no A-weighting). With Aweighting selected, the tester gave no indication of background noise even when it made listening to the wanted signal quite difficult. So we recommend leaving LK1 in position when making noise measurements. In this configuration, noise levels will be satisfactory if they are at -21dB or less, assuming the tester is correctly calibrated (ie, either the bottom-most LED or no LEDs should light). This measurement recommendation is actually more stringent than the AS60118.4-2007 environmental noise standard. (0dB = 0.1A/m) +6dB +3dB 0dB -3dB -6dB -9dB -12dB -15dB -18dB -21dB Hearing Loop Tester SILICON CHIP Final checks Once the signal levels have been set and the frequency response checked, siliconchip.com.au Control Equipment Temperature Sensor A DS18S20 1-Wire temperature sensor is fitted into a waterproof stainless steel probe. Accurate to ±0.5 °C over the range of -10 °C to +85 °C. Length 3.4 metres EDS-001 $49.50+GST Function Generator Kit Based around the XR-2206 function generator IC, it can produce sine, triangle, and 5V square waves with frequencies ranging from 15Hz to over 500kHz. SFK-001 $39.00+GST Field-strength measurements Field strength measurements should be made using a 1kHz sinewave as the signal source for the loop amplifier. If you do not have a signal generator, you can use one of the software generators described above. If the hearing loop is a part of a sound system which also uses loudspeakers, the 1kHz tone should be set to the normal listening level. The hearing loop amplifier is fed with a signal from the main sound system. It’s then just a matter of adjusting the signal level from the hearing loop amplifier so that the 0dB LED just lights on the meter. By increasing the driving frequency (but keeping the level the same), you can use the tester to check frequency response of the loop from 1-5kHz. This will show up any high-frequency drop-off in the field strength due to inductance effects in the loop. Generally, it’s not necessary to check the loop response below 1kHz since inductance effects do not affect low frequencies. It’s not necessary for the low-frequency response of the hearing loop to go below 100Hz. If you do decide to check the loop’s response down to 100Hz, remember that the tester rolls off its low-frequency response. For the wide setting, with LK1 inserted, its response is 3dB down at 200Hz and 6dB down at 100Hz. This means that if the meter reads -6dB at 100Hz, then the loop response is actually flat to 100Hz. Similarly, if the meter reads -9dB at 100Hz, then the loop response is -3dB at 100Hz. Helping to put you in Control Power Fig.9: here are the full-size artworks for the front and top panels. the loop can be tested with normal program material, such as speech. If the amplifier includes a VU meter, adjust the volume control to give the same average VU level as for the 1kHz sinewave signal. Peak levels on the VU meter should be ignored. The Hearing Loop Tester can also be used to set the amplifier output to provide the correct 0dB level with normal program material. In practice, measured loop field strength levels will vary depending on the signal applied to the loop. If the loop amplifier includes a compressor or if the SILICON CHIP Hearing Aid Loop Signal Preconditioner (to be described) is used, then the signal level will be relatively constant. Finally, note that the meter has a slow response. This has been done so that it averages the signal level over time. This allows it to display the longterm average level without indicating individual signal peaks (which would SC be misleading). Triple Axis Accelerometer. MMA7341L XYZaxis accelerometer, a great low-g sensor with analog voltage outputs and adjustable sensitivity (±3 g or ±11 g), and a 0g-detect signal when the board is in free-fall. POL-1252 $17.50+GST 1 axis AC Servo Kit Consists of a 400W Brushless AC Servo motor with 1000 line encoder, AC Servo Drive and 60V 8 A power supply. Great for CNC applications. CNC-145 $624 +GST 8 Relay Card on DIN Rail Mount. We have reduced our prices for these incredibly versatile cards. Available in both 12VDC and 24VDC RLD-128 $109.95+GST Anemometer Alarm Card. Converts a Davis Instruments Anemometer wind speed and direction into 4-20mA / 0-5V signals. Can program 2 alarm relays to operate outside specified wind speeds or direction. Also Modbus connection. KTA-250 $159.00+GST Ph: 03 9782 5882 Our Catalog is Coming! www.oceancontrols.com.au December 2010  91