Wireless Passive splitters

[headcrab]

We have 14 SLX4 receivers in our rack here, and every show I've done we have had problems with the mics dropping signal. It doesn't matter which channel I use. I thought that since the wavelength of the carrier is about the same as the width of the rack rails it might be that the rack is shielding the antennas. Would it help to use a non-conductive rack?
If I can't do that, can I put the antennas remotely and use passive signal splitters (i.e. from Radioshack) to distribute the RF to each receiver?

 

[mbenonis]

A couple of questions. First, do you currently have any antenna distribution or does each receiver have a pair of antennas on it? Are the antennas within a metal rack, or outside the rack? How far are the receivers from the stage (worst case distance)? What frequency band are they, and have you done any coordination? Are there any other major sources of RF in the UHF TV band, such as other wireless mics, IEM systems, wireless intercom, local TV stations, etc? Where are you located?

These questions will help us to diagnose the problem.

 

[museav]

You might take a look at this recent thread, http://www.controlbooth.com/forums/sound/12543-wireless-mic-trouble.html, it might answer some of your questions. An easy aspect to address is that the antennas should be mounted to the outside of the rack and I believe that is covered in the SLX manual.

This is not directed at this particular post, but this general topic seems to be rather common recently. Most here are dealing with problems thrust upon them and it seems that a number of people are encountering the same problem caused by people 'needing' to have a large number of wireless mics but then not being able to afford implementing them properly. I think that it is important to promote the perspective that if someone can't afford the proper implementation for the number of wireless mic systems wanted then what they are really saying is that they can't afford that many wireless systems.

 

[headcrab]

So we have antenna B inside the rack, and antenna A outside, on the front of the rack mount ears for each receiver. (No distribution system). I haven't set any specific antenna angle (someone mentioned 45 degrees). Worst case distance to stage is 100 feet.
Rack is earth-grounded
Frequency: Shure J3 (572-596MHz) and L4 (638-662MHz) bands. The only other wireless infrastructure in the building is the wireless network. I'm in Fort Collins, CO and as far as I know there are no major broadcasters nearby.

 

[museav]

Having all the "B" antennas inside the rack effectively defeats the diversity aspect and makes droputs much more likely. 100' shouldn't be a problem by itself but it depends on what is between the two points, people often forget that commercial construction can mean metal stud walls, walls filled with metal conduit or ductwork, etc. for the booth. You can also get multipath off metal ceiling decks and other RF reflective surfaces.

Remoting the antennas usually helps, but multiple generic RF splitters combined with long antenna cabling runs could mean unacceptable losses that more than offset any potential benefit.

I would try to get all the antennas mounted outside the rack and sticking above it as Shure recommends, You could also confirm with Shure the recommended operating frequencies.

 

[mbenonis]

I agree with Brad - see if you can get the B antennas outside of the rack, on the side facing the stage. Put the antennas at 90 degrees to each other (likely, 45 degrees from the horizontal facing the opposite direction, but it doesn't need to be that way).

With 14 systems, you'd definitely need an active antenna distribution system. You could probably get away with four active 4-way splitters, and split the antennas passively into each (and one splitter would feed the other one).

 

[FMEng]

Let's do a little math. I assume you have something like a four-way splitter feeding three more four-way splitters. The loss from input to any output port of a four-way splitter is around -7.5 dB (deciBels). Daisy chaining splitters adds up the losses. Therefore, each receiver is getting -15 dB of the signal from the antenna.

-15 dB loss means 1/32 of the original signal! As a frame of reference -3 dB is half the strength, and -10 dB is one-tenth the strength.

Go with a lossless splitter system. If you decide to remote the antennas any significant distance, say 50 feet or more, go with fatter coaxial cable such as RG-11. That reduces the loss in the cable. RG-11 is about a 1/2 inch thick versus RG-59, which is around 1/4 inch thick.

 

[Chris15]

Go with a lossless splitter system. If you decide to remote the antennas any significant distance, say 50 feet or more, go with fatter coaxial cable such as RG-11. That reduces the loss in the cable. RG-11 is about a 1/2 inch thick versus RG-59, which is around 1/4 inch thick.

If you are using a lossless splitter system and remoting antennas, then you might as well also use the right impedance cable - RG8 is thicker than 58 and then you start getting into the domain of Times Microwave and friends with things like LMR400. RGs 59, 6 & 11 are all 75 ohm coax - go search for the discussions about matching losses...

