Wireless Microphone System Design White Paper CNK Engineering, Mesa, AZ

Transcription

1 A case study where good equipment must be carefully integrated to get reliable performance. Matt Bigwood, the Technical Director at Fresh Start Church in Peoria, AZ (West side of metro Phoenix area) knew that their 600MHz wireless equipment needed to be replaced. T-Mobile was building out their band 71 (600MHz) system in Arizona and was making sure that 600MHz wireless microphone operators were advised that they had to move their operating frequencies. The church purchased high quality equipment for the transition as shown in Table 1 below: Equipment Model Quantity Band Wireless Microphone System Sennheiser EW500 G3 10 G Antenna for Receivers RF Venue Diversity Fin 1 N/A Antenna Splitter RF Venue Distro 4 3 N/A In Ear Monitor System Shure PSM300 4 J13 In Ear Monitor Combiner RF Venue Combine 4 1 N/A Antenna for IEM RF Venue CP Beam 1 N/A Table 1, new equipment list Additionally, the church had the following equipment re-deployed from their legacy equipment inventory that was usable in the 500MHz band: 1 each Shure UHFR band G1 1 each Sennheiser EW300 G3 IEM band G The wireless receiver equipment was installed by the church technical crew as shown in Figure 1. Figure 1, Wireless equipment block diagram

2 The church tech crew did a frequency plan by scanning one of the EW500 receivers configured as in Figure 1 and consulting the manufacturer provided channel selection charts. There were problems. When a microphone was powered off some of the receivers would output full level noise that occurs when an FM receiver squelch un-mutes. The Pilot Tone feature was active on the receiver. Figure 2 shows the condition of a receiver when the noise output occurred Squelch level setting RF Noise Full Level Audio Noise Output Pilot Tone Indicator Figure 2, EW500G3 Receiver display during noise condition Notice that the Squelch level is set to maximum and the detected RF level exceeds the squelch and the Pilot Tone indicator is lit. The noise was intermittent and, naturally, seemed to occur a lot during Sunday worship. If a microphone was powered on, it would operate normally. Mics were routinely powered off as worship team members would switch between handheld and lav s sharing the same receiver. Although an operation workaround of keeping the mixer channel muted when a mic was powered off would keep the noise out of the mix, it was determined that it would add too many additional cues and risk having a muted channel from an overloaded board operator and besides, that is not how the receiver is supposed to operate. At this point, Matt contacted CNK Engineering for assistance. CNK Engineering was contracted to provide a site survey and to troubleshoot the noise burst problem at Fresh Start Church. The steps we followed are outlined below: 1. A broadband discone antenna mounted at the same height as the Diversity Fin Antenna was used with a Rohde&Schwarz FSH6 spectrum analyzer to survey beyond the tuning range of the wireless mic receivers. The Spectrum Analyzer was operated in the Maximum Hold mode to capture all of the intermittent signals. The UHF band of interest result is shown in Figure 3, Wideband OTA scan in Fresh Start Church

3 Figure 3, Wideband OTA scan in Fresh Start Church 2. The scanned results was compared to the Post-Auction TV re-pack plan for Phoenix as shown in Figure 4 Phoenix AZ Post Spectrum Auction TV Stations KPPX KNXV KPHE KPHO K18JL KGRF KPAZ KGRY KPDF KTVK KUTP KASW KTAZ KPPX KTVW KFPH KAZT Radio Astronomy Figure 4, Phoenix AZ Post-Auction TV re-pack T Mobile 600MHz Verizon, Sprint, AT&T 700MHz 3. The re-pack exclusions were entered into Shure Wireless Workbench to generate a frequency plan with consideration to 3IM and 5IM exclusions. We used the post auction repack plan for TV channel exclusions to avoid having to re-plan after the repack is completed. Note that TV channel 19, presently active, is shown as clear in the wireless workbench as it is a Low Power station with the transmitter located over 60 miles away and a mountain range in between. The measured power of channel 19 was below the -84dBm exclusion level as described in FCC A1. Figures 5 and 6 shows the derived frequency plan.

4 Figure 5, Frequency Plan plot for 11 mics and 4 ears Figure 6, tabulated frequency plan

5 4. The derived frequency plan was programmed into the wireless mic receivers and the IEM transmitters. Note that the Sennheiser EW300IEM G3 was operated stand-alone on the platform with the worship team and not input into the Combine 4 combiner. 5. At this time it was noted that the noise observed on the receivers was still present when transmitters were turned off. We used the spectrum analyzer at several points from straight OTA and thru the antenna distribution system and discovered that the apparent noise floor was elevated on the outputs of the distribution system such that the output of the cascaded output of the distribution system was about 10dB higher than the OTA measurement. See figures 7 and We contacted Don Boomer, west coast representative from RF Venue, Vanessa Jensen from the tech support group at Sennheiser and advised Jerry Delgado, the Sennheiser in house rep for the Phoenix area of the problem in order to brainstorm what possible causes of the noise could be and explore possible solutions. Based on discussions with RF Venue and Sennheiser we prepared a list of possible causes and solutions as below: a) Coax shielding- RF Venue supplied RG58 jumper cables from Distro 4 to receivers have poor shielding. RS and RE EMI is possible. Sennheiser suggests better shielded cables, RF Venue concurs. We suggest RF Venue supply Fresh Start Church with better quality jumper cables, perhaps LMR195, RG214 or RG8 b) Ground Loop- Switching power supplies can be modulating the ground. Better star ground to Distro 4 s and EW 500 G3 receivers can reduce ground loop. c) Overdrive receiver LNA- Distro 4 appears to have more than advertised 3dB of gain. Do not know if this is creating overdrive condition, but with 60 distance between mic and RX antenna distro system gain is probably not necessary. Noise floor is raised with gain of Distro 4. See Figure 9. The right combination of IM mixed with adjacent digital TV signals spectral regrowth could get digital noise to a level where the pilot tone circuit could be fooled. Adding attenuation on receiver input could reduce the level. d) DC on RX input of EW 500 G3 receiver- Distro 4 probably doesn t care if DC present on outputs, but a DC path is probably not good. Could add DC blocks, but this is a costly solution and we don t know if this is root cause. BNC DC Blocks are about $25 ea. from Mini Circuits and it would require 22 of them.

