Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Nathan Riddle]

I have a number of questions that are mulling about in my head after working on this rack install (distro & combine) at a church.

I'll try to be clear/logical with my presented findings; but it's all jumbled in my head.

Info:

Before:
First, rack (senn RX Mics & AT TX IEMS) is slapped together all using 1/4 wave antennas.
Second, TX was in high power mode & in same rack.
Third, I ran WWB6 analysis on the frequencies; oh my goodness, half (6) of the 12 were incompatible.
Fourth, Rack was on stage and max of 50ft away from any TX/RX device

After:
-I of course installed RF Venue Distro4's for the RX. As per one of their blog posts I kept the RF cables 'disorganized.'
-I added a single combiner. I wanted to use RF Venue's Combine4; but I was trying to save a tiny bit of money for the client & use a 1/2 rack unit instead of full rack. This was a mistake because the AT combiner can't be racked with the AT TX units... 

-(It was very late/long day) I hadn't yet installed the remote 1/2λantenna (or received them) but I wanted to make sure they were still up and running. So I slapped on 1/4λ on the front of the rack for the RX (distro4 has bnc couplers) with the TX on the back side hoping to just 'get-by' until I had finished correctly. This made the RX units in the rack fare even worse than before.
-I then turned down the RF power on the AT TX IEMS in the rack so the RX wouldn't get overwhelmed.
-This didn't help all that much.
-I then put the 1/4λ back on and it worked 'okay' (i'm assuming as well as before)

Later (another day)
-I did the RF coordination and the 1/4λ antenna farm looked good.
-So I tried the 1/4λ with distro. Still issues even after coordination.
-I then grabbed my DiversityFin and remotely located the RX and still used the 2x 1/2λ on the back side of the rack. Worked beautifully.


Questions:
1) Why did the original rack work at all?
2) After adding distro [still using 1/4 wave & same frequencies] why did it seem to work worse than when they had an antenna farm?
3) Having nearly finished (still need to remotely mount the 1/2λ RX antenna). Anyone care to peer-review anything RF-wise that I might have done poorly? I tried to stick to all the best practices that I know of.


Before images:

[Nathan Riddle]

Freq before & front of almost finished rack.

[Nathan Riddle]

Noise floor RF Scans

[Henry Cohen]

Questions:
1) Why did the original rack work at all?

Luck mostly. Also remember that IM calculations predict where IM products occur, not that they necessarily will be destructive.

2) After adding distro [still using 1/4 wave & same frequencies] why did it seem to work worse than when they had an antenna farm?

Likely a combination of factors:
- There are reasons why low cost combiners and multi-couplers cost only a few hundred dollars while higher tier products cost north of $1k; port to port isolation, IP3, IM suppression, linearity, noise figure to name a few.
- Both antennas had near field obstructions (the rack) thus couldn't work efficiently.
- The IEM TX antenna being so close to the rack, with a high composite RF energy level, could have been exciting the metal parts of the rack enough to act sympathetically as a passive radiator, again due to the fact the rack is within the the reactive near field of the antenna(s).

3) Having nearly finished (still need to remotely mount the 1/2λ RX antenna). Anyone care to peer-review anything RF-wise that I might have done poorly? I tried to stick to all the best practices that I know of.

A few observations:
- Get the antennas away from the rack and up high into free field [as much as possible], such as mounting on mic stands.
- Unless you can provide a proper counterpoise for a 1/4 whip, use 1/2 wave / dipole or directional antennas.
- Use a better quality TX combiner.

[Nathan Riddle]

Luck mostly. Also remember that IM calculations predict where IM products occur, not that they necessarily will be destructive.


Likely a combination of factors:
- There are reasons why low cost combiners and multi-couplers cost only a few hundred dollars while higher tier products cost north of $1k; port to port isolation, IP3, IM suppression, linearity, noise figure to name a few.
- Both antennas had near field obstructions (the rack) thus couldn't work efficiently.
- The IEM TX antenna being so close to the rack, with a high composite RF energy level, could have been exciting the metal parts of the rack enough to act sympathetically as a passive radiator, again due to the fact the rack is within the the reactive near field of the antenna(s).

Thank you, that really helps since I'm guessing at best compared to knowledgeable guys/gals like you.

A bit like when I didn't understand sound/limiters. I can learn though

A few observations:
- Get the antennas away from the rack and up high into free field [as much as possible], such as mounting on mic stands.
- Unless you can provide a proper counterpoise for a 1/4 whip, use 1/2 wave / dipole or directional antennas.
- Use a better quality TX combiner.

Check, I plan on mounting on the stage walls about 8-10ft high on either side. (LMR-400 for the 50' run)
Check, using Sennheiser's omni 1/2 wave A1031U (They already had these not in use)
Danget, so I should have pushed for the Combine4 as a minimum?