 

[FMEng]

If you are using a lossless splitter system and remoting antennas, then you might as well also use the right impedance cable - RG8 is thicker than 58 and then you start getting into the domain of Times Microwave and friends with things like LMR400. RGs 59, 6 & 11 are all 75 ohm coax - go search for the discussions about matching losses...

Matching impedances is theoretically a good idea. In practice, it doesn't make any differences with receivers, provided all of the devices in the system are passive. After all, the impedance of a quarter wave whip antenna with no ground plane is not 50 ohms. I'd be willing to bet that if you measure the actual impedance of the receiver input, it isn't anything like 50 ohms either. The loss per foot of a 75 ohm cable will always be slightly lower than the 50 ohm cable of similar size

Where impedance matching becomes important is when you have an active device, like an antenna preamplifier or active, lossless splitter. There, the cable being driven should match the impedance of the source to keep the amplifier stage happy. What you use ahead of the input of an active device should not matter.

If you aren't sure, then go ahead and use 50 ohm cable. However, the cable and connectors are much harder to aquire. I can go to Home Depot and buy decent quality RG-11 and RG-6. Radio Shack might sell RG-8 and RG-58, but if you cut it open you will find that it is horrible quality. RS cable has lousy shielding with no foil and swiss cheese braid.

BTW- LMR400 is 50 ohm cable. It is essentially RG-8 with improved performance.

 

[Chris15]

Matching impedances is theoretically a good idea. In practice, it doesn't make any differences with receivers, provided all of the devices in the system are passive. After all, the impedance of a quarter wave whip antenna with no ground plane is not 50 ohms. I'd be willing to bet that if you measure the actual impedance of the receiver input, it isn't anything like 50 ohms either. The loss per foot of a 75 ohm cable will always be slightly lower than the 50 ohm cable of similar size

Where impedance matching becomes important is when you have an active device, like an antenna preamplifier or active, lossless splitter. There, the cable being driven should match the impedance of the source to keep the amplifier stage happy. What you use ahead of the input of an active device should not matter.

If you aren't sure, then go ahead and use 50 ohm cable. However, the cable and connectors are much harder to aquire. I can go to Home Depot and buy decent quality RG-11 and RG-6. Radio Shack might sell RG-8 and RG-58, but if you cut it open you will find that it is horrible quality. RS cable has lousy shielding with no foil and swiss cheese braid.

BTW- LMR400 is 50 ohm cable. It is essentially RG-8 with improved performance.

A 50 to 75 ohm mismatch results in a 0.8dB loss each time. Not the end of the world but tis better to avoid where one can.

The use of a 1/4 wave antenna without ground plane is not a recommended practice, so why would we be discussing it?

People buy cable off the shelf... oops I forgot we're talking about high schoolers here. Real RG58 and RG8 are much nicer. I've seen some pretty icky 75 ohm coax in my time - especially RG59. And if you get worried, use LMR195 in lieu of RG58.

Where did I imply that LMR400 and it's brothers were not a 50 ohm coax? I meant to imply that once you move beyond the loss budget allowance with RG8 you start looking at LMR400 (and then LMR600 and a bank loan).

I guess I look at it in the context of rentals where one really should run 50R coax so that things still play nice when you are using the same cables for transmit or receive as needed.

I guess in a number of ways it's like the use of AES spec cable for AES. Will it stop your show if you do it wrong over a short length - probably not. Try pushing the limits of your setup and you start to run a risk of things not working as well as they might otherwise...

 

[FMEng]

A 50 to 75 ohm mismatch results in a 0.8dB loss each time. Not the end of the world but tis better to avoid where one can.

(snip)

I guess in a number of ways it's like the use of AES spec cable for AES. Will it stop your show if you do it wrong over a short length - probably not. Try pushing the limits of your setup and you start to run a risk of things not working as well as they might otherwise...

Your point about losses across a mismatch is quite correct, but you are missing my point. The input of most receivers isn't matched to anything. They are not 50 ohms, they are not 75 ohms. If the antenna system connected to those inputs is totally passive, then there is no point in doing anything other than matching cable impedance to splitter impedance.

Manufacturers call them 50 ohms to prevent nosy technicians from calling them to ask what the impedance is. Take an RF network analyzer and measure some. It's more likely to be up in the kilohms or megohms and the world still spins.

We also need to recognize that receiver inputs are driven by signal voltage, not power. An impedance mismatch reduces signal power, but not necessarily voltage.

AES digital inputs and outputs are a different situation all together because they ARE a truly impedance matched system. The standards for AES digital connections are remarkably uniform and well adhered to.