6 Figure 7, OTA test configuration Figure 8, thru the RF Venue Distro 4 test configuration

7 Figure 9. Spectrum compared OTA and thru Distro 4 #3 loop thru output, same antenna 7. Ultimate Resolution of problem a) RF Venue agreed that cables with poor shielding on RG58 cables could certainly contribute to Radiated Susceptibility and supplied replacement RG8 cables for the connections between the Distro 4 units and the receiver inputs. They also supplied bandpass filters which were installed between the Antennas and the input of the first Distro 4 unit. This gave incremental improvement as TV Stations still in the 600MHz band (repack has not yet occurred in Phoenix) were reduced in level to minimize intermod contribution from those TV stations. Problem with noise bursts still occurred but with reduced frequency. As a temporary workaround, Fresh Start Church used the receivers with the factory antennas. There were no noise bursts but there were Line of Sight dropout issues which was the reason they wanted to add the Diversity Fin antenna and distribution system in the first place. b) The inputs to the first Distro 4 unit were illuminated with 2 6MHz spaced CW signals with the spectrum analyzer monitoring the loop thru path then the receiver output path (unused Receiver outputs were terminated with 50Ω) to measure the Input Third Order Intercept point of the LNA s in the Distro 4 unit. c) The Occupied Channel Power of the Intermod created by the nearby TV channels was measured and attenuation (15dB) was added to the input of the first Distro 4 unit to reduce the noise floor boost created by the intermod of the digital TV signals. Since the Distro 4 units have greater than unity gain, additional attenuation (6dB) was added to the loop thru output of the 2 nd Distro 4 unit to further reduce the drive level to the 3 rd Distro 4 unit. d) The TX power of the microphones were set to 20mW (+13dBm) and the distance from the platform in the church to the Diversity Fin RX antenna is 60 to 70 ft. (FSPL about 52 to 53dB). e) A gain stack up including added attenuation of the LPA path and the dipole path of the Diversity Fin antenna was modeled to minimize the effect of the intermods and is shown in Figure 10 f) The block diagram of the system configured with added filters and attenuation is shown in Figure 11 g) A picture of the receivers showing the reduced noise floor is shown in Figure 12 and 13. Fresh Start Church has had no more noise burst problems since the system re-configuration.

8 LPA Path gain(db) Power (dbm) Dipole Path gain(db) Power (dbm) Microphone transmitter 13 Microphone transmitter 13 User attenuation* User attenuation* FSPL 70'@560MHz FSPL LPA Gain 5-46 Dipole Gain ft RG8 (incl.2db loss for 2 BNC connector ft RG8 (incl.2db loss for 2 BNC connector Filter Filter dB attenuator dB attenuator Gain of Distro 4 # Gain of Distro 4 # ft RG8 cable (1) 1 ft RG8 cable (1) Gain of Distro 4 # Gain of Distro 4 # ft RG8 cable (2) 1 ft RG8 cable (2) 6dB attenuator dB attenuator ft RG8 cable (3) 1 ft RG8 cable (3) EW500 receiver sensitivity threshold spec from data sheet 2uV (-101dBm) * User Attenuation is for sweaty hand holding the mic over the antenna area (1), (2) and (3) are levels at the receiver input Figure 10, Gain stack up of final wireless mic system configuration Figure 11, Block diagram of final system configuration

9 Figure 12, Receivers show reduced noise after final system re-configuration Figure 13, before and after system re-configuration 8. Conclusions Prior to the Digital TV and Wireless Broadband spectrum re-packs that resulted in the loss of the 700MHz and 600MHz bands it was easy to setup wireless microphones and other

10 wireless equipment. It is now more difficult and takes system planning, modeling and measurement to get predictable, reliable system performance. Manufacturer provided specifications tend to show typical equipment performance and actual performance of a particular piece of equipment may vary. In order to properly integrate a Wireless Microphone, In Ear Monitors, Instrument Links and Production Comms it is necessary to use RF test equipment such as Calibrated Antennas, Spectrum Analyzers, Signal Generators and Network Analyzers to accurately characterize equipment performance which is then modeled then integrated and finally tested as a whole system to verify modeled performance. Use of high quality cables as well as filters and attenuators is necessary to successfully operate wireless equipment in the tightly packed, reduced spectrum now available. After the system is operational, it is additionally important to protect the system frequencies by licensing under an FCC part 74 license to reduce the chance of interference by other wireless equipment. We want to thank RF Venue and Sennheiser for the assistance they provided. For more information please contact: CNK Engineering PO Box 734 Mesa, AZ marty.pietz@cnkengineering.com (480)