Last related question:
Is concrete somewhat free field?
I'll be using wall boxes/plates to standoff the antenna from the wall as much as possible. But that will only be 2"-5".

[Don Boomer]

Don't worry so much about the noise floor by itself. The Distro 4 is probably supplying a little gain. So what s really important to staying on the air is the dynamic range. If you ran these again you would likely see that the Tx is stronger tha it was before.

I would highly recommend using a helical antenna for the IEMs.

If you are gonna stick with omni antennas for the receivers, they will be much happier about 6’ away from the wall a opposed to being directly mounted to the wall.

[Nathan Riddle]

Don't worry so much about the noise floor by itself. The Distro 4 is probably supplying a little gain. So what s really important to staying on the air is the dynamic range. If you ran these again you would likely see that the Tx is stronger than it was before.

Thanks Don! I love your gear, I was mainly in 'curiosity mode' (I don't know what I don't know).

I would highly recommend using a helical antenna for the IEMs.

While I agree, when I sent the quote to do this they were budget-savvy (hence the AT junk instead of Combine4) [Which I admit, I shouldn't have flexed there]. But I think the 1/2 wave's will work given the distance from the antenna to body pack will be less than 20ft.

If you are gonna stick with omni antennas for the receivers, they will be much happier about 6’ away from the wall a opposed to being directly mounted to the wall.

Dang, I was afraid of that. I did stress that the 1/2 waves should be substituted for LPDA's, but it's hard to beat free (already owned).

For future reference do LPDA's fare better being mounted to a wall since they have decreased sensitivity to the rear?

[Luke Geis]

I am not an RF master, but I have been doing a LOT of studying lately since I seem to be using RF at every show I do lately.

My understanding is this:

1. More RF power is not better, it just helps in certain situations. Too much is not a good thing as it can make things worse.

2. 1/4 wave antennas I believe require a ground plane to really work at all and that ground plane is the wireless unit. When it is moves away remotely in any way, it looses ability.

3. Half wave antennas are actually REALLY good......... if properly deployed. They are usually best placed up as high as you can get them. As long as there is line of sight, they should work fine. The second plus is using as short a BNC cable ( at the correct ohm rating ) as possible. Thie reduces signal loss.

4. Directional antennas and distros do not guarantee better performance if the basic rules are not followed. Line of sight is key, short BNC runs are important and not having too much gain is also key. Aside from that, too much of a good thing is still too much. If you have a directional antenna and you are only 10' away from the receiver, you may not need a distro at all and using one may make the situation worse.

I have found that it is all about making things as simple as possible and following basic rules. If you are in an environment that is very dirty, you have to get the receiver as close to the transmitter as is practical using as low gain an antenna as you can. Next is making sure there is line of sight and nothing that can interfere with RF nearby. LED walls are not helpful..... If you can reduce the noise the receiver hears while increasing the opportunity that the transmitter is the only thing that gets through, you better the odds of a good experience. The answer is not always more of anything, but actually finding a way to lessen things.

Next, I have learned that boundaries play a HUGE role. The RF wave emits in two planes and the idea is to make it so that the antennas can only catch any one plane at a time. When near boundaries such as the floor or walls, there is a bounce that occurs and can come back at near the same power. This bounce can be heard by the receiver and will affect how it " hears " things. If significantly out of phase with either of the two planes, it will wreak havoc on RF diversity. You want the direct signal path to be what the receiver hears and preferably only able to " hear " the transmitter's signal.

[Henry Cohen]

1. More RF power is not better, it just helps in certain situations. Too much is not a good thing as it can make things worse.

Correct

2. 1/4 wave antennas I believe require a ground plane to really work at all and that ground plane is the wireless unit. When it is moves away remotely in any way, it looses ability.

1/4 wave antennas do in fact require a ground plane (or more accurately, a counterpoise), but any metal surface orthogonally opposed to the radiating element can be the ground plane; it doesn't have to be the receiver chassis.

3. Half wave antennas are actually REALLY good......... if properly deployed. They are usually best placed up as high as you can get them. As long as there is line of sight, they should work fine. The second plus is using as short a BNC cable ( at the correct ohm rating ) as possible. Thie reduces signal loss.

Any antenna performs better when up high above obstructions, regardless of type. Short coax cable is not nearly as important as using low loss coax (double shielded) of the correct size to minimize losses. It's far better to run a longer piece of low loss coax to get the antennas closer to the transmitters; at the distances we work, free space path loss is always greater (by a significant magnitude) than the loss through good coax. (And BTW, RG58, RG8X, and RG213 are not good coax types for UHF frequencies.)

4. Directional antennas and distros do not guarantee better performance if the basic rules are not followed. Line of sight is key, short BNC runs are important and not having too much gain is also key. Aside from that, too much of a good thing is still too much. If you have a directional antenna and you are only 10' away from the receiver, you may not need a distro at all and using one may make the situation worse.

Mostly correct, but see my comment about coax length vs. type above.

I have found that it is all about making things as simple as possible and following basic rules. If you are in an environment that is very dirty, you have to get the receiver as close to the transmitter as is practical

No; get the receiver's antennas as close to the transmitters as possible.

using as low gain an antenna as you can.

Not necessarily. A high gain passive directional antenna will provide off axis rejection which could be as critical to good reception, if not more so, than a unity gain antenna.

Next is making sure there is line of sight and nothing that can interfere with RF nearby. LED walls are not helpful..... If you can reduce the noise the receiver hears while increasing the opportunity that the transmitter is the only thing that gets through, you better the odds of a good experience. The answer is not always more of anything, but actually finding a way to lessen things.

Mostly correct, but there are finer points to this that are dependent on the given deployment.

Next, I have learned that boundaries play a HUGE role. The RF wave emits in two planes and the idea is to make it so that the antennas can only catch any one plane at a time. When near boundaries such as the floor or walls, there is a bounce that occurs and can come back at near the same power. This bounce can be heard by the receiver and will affect how it " hears " things. If significantly out of phase with either of the two planes, it will wreak havoc on RF diversity. You want the direct signal path to be what the receiver hears and preferably only able to " hear " the transmitter's signal.

Note sure what you mean by "two planes". If you mean the electric field and the magnetic field (the two fields are orhtogonally opposed), both are required for proper RF propogation and reception at the RX antenna. If you mean azimuth (horizontal) beamwidth versus elevation (vertical) beamwidth of the antennas, these parameters are a known spec of the antenna and are part of the decision making process about which antenna type is best suited to a given application. Both patterns need to be considered when determining antenna deployment so that as much far field freespace as possible is provided.

[Larry Sheehan]

distances we work, free space path loss is always greater (by a significant magnitude) than the loss through good coax. (And BTW, RG58, RG8X, and RG213 are not good coax types for UHF frequencies.)

Henry, could you please elaborate? If not these types, what are the correct cable types?

[Henry Cohen]

Henry, could you please elaborate? If not these types, what are the correct cable types?

Coaxial cable losses at the distances we're working with will always be less than than free space path loss.
 
Pick a coax cable, any coax cable (and a frequency): http://www.qsl.net/co8tw/Coax_Calculator.htm. For safe measure, add 2dB to account for worn connector losses.

Now check the path loss at that same frequency (use "0" for transmitter and receiver gain): https://www.pasternack.com/t-calculator-fspl.aspx.

This is why it's virtually always better to get the antennas as close to the transmitters as possible, running longer lengths of coax if necessary.

As far as low loss coaxial cables are concerned, a double shield (braid over foil, or less so double braid) construction is desired as it's less lossy AND less permeable to external RFI.
Belden 9913, 9913F7, RF195, RF240, RF400
PWS S9046
Times Microwave LMR series (195, 240, 400, 600)
The Wireman CQ118 and CQ106


Edited to discount the 9913 type, but show the difference in part number for 9913F7

[Scott Holtzman]

Coaxial cable losses at the distances we're working with will always be less than than free space path loss.
 
Pick a coax cable, any coax cable (and a frequency): http://www.qsl.net/co8tw/Coax_Calculator.htm. For safe measure, add 2dB to account for worn connector losses.

Now check the path loss at that same frequency (use "0" for transmitter and receiver gain): https://www.pasternack.com/t-calculator-fspl.aspx.

This is why it's virtually always better to get the antennas as close to the transmitters as possible, running longer lengths of coax if necessary.

As far as low loss coaxial cables are concerned, a double shield (braid over foil, or less so double braid) construction is desired as it's less lossy AND less permeable to external RFI.
Belden 9913, 9913F7, RF195, RF240, RF400
PWS S9046
Times Microwave LMR series (195, 240, 400, 600)
The Wireman CQ118 and CQ106

LMR400 3/8" has always been my go to for low power apps.  We even used it to connect exciters to PA's in 900Mhz paging transmitters and it held up.  Easy to terminate too.

[Jason Glass]

As far as low loss coaxial cables are concerned, a double shield (braid over foil, or less so double braid) construction is desired as it's less lossy AND less permeable to external RFI.
Belden 9913, 9913F7, RF195, RF240, RF400
PWS S9046
Times Microwave LMR series (195, 240, 400, 600)
The Wireman CQ118 and CQ106

FWIW, in addition to using only foil + braid 100% sheild variants, my policy (and recommendation) is to avoid all "air spaced" dielectrics and stay with PE foam dielectric variants like LMR400UF or Belden 9913F7.

Why?  First and most crucial is that they are far more resistant to permanent damage caused by crushing and kinking, which unsupervised stagehands will do EVERY TIME during load out.  EVERY TIME.  100 OUT OF 100 TIMES. Because they pull, then pull, then pull again, and then go look to see what's hanging up. After the coil out of their sight has collapsed into a tight hard kink.  No matter how explicitly you instruct them and beg them to be careful.  EVERY.  DAMNED.  TIME.  It can of course be straightened to visually passable, but an air spaced center conductor remains internally and invisibly kinked, permanently compromising the cable's continuous impedance and loss specs.  Second, air spaced variants suffer from pushed and pulled connector pins because their center conductors aren't bonded and are free to float and migrate under repeated coiling, uncoiling, and temperature cycling. This induces enormous tensile and compressive forces on their connectors' components.  Third, their center conductors are exposed to moisture and corrode at an accelerated rate compared to bonded foam variants.  This failure mode has destroyed many of my clients' cables beyond repair in only several years of service, while bonded PE foam dielectric variants last decades without any corrosion at all.

And beware, EVERY single digit and alpha in a cable model # is crucial. For example, Belden 9913 is a completely different cable than 9913F7.  Just because their loss per ft. specs are similar means nothing regarding all their other properties.

Sent from my mobile phone using Tapatalk

[Larry Sheehan]

Coaxial cable losses at the distances we're working with will always be less than than free space path loss.
 
Pick a coax cable, any coax cable (and a frequency): http://www.qsl.net/co8tw/Coax_Calculator.htm. For safe measure, add 2dB to account for worn connector losses.

Now check the path loss at that same frequency (use "0" for transmitter and receiver gain): https://www.pasternack.com/t-calculator-fspl.aspx.

This is why it's virtually always better to get the antennas as close to the transmitters as possible, running longer lengths of coax if necessary.

As far as low loss coaxial cables are concerned, a double shield (braid over foil, or less so double braid) construction is desired as it's less lossy AND less permeable to external RFI.
Belden 9913, 9913F7, RF195, RF240, RF400
PWS S9046
Times Microwave LMR series (195, 240, 400, 600)
The Wireman CQ118 and CQ106

Thanks, Henry.

[Henry Cohen]

FWIW, in addition to using only foil + braid 100% sheild variants, my policy (and recommendation) is to avoid all "air spaced" dielectrics and stay with PE foam dielectric variants like LMR400UF or Belden 9913F7.

Why?  First and most crucial is that they are far more resistant to permanent damage caused by crushing and kinking, which unsupervised stagehands will do EVERY TIME during load out.  EVERY TIME.  100 OUT OF 100 TIMES. Because they pull, then pull, then pull again, and then go look to see what's hanging up. After the coil out of their sight has collapsed into a tight hard kink.  No matter how explicitly you instruct them and beg them to be careful.  EVERY.  DAMNED.  TIME.  It can of course be straightened to visually passable, but an air spaced center conductor remains internally and invisibly kinked, permanently compromising the cable's continuous impedance and loss specs.  Second, air spaced variants suffer from pushed and pulled connector pins because their center conductors aren't bonded and are free to float and migrate under repeated coiling, uncoiling, and temperature cycling. This induces enormous tensile and compressive forces on their connectors' components.  Third, their center conductors are exposed to moisture and corrode at an accelerated rate compared to bonded foam variants.  This failure mode has destroyed many of my clients' cables beyond repair in only several years of service, while bonded PE foam dielectric variants last decades without any corrosion at all.

And beware, EVERY single digit and alpha in a cable model # is crucial. For example, Belden 9913 is a completely different cable than 9913F7.  Just because their loss per ft. specs are similar means nothing regarding all their other properties.

Jason is absolutely correct, and my error for including 9913. 9913F7 is the variant to consider.

[Mark Cadwallader]

Jason is absolutely correct, and my error for including 9913. 9913F7 is the variant to consider.

Henry, would you be willing to edit your prior post to remove the incorrect part number?  I try to always read the entire string of posts, but I sometimes am lazy when searching for a specific reference in an old thread I only generally recollect. Thank you. (And thank you for sharing your knowledge and experience with RF newbies like me.)

[Henry Cohen]

Henry, would you be willing to edit your prior post to remove the incorrect part number?  I try to always read the entire string of posts, but I sometimes am lazy when searching for a specific reference in an old thread I only generally recollect.

Sensible request. Done.

[ProSoundWeb Community]